Mitochondrial genome data support the basal position of Acoelomorpha and the polyphyly of the Platyhelminthes

MOLECULAR

Molecular Phylogenetics and Evolution 33 (2004) 321–332

PHYLOGENETICSANDEVOLUTION

www.elsevier.com/locate/ympev

Mitochondrial genome data support the basal positionof Acoelomorpha and the polyphyly of the Platyhelminthes

Inaki Ruiz-Trilloa, Marta Riutorta, H. Matthew Fourcadeb,Jaume Bagunaa, Jeffrey L. Booreb,c,*

a Departament de Genetica, Universitat de Barcelona, Av. Diagonal, 645, 08028 Barcelona, Spainb Evolutionary Genomics, DOE Joint Genome Institute and Lawrence Berkeley National Laboratory, Walnut Creek, CA, USA

c Department of Integrative Biology, University of California, Berkeley, CA, USA

Received 5 November 2003; revised 2 March 2004

Available online 29 July 2004

Abstract

We determined 9.7, 5.2, and 6.8kb, respectively, of the mitochondrial genomes of the acoel Paratomella rubra, the nemertoder-

matid Nemertoderma westbladi, and the free-living rhabditophoran platyhelminthMicrostomum lineare. The identified gene arrange-

ments are unique among metazoans, including each other, sharing no more than one or two single gene boundaries with a few

distantly related taxa. Phylogenetic analysis of the amino acid sequences inferred from the sequenced genes confirms that the aco-

elomorph flatworms (acoels+nemertodermatids) do not belong to the Platyhelminthes, but are, instead, the most basal extant bi-

laterian group. Therefore, the Platyhelminthes, as traditionally constituted, is a polyphyletic phylum.

� 2004 Elsevier Inc. All rights reserved.

Keywords: Platyhelminth; Acoel; Mitochondria; Evolution; Genome; Metazoa

1. Introduction

The Acoelomorpha, i.e., Acoela plus Nemertoder-matida, is a group of acoelomate worms that has tradi-

tionally been included in the phylum Platyhelminthes.

Recent studies of small subunit ribosomal RNA gene

(SSU) sequences, however, showed the Platyhelminthes

is a polyphyletic assemblage, with the acoelomorphs

representing the most basal extant bilaterian clade,

and the bulk of the Platyhelminthes (i.e., the Catenulida

and the Rhabditophora, which includes all parasiticclasses) branching within the Protostomia (Jondelius

et al., 2002; Ruiz-Trillo et al., 1999).

The basal position of the Acoelomorpha and the

consequent polyphyly of the Platyhelminthes have

1055-7903/$ - see front matter � 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ympev.2004.06.002

* Corresponding author. Fax : 1-925-296-5666.

E-mail address: [email protected] (J.L. Boore).

been further tested and corroborated using other mo-

lecular characters. First, the Hox clusters of rhabdi-

tophoran platyhelminths are now known to have analmost full set (seven or eight) of Hox genes (Bayascas

et al., 1998; Orii et al., 1999), some bearing signature

peptides indicative of lophotrochozoan and proto-

stome affinities (De Rosa et al., 1999). By contrast, re-

cent work shows acoels to have a limited set of Hox

genes (four or five), which do not bear such signature

peptides (Cook et al., 2004). Secondly, these same evo-

lutionary relationships have been found in compari-sons of myosin heavy chain type II (myosin II) gene

sequences from a large set of metazoans (both in iso-

lation and when combined with SSU sequences;

Ruiz-Trillo et al., 2002) and in a combined analysis

of SSU+LSU gene sequences (Telford et al., 2003). Fi-

nally, like all diploblasts examined so far, acoels do

not express the heterochronic gene let-7, an essential

regulator of developmental timing in all studied bilate-

322 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

rians including rhabditophoran Platyhelminthes (Pas-

quinelli et al., 2003).

It remains uncertain whether the Acoelomorpha itself

is a monophyletic group. Molecular data support either

a monophyletic (myosin II; Ruiz-Trillo et al., 2002) or a

paraphyletic (SSU and SSU+LSU; Jondelius et al.,2002; Telford et al., 2003) Acoelomorpha. The mono-

phyly of the Acoelomorpha is supported by some puta-

tive morphological synapomorphies, such as the specific

structure of its basal body–rootlet system of cilia and of

their ciliary tips, and fine structure of frontal organs

(Smith et al., 1986), the lack of a true brain with neuro-

pile (Raikova et al., 1998; Reuter et al., 1998), reduction

of ECM in the body wall (Rieger et al., 1991; Tyler andRieger, 1999) and, potentially, if shown for the nemerto-

dermatids, the duet-spiral type of embryonic cell cleav-

age (Henry et al., 2000).

Furthermore, although all molecular data clearly

show that the bulk of the Platyhelminthes is associated

with the Lophotrochozoa (Adoutte et al., 2000; Car-

ranza et al., 1997; von Nickisch-Rosenegk et al., 2001;

Ruiz-Trillo et al., 1999, 2002), whether they are derivedor basal lophotrochozoans is still uncertain, since none

of the genes used to infer phylogenetic relationships

has provided resolution at this level. The flatworm body

plan figures prominently in nearly all imagined scenarios

for the evolution of the Metazoa, so solving these issues

addresses questions not only in systematics, but in com-

parative development and in understanding the so-called

Cambrian explosion.Comparing complete mitochondrial genome se-

quences (including the comparison of gene order) can

be very powerful at addressing phylogenetic relation-

ships (Boore and Brown, 1998). Now, over 50 complete

mitochondrial genome sequences from invertebrate an-

imals are available (Boore, 1999; from Evolutionary

Genomics link at http://www.jgi.doe.gov). Nearly all

animal mitochondrial DNAs (mtDNAs) are circularmolecules, about 16kb in size, containing genes for

two ribosomal RNAs, 22 tRNAs, and 13 protein sub-

units. Among triploblast animals, the only gene known

to have been lost is atp8; this has occurred indepen-

dently in several lineages: nematodes (Keddie et al.,

1998; Okimoto et al., 1992), parasitic platyhelminths

(Le et al., 2002; von Nickisch-Rosenegk et al., 2001),

and the bivalve mollusk Mytilus edulis (Hoffmannet al., 1992).

In addition, the study of mitochondrial genomes also

provides additional information useful for phylogenetic

inference, the nucleotide and inferred amino acid se-

quences of genes. The comparison of the concatenated

nucleotide or amino acid sequences of mitochondrial

genes has been used successfully in some studies to solve

deep level relationships (Arnason and Janke, 2002; Bo-ore and Brown, 2000; Boore and Staton, 2002; Inoue

et al., 2003; Miya et al., 2003; Murata et al., 2003). How-

ever, this data set has not been well tested for its power

of resolution at the most basal bilaterian level.

