The first sequenced mitochondrial genome of a placozoan, Trichoplax adhaerens, challenged the conventional wisdom that a compact mitochondrial genome is a common feature among all animals. Three additional placozoan mitochondrial genomes representing highly divergent clades have been sequenced to determine whether the large Trichoplax mtDNA is a shared feature among members of the phylum Placozoa or a uniquely derived condition. All three mitochondrial genomes were found to be very large, 32- to 37-kb, circular molecules, having the typical 12 respiratory chain genes, 24 tRNAs, rnS, and rnL. They share with the Trichoplax mitochondrial genome the absence of atp8, atp9, and all ribosomal protein genes, the presence of several cox1 introns, and a large open reading frame containing an intron group I LAGLIDADG endonuclease domain. The differences in mtDNA size within Placozoa are due to variation in intergenic spacer regions and the presence or absence of long open reading frames of unknown function. Phylogenetic analyses of the 12 respiratory chain genes support the monophyly of Placozoa. The similarities in composition and structure between the three mitochondrial genomes reported here and that of Trichoplax's mtDNA suggest that their uncompacted state is a shared ancestral feature to other nonmetazoans while their gene content is a derived feature shared only among the Metazoa.
Animals typically have much smaller mitochondrial genomes than do nonanimal single-celled organisms and fungi. Whereas animal mitochondrial genomes are characterized by a tightly packed collection of conserved genes and other functional elements, the larger nonanimal mitochondrial genomes generally contain noncoding regions, such as introns and additional genes not present among animals. It has thus been argued that drastic mitochondrial size reduction occurred deep in evolutionary history, before the emergence of animals. In this study, however, we show that the phylum Placozoa, arguably one of the most ancient animal groups, possesses mitochondrial genomes of intermediate size, smaller than the typical nonanimal genome yet much larger than the mitochondrial genome found in typical animals. As in nonanimals, the increased size of the placozoan mitochondrial genome is due to the presence of additional genes, introns, and long noncoding regions. Although other large animal mitochondrial genomes have been discovered, they have been isolated findings in particular species and none encompassed as large a taxonomic group as the level of phylum. Because large mitochondrial genomes are a shared feature among all placozoans and given this phylum's phylogenetic position in the animal tree, we conclude that, contrary to conventional wisdom, the ancestral animal mitochondria was likely a large, noncompacted molecule.
Citation: Signorovitch AY, Buss LW, Dellaporta SL (2007) Comparative Genomics of Large Mitochondria in Placozoans. PLoS Genet 3(1): e13. doi:10.1371/journal.pgen.0030013
Editor: Gil McVean, University of Oxford, United Kingdom
Received: September 11, 2006; Accepted: December 5, 2006; Published: January 12, 2007
Copyright: © 2007 Signorovitch et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by National Science Foundation grant EF-0319076 and in part by a Caribbean Coral Reed Ecosystems Program Award (contribution No. 781), National Institutes of Health Genetics training grant 5 T32 GM07499–28, and National Institutes of Health grant R21-AI066242-01A1.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: ML, maximum likelihood; ORF, open reading frame
Comparative mitochondrial genomics is becoming a powerful approach to resolving phylogenetic relationships among distantly related taxa (e.g., [1–7], and reviewed in [8,9]). A problem of particular interest to evolutionary biologists has been the order in which animal phyla comprising the lower Metazoa (cnidarians, ctenophores, placozoans, and sponges) diverged. Since these animal phyla are believed to belong to the earliest diverging branches in the animal tree, learning which characteristics they share with our nonanimal relatives and which traits are unique to animals is essential to understanding metazoan evolution. Recently, sequencing of the entire Trichoplax adhaerens (phylum Placozoa) mitochondrial genome  made possible phylogenetic comparisons using all shared mitochondrial coding sequences (12 respiratory chain genes in all) across three lower metazoan phyla—Cnidaria, Placozoa, and Porifera—and two outgroup species, the choanoflagellate Monosiga brevicollis and the chytrid fungus Monoblepharella. This revised phylogeny not only provides support for the placement of the lineage leading to the placozoan Trichoplax as a basal animal phylum but also raises the possibility that the ancestral animal mitochondrial genome could have actually been a large molecule akin to that of Trichoplax, instead of a compact molecule similar to that of all other animals.
Trichoplax has one of the smallest animal nuclear genomes [10–12] and yet the largest animal mitochondrial genome . Most animal mitochondrial genomes are small, 15- to 20-kb, circular molecules encoding the typical respiratory chain genes (ATP synthase: atp6 and atp8, apocytochrome b: cob, cytochrome oxidase: cox1, cox2, and cox3, and reduced nicotinamide adenine dinucleotide ubiquinone oxireductase: nad1–6, and nad4L), 22 tRNA genes, and two rRNA genes. These genes are compactly arranged, sometimes overlapping, and usually lacking intronic and intergenic spacer regions. Although large animal mitochondrial genomes have been discovered [13–16], they are relatively rare and owe their large size to secondary expansions such as duplications, AT-rich regions, and multiple short tandem repeats. In contrast, plant, fungi, and protist mitochondrial genomes are often very large, being up to an order of magnitude larger in size than mtDNA found in the typical animal mitochondria. These nonanimal mtDNA encode many additional proteins—in particular, ribosomal proteins, which are completely absent in animal mitochondria—sometimes extra tRNAs, and they often possess intronic and large intergenic spacer regions. The Trichoplax mtDNA structurally resembles an intermediate between the large nonanimal and the compact animal mitochondrial genome. Similar to nonanimals, Trichoplax has a large, 43,079–base pair mitochondrial genome and extensive intergenic spacer regions, open reading frames (ORFs), and several introns, but, like all other animals, its genome also lacks ribosomal protein genes.
To understand the ancestral animal condition and the unique features that define Metazoa, the molecular and structural features shared between Trichoplax and its relatives must be investigated not only in this particular species but in the entire phylum Placozoa. To this end, we sequenced the mitochondrial genomes of three additional, highly divergent placozoans [17,18], and we here report on our findings. Briefly, all placozoan mtDNA sequenced resembled that of Trichoplax in that they possess very large, 32- to 37-kb, circular molecules, an identical set of respiratory chain genes and structural RNAs, as well as an intron group I LAGLIDADG endonuclease domain, conserved intron positions in cox1, and large intergenic spacer regions. The differences in genome size among the four placozoan mitochondria can be attributed to length variation in spacer regions and the presence or absence of long open reading frames. Our phylogenetic analyses support the monophyly of Placozoa as well as a basal placement among the Metazoa. Mitochondrial genome size reduction thus likely occurred after the emergence of animals.
Mitochondrial Genome Comparisons within Placozoa
Recent work has shown that Placozoa, once thought to be a monotypic taxon, is actually a phylum composed of no fewer than five highly divergent clades [17,18]. T. adhaerens is a member of Clade I . The complete mitochondrial genome sequences of three placozoan strains: BZ2423, BZ10101, and BZ49, belonging to Clades II, III, and V, respectively , were determined and analyzed. They are characteristically large, 32- to 37-kb, circular molecules that contain a common set of 12 respiratory chain genes (atp6, cob, cox