To date, complete or near-complete mtDNA se-

quences are available for 11 species of Platyhelminthes,

all belonging to the parasitic classes (Le et al., 2002; von

Nickisch-Rosenegk et al., 2001). No data are presentlyavailable for any of the free-living Rhabditophora (sev-

en orders), nor for any catenulid or acoelomorph. The

characterization of mitochondrial genomes from free-

living Rhabditophora species and from the basal Acoel-

omorpha is necessary to both elucidate the primitive

traits for the mitochondrial genomes of metazoans, as

well as to assess phylogenetic relationships.

In this study, we describe the partial mitochondrialgenomes from two acoelomorphs, the acoel Paratomella

rubra and the nemertodermatid Nemertoderma westbla-

di, and from a free-living rhabditophoran flatworm,

the macrostomid Microstomum lineare (Order Macro-

stomida), the first representatives of these groups to be

examined. We analyze genomic features in comparison

with mtDNAs of a variety of metazoans and perform

a phylogenetic analysis with their inferred amino acid se-quences. Published mtDNA sequences from cnidarians

(Beagley et al., 1998; Beaton et al., 1998) were used as

outgroups. Our aims are to: (1) further test the phyloge-

netic position of acoels and nemertodermatids being

separated from the rest of the Platyhelminthes, (2) vali-

date the monophyly of the Acoelomorpha, (3) assess the

similarities and differences between free-living and para-

sitic rhabditophoran Platyhelminthes, and (4) test thephylogenetic value of mitochondrial sequences for deep

evolutionary relationships.

2. Materials and methods

2.1. Molecular analysis

Live specimens of P. rubra were obtained from Sitges

(Spain). Specimens of M. lineare and N. westbladi were

kindly provided by Dr. M. Reuter (Abo, Finland) and

Dr. Ulf Jondelius (Uppsala, Sweden). We isolated total

DNA using the Qiagen DNA extraction kit. Initially, we

used universal primers to amplify short fragments of the

genes cox1 (primers LCO1490 and HCO2198; Folmer

et al., 1994), cox3 (primers COIIIF and COIIIB; Booreand Brown, 2000), cob (primers Cytb424 and Cytb876;

Boore and Brown, 2000), and rrnL (primers 16ARL

and 16SBRH; Palumbi, 1996). We used a standard

PCR protocol (50ll, with 1U Dynazyme polymerase

of Finnzimes, 35 cycles of 20s at 94 �C, 45s at 48 �C,and 45s at 72 �C).

The products of those PCRs which amplified bands

of expected size (all in P. rubra, cox1, and rrnL in N.

westbladi, cox1, cob, and rrnL in M. lineare) were puri-

fied using Microcon PCR columns. Purified products

I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332 323

were cycle-sequenced directly from both strands using

BigDye chemistry, precipitated using the DyeEx Spin

Kit (Qiagen) column, and run on ABI Prism 373 or

377 automated sequencers. Contigs were assembled us-

ing SeqEd v1.03. DNA sequences obtained from these

fragments were used to design species-specific oligonu-cleotides that face outwards from the fragments. These

primers were then employed in long-PCR (Barnes,

1994) in all possible combinations with the Advantage

polymerase kit (Clontech). Reaction conditions were

as follows: 35 cycles of 8s at 94 �C and 15min at

70 �C, with 1U Advantage polymerase per 50llreaction.

This generated single fragments of approximately 3.7,6, and 9.7kb (from cox1 to rrnL, rrnL to cob, and cox1

to cob, respectively) in P. rubra; 5.3kb (from cox1 to

rrnL) in N. westbladi; and 3.1 and 3.7kb (from cox1 to

rrnL and from rrnL to cob, respectively) in M. lineare.

Multiple attempts at amplifying the other portions of

these mtDNAs while varying many parameters for the

reactions all failed.

Products were purified and cloned using TOPO XLPCR Cloning kit (Invitrogen). Cloned fragments were

purified by a miniprep extraction (High Pure Plasmid

Isolation Kit, Roche), and sequenced as above, with ad-

ditional primers used to walk through both strands of

each fragment. Sequences were assembled using the Seq-

man II program (DNASTAR).

2.2. Gene annotation

The protein and ribosomal RNA-encoding genes

were identified by comparisons with other pub-

lished sequences using BLAST programs at NCBI

(www.ncbi.nlm.nih.gov/BLAST; Zhang and Madden,

1997). tRNA genes were identified either by using

tRNAscan-SE (version 1.1, www.genetics.wustl.edu/

eddy/tRNAscan-SE; Lowe and Eddy, 1997) or, wheretRNAs were not found using this program, by recogniz-

ing potential secondary structures by eye. The 50 ends of

protein genes were inferred to be at the first legitimate,

in-frame start codon (ATN, GTG, TTG, and GTT),

even if this appeared to overlap by a few nucleotides

with the preceding gene. Protein gene termini were in-

ferred to be at the first in-frame stop codon unless that

codon was located within the sequence of a downstreamgene. Otherwise, a truncated stop codon (T or TA) ad-

jacent to the beginning of the next gene was designated

as the termination codon and was assumed to be com-

pleted by polyadenylation after transcript cleavage (Oj-

ala et al., 1981). The 50 and 30 ends of both rrnL and

rrnS genes were assumed to be adjacent to the ends of

bordering tRNA genes.

The nucleotide sequences reported in this article havebeen deposited in GenBank under Accession Nos.

AY228756, AY228757, and AY228758.

2.3. Alignment

Either the general invertebrate (for P. rubra and N.

westbladi) or the flatworm (M. lineare) mitochondrial

genetic codes were used to infer the amino acid sequence

of the protein-encoding genes. Amino acid and nucleo-tide sequences were aligned first by clustalX as imple-

mented in BioEdit 5.0.6 (http://www.mbio.ncsu.edu/

BioEdit/bioedit.html), and then revised by eye using

the GDE2.0 sequence editor (Smith et al., 1994). The ge-

netic code was tested, by eye, at positions that are clearly

conserved through all protein-encoding genes.

The inferred amino acid sequences of the three taxa

were imported into an aligned matrix that includes 42taxa representing a wide range of published metazoan

mitochondrial genomes (see Table 1 for species and

GenBank accession numbers). Ambiguously aligned po-

sitions and gaps were excluded from the analyses result-

ing in a total of 1383 aligned amino acid positions

(including 1069, 713, and 454 amino acid characters

for P. rubra, M. lineare, and N. westbladi, respectively).

2.4. Phylogenetic analyses

A relative rate test on all taxa was performed in

RRTree (Robinson-Rechavi and Huchon, 2000) in

which each taxon was considered to be a separate line-

age. For phylogenetic inference, three different data sets

were analyzed: (1) all taxa; (2) without N. westbladi; and

(3) without the taxa that did not pass the RRT. Thesedata sets were subjected to maximum likelihood (ML)

and neighbor-joining (NJ) analyses. ML analyses were

performed with TREE-PUZZLE 5.0 (Strimmer and

von Haeseler, 1996) using the mtREV24 model (Adachi

and Hasegawa, 1996), the gamma distribution (eight

categories) and 10,000 quartet puzzling (QP) replicates.

NJ analyses were performed in Mega 2.1 (Kumar

et al., 2001) using the gamma model, with the parameteralpha as previously calculated in TREE-PUZZLE and

1000 bootstrap replicates.

3. Results

3.1. Gene content and organization

The 9795nt portion of P. rubra contains a large non-

coding region and 21 genes: nine for proteins (with cox1

and cob being incomplete at the 50 and 30 ends, respec-

tively), 10 for tRNAs, and two for rRNAs. The

5243nt portion of N. westbladi contains 13 genes: four

for proteins (with cox1 being incomplete at the 30 end),

seven for tRNAs, and two for rRNAs (with rrnL being

incomplete at the 50 end). The 6882nt portion ofM. line-

are contains a large non-coding region and 13 genes: five

for proteins (with cob and cox1 being incomplete at the

Table 1

List of species used in this study with GenBank accession numbers

Higher taxon Species GenBank

Cnidaria

Cnidaria-1 Sarcophyton glaucum AF064823

Cnidaria-2 Metridium senile NC_000933

Deuterostomia

Vertebrata-1 Corvus frugilegus NC_002069

Vertebrata-2 Alligator mississippiensis NC_001922

Vertebrata-3 Oncorhynchus mykiss NC_001717

Vertebrata-4 Eumeces egregius NC_000888

Vertebrata-5 Mustelus manazo NC_000890

Vertebrata-6 Homo sapiens NC_001807

Cephalochordata Branchiostoma floridae NC_000834

Hemichordata Balanoglossus carnosus NC_001887

Echinodermata-1 Florometra serratissima NC_001878

Echinodermata-2 Asterina pectinifera NC_001627

Echinodermata-3 Strongylocentrotus purpuratus NC_001453

Echinodermata-4 Paracentrotus lividus NC_001572

Echinodermata-5 Arbacia lixula NC_001770

Lophotrochozoa

Rhabditophora-1 Fasciola hepatica NC_002546

Rhabditophora-2 Hymenolepis diminuta NC_002767

Rhabditophora-3 Microstomum linearea AY228756

Rhabditophora-4 Taenia crassiceps NC_002547

Rhabditophora-5 Echinococcus multilocularis AB018440

Mollusca-1 Albinaria coerulea NC_001761

Mollusca-2 Loligo bleekeri NC_002507

Mollusca-3 Katharina tunicata NC_001636

Mollusca-4 Cepaea nemoralis NC_001816

Mollusca-5 Mytilus edulis M83756-62

Annelida-1 Platynereis dumerilii NC_000931

Annelida-2 Lumbricus terrestris NC_001673

Brachiopoda-1 Terebratulina retusa NC_000941

Brachiopoda-2 Terebratalia transversa NC_003086

Ecdysozoa

Nematoda-1 Caenorhabditis elegans X54252

Nematoda-2 Ascaris suum X54253

Nematoda-3 Trichinella spiralis NC_002681

Nematoda-4 Onchocerca volvulus NC_001861

Chelicerata-1 Limulus polyphemus NC_003057

Chelicerata-2 Ixodes hexagonus NC_002010

Chelicerata-3 Rhipicephalus sanguineus NC_002074

Myriapoda Lithobius forficatus NC_002629

Hexapoda-1 Ceratitis capitata NC_000857

Hexapoda-2 Drosophila melanogaster NC_001709

Hexapoda-3 Anopheles gambiae NC_002084

Hexapoda-4 Locusta migratoria NC_001712

Crustacea-1 Daphnia pulex NC_000844

Crustacea-2 Artemia franciscana NC_001620

Acoela

Acoela Paratomella rubraa AY228758

Nemertodermatida

Nemertodermatida Nemertoderma westbladia AY228757

a Taxa sequenced in this study.

324 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

50 and 30 ends, respectively), six for tRNAs, and two for

rRNAs. All of the genes identified are found in the same

transcriptional orientation (Fig. 1).

All taxa have a compact mitochondrial genome orga-

nization, especially the acoel, in which most genes abut

directly. There are two start codons used in the P. rubra

Fig. 1. Gene map of the sequenced portions of the mtDNAs of the acoel Paratomella rubra, the nemertodermatid Nemertoderma westbladi, and the

rhabditophoran Microstomum lineare. Genes are not to scale. All genes are transcribed from left to right. Transfer RNA genes are designated by a

single letter for the corresponding amino acid, with the two serine tRNA genes differentiated by numeral; all other genes are designated by standard

annotation. ‘‘NC’’ for M. lineare refers to a large non-coding region. Lines are shown connecting homologous genes other than those specifying

tRNAs.

I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332 325

protein genes: ATT in six cases and ATA in cob and

nad1. ATG, ATT, and GTT are the start codons used

in N. westbladi protein genes. Finally, in M. lineare allfour start codons are different: GTT, ATT, ATA, and

ATG. Nearly all protein genes in the three taxa end

on a complete stop codon, either TAA or TAG. P. rubra

atp8, atp6, and nad1 and M. lineare cox3 end on an ab-

breviated stop codon (T or TA) that is presumably com-

pleted by polyadenylation of the mRNA (Ojala et al.,

1981).

All taxa presented here bear a gene arrangement un-known in any other metazoan mtDNA (Fig. 1). A thor-

ough comparison of each individual boundary with

those of all other published mtDNAs reveals nothing

but a few single adjacencies shared with a few distantly

related taxa, all certainly by coincidence. The two

ribosomal genes are found to be separated in the

acoelomorphs, which is uncommon among metazoan

Table 2

Nucleotide composition and AT- and GC-skew values for P. rubra, N. westbl

non-coding nucleotides

Number of

nucleotides

T (%) C (%) A (%

MtDNA P. rubra 9795 47.5 8.0 28.9

N. westbladi 5243 37.5 13.8 27.1

M. lineare 6882 47.9 8.6 29.5

Proteins P. rubra 6727 49.3 9.0 25.3

N. westbladi 3278 39.7 14.4 23.3

M. lineare 4091 50.8 9.1 25.4

rRNAs P. rubra 2207 43.0 5.9 36.3

N. westbladi 1418 33.4 13.6 33.0

M. lineare 1723 38.9 8.7 39.1

tRNAs P. rubra 653 42.6 6.1 39.6

N. westbladi 475 34.5 11.2 34.7

M. lineare 366 45.5 5.9 34.6

Non-coding P. rubra 200 52.0 4.5 32.0

N. westbladi —a — — —

M. lineare 667 54.3 6.8 27.4

a N. westbladi lacks information on non-coding regions, since the fragme

mtDNAs. M. lineare conserves the typical metazoan ar-

rangement rrnS-rrnL, but with trnS2 intervening instead

of the commonly found trnV (Fig. 1).

3.2. Base composition and codon usage

The 5243nt determined for N. westbladi mtDNA is

A+T rich, 64.6%, similar to many other mtDNAs.

The 9795 and 6882nt determined for P. rubra and M.

lineare, respectively, are very A+T rich (76.4 and

77.4%), even for a mtDNA (Table 2). In each case, thereported strand is very rich in T and G, showing that

the orientation of each TA and GC pair between the

two strands is highly asymmetric (especially in the acoel

and the rhabditophoran). This asymmetry can be quan-

tified by a measure of skewness devised by Perna and

Kocher (1995), which measures the relative number of

A�s to T�s by AT-skew ([A�T]/[A+T]) and G�s to C�s

adi, and M. lineare mtDNA, protein encoding, ribosomal, tRNAs, and

) G (%) A+T (%) G+C (%) AT-skew GC-skew

15.6 76.4 23.6 �0.24 0.32

21.6 64.6 35.4 �0.16 0.22

14.1 77.4 22.7 �0.23 0.24

16.4 74.6 25.4 �0.32 0.29

22.7 63.0 37.1 �0.26 0.22

14.7 76.2 23.8 �0.33 0.23

14.7 79.3 20.6 �0.08 0.42

20.0 66.4 33.7 �0.006 0.18

13.3 78.0 22.0 0.0025 0.20

11.7 82.2 17.8 �0.03 0.31

19.5 69.2 30.7 �0.002 0.27

14.0 80.1 19.9 �0.13 0.40

11.5 84.0 16.0 �0.23 0.43

— — — — —

11.6 81.7 18.4 �0.32 0.26

nt analyzed has only a few non-coding nucleotides.

326 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

by GC-skew ([G�C]/[G+C]). According to these for-

mulae, skew values can range from �1 to +1, with great-

er compositional asymmetry giving skew values (positive

or negative) closer to 1; a skew value of zero indicates

that the distribution is equal between the strands. AT-

and GC-skew values are, respectively, �0.24 and 0.32for the acoel; �0.16 and 0.22 for the nemertodermatid,

and �0.23 and 0.24 for the rhabditophoran (Table 2).

As expected, GC-skew values are, in general, higher

for tRNAs, ribosomal and non-coding nucleotides and

lower for protein-encoding nucleotides. However AT-

skew values are lower than expected for ribosomal,

Fig. 2. Potential secondary structure of the putative mt tRNAs found for P

tRNAs and non-coding nucleotides (Table 2). A possi-

ble explanation is that among ribosomal and tRNAs

genes a similar number of A�s and T�s are required for

stem structure formation in their products. This may

not have so strong an effect on GC-skew due to the po-

tential formation of GT pairs.

3.3. Transfer RNAs

There are 10 sequences in P. rubra, seven in N. westb-

ladi, and six in M. lineare identified with the potential

for folding into tRNA-like structures (Fig. 2). Each

aratomella rubra, Nemertoderma westbladi, and Microstomum lineare.

I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332 327

has a seven-member amino-acyl acceptor stem and a

five-member anticodon stem, some with one or two mis-

matches. The extra arms have three to five nucleotides.

There are usually three to five nucleotide pairs in both

the DHU and TwC arms. The putative tRNA(K) of

N. westbladi and tRNA(I) of M. lineare have mismatch-es and ambiguously determined nucleotides, and so may

not be real tRNAs. The tRNAs S1 of both acoelomorph

species lack a paired DHU arm. The nemertodermatid

and the macrostomid differ in the anticodon sequence

of tRNA(K), being TTT in N. westbladi (as in most

metazoans) and CTT in M. lineare (as in the flatworm

parasitic species and some others).

3.4. Genetic code

The mitochondrial genetic code for most inverte-

brates appears to deviate from the ‘‘universal’’ genetic

code with regard to the identities of ATA (methionine,

M, instead of isoleucine, I), TGA (tryptophan,W, in-

stead of a stop codon) and AGR codons (serine, S, in-

stead of arginine, R) (Table 3; see Wolstenholme,1992). Diploblasts share only the deviation for the co-

don TGA. Platyhelminthes have been inferred to have

two further differences: AAA encodes asparagine (N)

rather than lysine (K), and ATA encodes I rather than

M (as in the universal code) (Telford et al., 2000).

Our data seem to corroborate previous results (Tel-

ford et al., 2000), in which acoels and nemertodermatids

share the invertebrate mitochondrial genetic code with-out the flatworm variations (Table 3). This has been ob-

served in some amino acid positions throughout the

alignment, such as, for example, at amino acid position

271 (position corresponding to the Homo sequence) of

the Cob gene there is a very conserved I that corresponds

to an ATA codon in the nucleotide sequence from M.

lineare; and at amino acid position 136 of the Cox3 gene

there is a conserved N that corresponds to an AAA co-don in the nucleotide sequence from M. lineare. In the

two positions P. rubra has the codon corresponding to

the invertebrate mitochondrial genetic code for those

amino acids (ATC and AAT respectively; N. westbladi

has AAC for the second site).

However, the use of genetic code changes as phyloge-

netic characters should be used only with great caution,

since: (1) it remains undetermined how common suchchanges are, (2) inference of these deviations is often

Table 3

The amino acid specifications for four codons that commonly vary

TGA

‘‘Universal’’ genetic code Stop

Diploblastica W

Acoelomorpha W

Most invertebrates W

Rhabditophora W

speculative, and (3) there are obvious convergences,

such as those between platyhelminths and echinoderms

for the changes of AAA and ATA.

3.5. Phylogenetic analyses

In order to test whether the clade Acoelomorpha is

valid and whether M. lineare groups with the parasitic

flatworms, we performed initial ML and NJ analyses in-

cluding all of the taxa in Table 1 (see supplementary da-

ta), but the results are poorly supported and clearly

show artifactual associations. P. rubra and N. westbladi

are sister groups, forming the clade Acoelomorpha in

the NJ tree (but with only 60% bootstrap support),but not in the ML tree. The Acoelomorpha appears as

sister group of all the other bilaterians, as found in ear-

lier studies (Ruiz-Trillo et al., 2002). Both NJ and ML

analyses group the rhabditophoran M. lineare with the

parasitic flatworms (53% QP and 89% NJ bootstrap sup-

port). There is an artifactual grouping of nematodes

with platyhelminths and Mytilus within the Lophotro-

chozoa due to the effect of long-branch attraction.The low support for the basal position of Acoelo-

morpha and the difference between NJ and ML analyses

on the Acoelomorpha clade might be due to the fact

that the N. westbladi mitochondrial genome sequence

includes only three complete and one partial protein

coding genes, that is, 454 amino acid positions out of

the total 1383. Thus, we performed an additional anal-

ysis without the nemertodermatid N. westbladi, whichresulted in a similar topology, but with a general in-

crease in branch support (see supplementary data).

Again, nematodes group artifactually with the other

long-branch taxa Platyhelminthes and Mytilus. The

rhabditophoran M. lineare groups with the other platy-

helminths (now with 72% QP and 92% NJ bootstrap

value) and the acoel P. rubra appears as sister group

of all the other bilaterians (now with 83% QP and NJbootstrap support). These analyses corroborate previ-

ous results that acoelomorphs are the most basal extant

bilaterian clade.

We performed a relative rate test for all taxa. At the

1% confidence level, several protostome taxa appear to

have much faster evolutionary rates than the other bila-

terians, and were then eliminated from subsequent phy-

logenetic analysis. Such is the case of the nematodes, theplatyhelminths, the ticks Ixodes and Rhiphicephalus, the

ATA AGR AAA

I R K

I R K

M S K

M S K

I S N

328 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

acoelomorph N. westbladi, and the mollusks Mytilus

and Cepaea. The acoel P. rubra and the lophotrochozo-

ans Albinaria, Loligo, and Terebratalia have only faster

evolutionary rates compared to a few (two to five) short-

er-branched deuterostomes. Thus, they were not consid-

ered fast-clock taxa, and were included in the finalanalysis. A table summarizing this analysis can be found

in the supplementary material.

Fig. 3 shows the results of phylogenetic analysis with

the dataset limited to taxa of more moderate rates of se-

quence change. The ML tree used QP with 10,000 repli-

cates and the NJ tree used 1000 bootstrap replicates for

branch support. The topology corroborates previous

SSU and myosin II analyses (Ruiz-Trillo et al., 1999,2002). Acoela is the most basal extant bilaterian clade

(97 and 96% QP and NJ branch support) and the rest

of the bilaterians, or Eubilateria, are divided into the

three super-clades, Deuterostomia (82% QP and 99%

Fig. 3. Phylogenetic tree based on mitochondrial protein-coding gene

sequences using quartet-puzzling (QP), a form of maximum likelihood

analysis. Numbers above key nodes refers to the percentage obtained

from a QP with 10,000 replicates followed by support from a neighbor-

joining (NJ) analysis with 1000 bootstrap replicates. Asterisks indicate

non-key nodes with more than a 90% for both QP and NJ support.

Taxa with increased evolutionary rates as detected in the RRT have

been excluded. For species names and taxa excluded see Table 1. Ec,

Ecdysozoa; Lo, Lophotrochozoa; De, Deuterostomia; Ac, Acoelo-

morpha; and Di, Diploblastica.

NJ), Lophotrochozoa (90% QP and 81% NJ), and Ec-

dysozoa (100% QP and 100% NJ). The support for a

protostomian clade is very high (91% QP and 98%

NJ). Of course, taxon sampling for many of these

groups remains low, e.g., it is not evaluated in this anal-

ysis whether Ecdysozoa includes the nematodes.

4. Discussion

The results of this study constitute the first attempt to

analyze at the mitochondrial genome level any acoel,

nemertodermatid, or free-living flatworm taxa, the first

molecular evolutionary analysis of their mitochondrialgenome organization, and a test of using mitochondrial

amino acid sequences to address very deep evolutionary

relationships among major animal groups.

4.1. The basal position of the Acoelomorpha and the

polyphyly of the Platyhelminthes

The amino acid sequence comparisons presented here(see Fig. 3 and supplementary data) add to a growing

list of molecular data backing a basal position for the

Acoelomorpha (initially reported using 18S rDNA se-

quences by Ruiz-Trillo et al. (1999) and Jondelius et

al. (2002)) and a derived, lophotrochozoan position,

for the rest of the Platyhelminthes (anticipated in Car-

ranza et al., 1997). These data include SSU sequences,

Hox genes, myosin II and myosin II+SSU sequences,LSU+SSU sequences, and let-7 gene expression (see

Section 1 for a detailed description and references).

How can we reconcile this with the morphological

evidence traditionally used to argue for monophyly of

Platyhelminthes (e.g., Hyman, 1951)? The three mono-

phyletic groups recognized within the Platyhelminthes

as traditionally constructed—Acoelomorpha, Catenul-

ida, and Rhabditophora (Smith et al., 1986)—are eachdefined by well-accepted synapomorphies. However,

lack of robust morphological synapomorphies that

would unite these three major clades led Smith et al.

(1986) to question the monophyly of the Platyhelmin-

thes. Nonetheless, the phylum was generally believed

to be monophyletic because they share a long list of

characters, including having soft bodies, hermaphrodit-

ism, internal fertilization, filiform and biflagellatesperm, lack of acrosomes, and multiciliated epidermal

cells. Other characters, such as spiral cleavage and a

ladder-like nervous system (see Brusca and Brusca,

1990 for general references), have been also considered,

but have been scored for only a limited range of taxa.

However, most of these characters could more

appropriately be argued to be symplesiomorphies or

homoplasies.Considering potential characters that would unite

Acoelomorpha, Catenulida, and Rhabditophora, there

I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332 329

are two others that bear serious consideration: lack of

mitosis in somatic cells (Ehlers, 1985) and presence of

a sack-like gut (which means lack of a proper anus).

The first refers to the inability of most somatic cells to

divide, with new cells being generated from undifferenti-

ated cells which, in Platyhelminthes, are known as neo-blasts (Baguna, 1981; Baguna et al., 1989). However,

this character is imprecisely defined, since all animals

have differentiated cells that lack the ability to divide.

Moreover, it has not been properly scored for many

phyla. As for the second character, it has been suggested

that the sack-like gut of Platyhelminthes (including

Acoelomorpha) has been derived by paedomorphosis

from the one-way through gut of a more complex ances-tor. A parsimonious reinterpretation of this feature on a

tree that places Acoelomorpha as basal to Bilateria is

that it is a symplesiomorphy for the Acoelomorpha

and separately derived for the clade of Catenulida and

Rhabditophora within the Lophotrochozoa.

Alternatively, there are characters that support a

basal position of Acoelomorpha. First, acoels generate

only endomesoderm (Henry et al., 2000), in contrastwith the rest of the Platyhelminthes (and protostomes

in general) that have both ecto- and endomesoderm.

Endomesoderm is considered the ancestral form. Sec-

ond, acoelomorphs have an anterior concentration of

nerve cells without forming a ‘‘true brain’’ with neuro-

pile (Raikova et al., 2000; Reuter et al., 1998; but see

Tyler, 2001 for a contrary view). Moreover, while

other bilaterians, including catenulids and rhabdito-phorans, have longitudinal nerve cords that are dis-

tinctly dorsal or ventral, acoelomorphs have a radial

arrangement, interpreted here as being primitive or

separately derived. Finally, limited and preliminary

study indicates that aceolomorphs have only few (four

to five) Hox genes (Cook et al., 2004), so that the ex-

panded repertoire may be a synapomorphy of the

other bilaterians.This revised phylogeny compels a reinterpretation of

the evolution of some features. First, the lack of proto-

nephridia for Acoelomorpha is regarded in traditional

schemes as having been derived by loss from their

platyhelminth ancestor. However, under the new sce-

nario, this lack of protonephridia may be the retention

of a primitive condition, a state shared with diplo-

blasts. Second, the ‘‘duet-spiral’’ type of embryonic cellcleavage has been usually considered to be derived

from the quartet type spiral cleavage of other Platyhel-

minthes. As discussed by Henry et al. (2000), duet-spir-

al cleavage may have arose alternatively from a form

of radial or biradial cleavage characteristic of the more

primitive programs in the Metazoa, whereas quartet

spiral cleavage would have originated independently

within the Lophotrochozoa. Finally, as mentionedabove, the sack-like gut of the Acoelomorpha may be

a symplesiomorphy shared with the similar state of dip-

loblasts, followed by the appearance of a one-way

through gut in the rest of Bilateria, within which the

Platyhelminthes (without the Acoelomorpha) separate-

ly adopted a similar condition.

Finally, is there enough evidence to support a mono-

phyletic Acoelomorpha? Potential morphological syna-pomorphies of the Acoelomorpha are: (1) a network

formed by interconnecting rootlets of epidermal cilia;

(2) a shaft region in epidermal cilia; (3) similar fine struc-

ture of frontal organs; (4) reduced extracellular matrix

(ECM); and (5) absence of protonephridia (Ax, 1996;

Ehlers, 1985; Smith et al., 1986; Tyler and Rieger,

1999). Another suggested synapomorphy, the spiral-du-

et type of cleavage, has not been unequivocally deter-mined for nemertodermatids, though it seems to be

present (Ulf Jondelius, personal communication). Al-

though the comparison of ribosomal genes results in a

paraphyletic relationship (Jondelius et al., 2002; Telford

et al., 2003) for the Acoelomorpha, phylogenetic analy-

ses of the amino acid sequences of mitochondrial genes

(this work) and myosin II sequences (Ruiz-Trillo et al.,

2002) show a monophyletic Acoelomorpha. However,the fact that acoels and nemertodermatids do not share

any gene boundaries between them and the long branch-

es separating them in the phylogenetic tree (supplemen-

tary data), argue for an extended period since the split of

these lineages or, alternatively, for a very rapidly chang-

ing mode of gene arrangement and nucleotide substitu-

tion. Additional support for an extended period since

acoels and nemertodermatids diverged comes from mor-phology. Acoels and nemertodermatids have significant

differences in the pattern of neurotransmitters (Raikova

et al., 2000), statocyst structure (Ax, 1996) and in the

fact that acoels digestive tract structure is either syncy-

tial or cellular while in nemertodermatids is always epi-

thelial (Rieger et al., 1991; Smith and Tyler, 1985).

Sperm morphology is also different in both lineages:

Nemertodermatids have uniflagellate spermatozoa withthe common 9+2 axonemal pattern (Lundin and Hen-

delberg, 1998), while Acoel spermatozoa bear two flagel-

la with reverse orientation and sometimes a modified

axoneme structure (Raikova et al., 2001). A short com-

mon period followed by a long divergence could explain

the failure of ribosomal genes to recover the Acoelomor-

pha as a monophyletic group. Acoela and Nemertoder-

matida (specially the first) present a high rate ofsubstitutions for these genes hence it makes more prob-

able that any signal that could have accumulated in their

common ancestor would have been erased by the subse-

quent divergence period.

To summarize, our analyses of mitochondrial amino

acid sequences added to morphological characters and a

growing number of molecular evidences argue for a

monophyletic Acoelomorpha as a basal bilateriangroup, occupying a pivotal position between diploblast

and triploblasts. Consequently, under this evolutionary

330 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

scenario, we believe Acoelomorpha merits establishment

as a new phylum.

4.2. Rhabditophoran mitochondrial variability

The fact that M. lineare does not share any geneboundaries with any of the parasitic rhabditophoran

flatworms sequenced to date demonstrates that there is

great variability in mitochondrial genome structure

within the Platyhelminthes. This bolsters confidence that

mitochondrial gene arrangements might be a useful tool

for inferring evolutionary relationships within the phy-

lum, many of which remain unresolved (Baguna and

Riutort, 2004; Littlewood and Olson, 2001).

4.3. Phylogenetic value of mitochondrial sequences at

deep evolutionary levels

Our phylogenetic analyses of mitochondrial protein-

coding sequences demonstrate their power for inferring

ancient evolutionary relationships, although they have

difficulty handling problems with long-branch taxa asis the case with all molecular sequence comparisons.

Mitochondrial sequence comparisons support the

monophyly of the three eubilaterian super-clades, Deu-

terostomia, Lophotrochozoa, and Ecdysozoa, at least

so far as the constituent taxa have been sampled. Also,

this analysis well resolves the branching pattern be-

tween the three eubilaterian super-clades, in which

protostomes (ecdysozoans and lophotrochozoans)form a monophyletic group (Fig. 3). Many previous

molecular studies were not able to fully resolve the

branching order among the three super-clades (Ado-

utte et al., 2000; Ruiz-Trillo et al., 1999, 2002; Telford

et al., 2003).

In summary, the high branch support for: (1) the

three bilaterian super-clades, (2) the basal position of ac-

oels, and (3) the branching order of the three eubilateri-an super-clades, indicates mitochondrial sequences may

be indeed a good and promising phylogenetic marker

for deep evolutionary events. This phylogenetic value

of mitochondrial sequences should be carefully tested,

however, with the inclusion of enigmatic taxa, such as

gastrotrichs, rotifers, and chaetognaths.

Acknowledgments

We thank Ulf Jondelius and Maria Reuter for pro-

viding N. westbladi and M. lineare specimens and Clint

Turbeville, Kevin Helfenbein, and Jordi Paps for techni-

cal assistance. We thank the people in the sequencing

unit of the ‘‘Serveis cientıfico-tecnics’’ of the Universitat

de Barcelona. I.R.-T., M.R., and J.B. were supported byCIRIT (Generalitat de Catalunya) Grants 1999SGR-

00026 and 2001SGR-00102 and M.R. and I.R.-T. by

DGICYT (Ministerio de Ciencia y Tecnologıa) Grant

PB97-0937. I.R.-T. was sponsored by a predoctoral

grant from the Universitat de Barcelona. Part of this

work was performed under the auspices of the US De-

partment of Energy, Office of Biological and Environ-

mental Research, in the University of California,Lawrence Berkeley National Laboratory, under Con-

tract No. DE-AC03-76SF00098.

Appendix A. Supplementary material

Supplementary data associated with this article can

be found, in the online version, at doi:10.1016/j.ymp-ev.2004.06.002.

References

Adachi, J., Hasegawa, M., 1996. Model of amino acid substitution in

proteins encoded by mitochondrial DNA. J. Mol. Evol. 42, 459–

468.

Adoutte, A., Balavoine, G., Lartillot, N., Lespinet, O., Prud�homme,

B., de Rosa, R., 2000. The new animal phylogeny: reliability and

implications. Proc. Natl. Acad. Sci. USA 97, 4453–4456.

Arnason, U., Janke, A., 2002. Mitogenomic analyses of eutherian

relationships. Cytogenet. Genome Res. 96, 20–32.

Ax, P., 1996. in: Multicellular Animals. A New Approach to the

Phylogenetic Order in Nature, vol. I. Springer-Verlag, Berlin,

Germany.

Baguna, J., 1981. Planarian neoblasts. Nature 290, 14–15.

Baguna, J., Salo, E., Auladell, C., 1989. Regeneration and pattern

formation in planarians III. Evidence that neoblasts are totipotent

stem cells. Development 107, 77–86.

Baguna, J., Riutort, M., 2004. Molecular phylogeny of the platyhel-

minthes. Can. J. Zool. 82, 168–193.

Barnes, W.M., 1994. PCR amplification of up to 35-kb DNA with high

fidelity and high yield from bacteriophage templates. Proc. Natl.

Acad. Sci. USA 91, 2216–2220.

Bayascas, J.R., Castillo, E., Salo, E., 1998. Platyhelminthes have a

Hox code differentially activated during regeneration, with genes

closely related to those of spiralian protostomes. Dev. Genes Evol.

208, 467–473.

Beagley, C.T., Okimoto, R., Wolstenholme, D.R., 1998. The mito-

chondrial genome of the sea anemone Metridium senile (Cnidaria):

introns, a paucity of tRNA genes, and a near-standard genetic

code. Genetics 148, 1091–1108.

Beaton, M.J., Roger, A.J., Cavalier-Smith, T., 1998. Sequence analysis

of the mitochondrial genome of Sarcophyton glaucum: conserved

gene order among octocorals. J. Mol. Evol. 47, 697–708.

Boore, J.L., Brown, W.M., 1998. Big trees from little genomes:

mitochondrial gene order as a phylogenetic tool. Curr. Opin.

Genet. Dev. 8, 668–674.

Boore, J.L., 1999. Animal mitochondrial genomes. Nucleic Acids Res.

27, 1767–1780.

Boore, J.L., Brown, W.M., 2000. Mitochondrial genomes of Galathe-

olinum, Helobdella, and Platynereis: sequence and gene arrange-

ment comparisons indicate that Pogonophora is not a phylum and

Annelida and Arthropoda are not sister taxa. Mol. Biol. Evol. 17,

87–106.

Boore, J.L., Staton, J.L., 2002. The mitochondrial genome of the

Sipunculid Phascolopsis gouldii supports its association with

Annelida rather than Mollusca. Mol. Biol. Evol. 19, 127–137.

I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332 331

Brusca, R.C., Brusca, G.J., 1990. Invertebrates. Sinauer Associates,

Sunderland.

Carranza, S., Baguna, J., Riutort, M., 1997. Are the Platyhelminthes a

monophyletic primitive group? An assessment using 18S rDNA

sequences. Mol. Biol. Evol. 14, 485–497.

Cook, C.E., Jimenez, E., Akam, M., Salo, E., 2004. The Hox

complement of acoel flatworms, a basal bilaterian clade. Evol.

Dev. 6, 154–163.

De Rosa, R., Grenier, J.K., Andreeva, T., Cook, C.E., Adoutte, A.,

Akam, M., Carroll, S.B., Balavoine, G., 1999. Hox genes in

brachiopods and priapulids and protostome evolution. Nature 399,

772–776.

Ehlers, U., 1985. Phylogenetic relationships within the Platyhelmin-

thes. In: Morris, S.C., George, J.D., Gibson, R., Platt, H.M. (Eds.),

The Origin and Relationships of Lower Invertebrates Groups.

Oxford University Press, Oxford, pp. 143–158.

Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R., 1994.

DNA primers for amplification of mitochondrial cytochrome c

oxidase subunit I from diverse metazoan invertebrates. Mol. Mar.

Biol. Biotechnol. 3, 294–299.

Henry, J.Q., Martindale, M.Q., Boyer, B.C., 2000. The unique

developmental program of the acoel flatworm, Neochildia fusca.

Dev. Biol. 220, 285–295.

Hyman, L.H., 1951. The Invertebrates. II. Platyhelminthes and

Rhynchocoela. The Acoelomate Bilateria. McGraw-Hill, New

York.

Hoffmann, R.J., Boore, J.L., Brown, W.M., 1992. A novel mitochon-

drial genome organization for the blue mussel, Mytilus edulis.

Genetics 131, 397–412.

Inoue, J.G., Miya, M., Nishida, M., 2003. Basal actinopterygian

relationships: a mitogenomic perspective on the phylogeny of the

ancient fish. Mol. Phylogenet. Evol. 26, 110–120.

Jondelius, U., Ruiz-Trillo, I., Baguna, J., Riutort, M., 2002. The

Nemertodermatida are basal bilaterians and not members of the

Platyhelminthes. Zool. Scr. 31, 201–215.

Keddie, E.M., Higazi, T., Unnasch, T.R., 1998. The mitochondrial

genome of Onchocerca volvulus: sequence, structure and phyloge-

netic analysis. Mol. Biochem. Parasitol. 95, 111–127.

Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. MEGA2:

molecular evolutionary genetics analysis software. Bioinformatics

17, 1244–1245.

Le, T.H., Blair, D., McManus, D.P., 2002. Mitochondrial genomes of

parasitic flatworms. Trends Parasitol. 18, 206–213.

Littlewood, D.T.J., Olson, P.D., 2001. Small subunit rDNA and the

Platyhelminthes: signal, noise and compromise. In: Littlewood,

D.T.J., Bray, R.D. (Eds.), The Interrelationships of the Platyhel-

minthes. Taylor and Francis, London, pp. 252–278.

Lowe, T.M., Eddy, S.R., 1997. tRNAscan-SE: a program for

improved detection of transfer RNA genes in genomic sequence.

Nucleic Acids Res. 25 (5), 955–964.

Lundin, K., Hendelberg, J., 1998. Is the sperm type of the Nemerto-

dermatida close to that of the ancestral Platyhelminthes?. Hydro-

biologia 383, 197–205.

Miya, M., Takeshima, H., Endo, H., Ishiguro, N.B., Inoue, J.G.,

Mukai, T., Satoh, T.P., Yamaguchi, M., Kawaguchi, A., Mabuchi,

K., Nishida, M., 2003. Major patterns of higher teleostean

phylogenies: a new perspective based on 100 complete mitochon-

drial DNA sequences. Mol. Phylogenet. Evol. 26, 121–138.

Murata, Y., Nikaido, M., Sasaki, T., Cao, Y., Fukumoto, Y., Okada,

N., 2003. Afrotherian phylogeny as inferred from complete

mitochondrial genomes. Mol. Phylogenet. Evol. 28, 253–260.

von Nickisch-Rosenegk, M., Brown, W.M., Boore, J.L., 2001.

Complete sequence of the mitochondrial genome of the tapeworm

Hymenolepis diminuta: gene arrangements indicate that platyhel-

minthes are eutrochozoans. Mol. Biol. Evol. 18, 721–730.

Ojala, D., Montoya, J., Attardi, G., 1981. tRNA punctuation model of

RNA processing in human mitochondria. Nature 290, 470–474.

Okimoto, R., MacFarlane, J.L., Clary, D.O., Wolstenholme, D.R.,

1992. The mitochondrial genomes of two nematodes, Caenorhabd-

itis elegans and Ascaris suum. Genetics 130, 471–498.

Orii, H., Kato, K., Umesono, Y., Sakurai, T., Agata, K., Watanabe,

K., 1999. The planarian HOM/HOX homeobox genes (Plox)

expressed along the anteroposterior axis. Dev. Biol. 210, 456–468.

Palumbi, S.R., 1996. Nucleic acids II: the polymerase chain reaction.

In: Hillis, D.M., Moritz, C., Mable, B.K. (Eds.), Molecular

Systematics. Sinauer, Sunderland, MA, pp. 205–247.

Pasquinelli, A.E., McCoy, A., Jimenez, E., Salo, E., Ruvkun, G.,

Martindale, M.Q., Baguna, J., 2003. Expression of the 22 nucle-

otide let-7 heterochronic RNA throughout the Metazoa: a role in

life history evolution?. Evol. Dev. 5, 372–378.

Perna, N.T., Kocher, T.D., 1995. Patterns of nucleotide composition

at fourfold degenerate sites of animal mitochondrial genomes. J.

Mol. Evol. 41, 353–358.

Raikova, O.I., Reuter, M., Kotikova, E.A., Gustafsson, M.K.S., 1998.

A commisural brain! The pattern of 5-HT immunoreactivity in

Acoela (Platyhelminthes). Zoomorphology 118, 69–77.

Raikova, O.I., Reuter, M., Jondelius, U., Gustafsson, M.K.S., 2000.

The brain of the Nemertodermatida (Platyhelminthes) as revealed

by anti-5HT and anti-FMRMamide immunostainings. Tissue Cell

32, 358–365.

Raikova, O.I., Reuter, M., Justine, J.L., 2001. Contributions to the

phylogeny and systematics of the Acoelomorpha. In: Littlewood,

D.T.J., Bray, R.A. (Eds.), Interrelationships of the Platyhelmin-

thes. Taylor and Francis, London, pp. 13–23.

Reuter, M., Raikova, O.I., Gustafsson, M.K., 1998. An endocrine

brain? The pattern of FMRF-amide immunoreactivity in Acoela

(Plathelminthes). Tissue Cell 30, 57–63.

Rieger, R.M., Tyler, S., Smith III, J.P.S., Rieger, G.E., 1991.

Platyhelminthes: Turbellaria. In: Harrison, F.W., Bogitsh, B.J.

(Eds.), Microscopic Anatomy of Invertebrates. vol. 3: Platyhel-

minthes and Nemertinea. Wiley-Liss, New York, pp. 7–140.

Robinson-Rechavi, M., Huchon, D., 2000. RRTree: relative-rate tests

between groups of sequences on a phylogenetic tree. Bioinformatics

16, 296–297.

Ruiz-Trillo, I., Riutort, M., Littlewood, D.T.J., Herniou, E.A.,

Baguna, J., 1999. Acoel flatworms: earliest extant bilaterian

metazoans, not members of the Platyhelminthes. Science 283,

1919–1923.

Ruiz-Trillo, I., Paps, J., Loukota, M., Ribera, C., Jondelius, U.,

Baguna, J., Riutort, M., 2002. A phylogenetic analysis of myosin

heavy chain type II sequences corroborates that Acoela and

Nemertodermatida are basal bilaterians. Proc. Natl. Acad. Sci.

USA 99, 11246–11251.

Smith, J.P.S., Tyler, S., 1985. The acoel turbellarians: kingpins of

metazoan evolution or a specialized offshoot. In: Conway Morris,

C., George, J.D., Gibson, R., Platt, H.M. (Eds.), The Origins and

Relationships of Lower Invertebrates. Oxford University Press,

Oxford, pp. 123–142.

Smith, J.P.S., Tyler, S., Rieger, R.R., 1986. Is the Turbellaria

polyphyletic?. Hydrobiologia 132, 13–21.

Smith, S.W., Overbeek, R., Woese, C.R., Gilbert, W., Gillevet, P.M.,

1994. The genetic data environment: an expandable GUI for

multiple sequence-analysis. Comp. Appl. Biosci. 10, 671–675.

Strimmer, K., von Haeseler, A., 1996. Quartet puzzling: a quartet

maximum likelihood method for reconstructing tree topologies.

Mol. Biol. Evol. 13, 964–969.

Telford, M.J., Herniou, E.A., Russell, R.B., Littlewood, D.T.J., 2000.

Changes in mitochondrial gene tic code as phylogenetic characters:

two examples from the flatworms. Proc. Natl. Acad. Sci. USA 97,

11359–11364.

Telford, M.J., Lockyer, A.E., Cartwright-Finch, C., Littlewood,

D.T.J., 2003. Combined large and small subunit ribosomal RNA

phylogenies support a basal position of the acoelomorph flat-

worms. Proc. R. Soc. (Lond.) B 270, 1077–1083.

332 I. Ruiz-Trillo et al. / Molecular Phylogenetics and Evolution 33 (2004) 321–332

Tyler, S., 2001. The early worm: origins and relationships of the lower

flatworms. In: Littlewood, D.T.J., Bray, R. (Eds.), Interrelation-

ships of the Platyhelminthes. Taylor & Francis, London, pp. 3–12.

Tyler, S., Rieger, R.M., 1999. Functional morphology of musculature

in the acoelomate worm, Convoluta pulchra (Platyhelminthes).

Zoomorphology 119, 127–141.

Wolstenholme, D.R., 1992. Genetic novelties in mitochondrial ge-

nomes of multicellular animals. Curr. Opin. Genet. Dev. 2, 918–

925.

Zhang, J., Madden, T.L., 1997. Power BLAST: a new network BLAST

application for interactive or automated sequence analysis and

annotation. Genome Res. 7, 649–656.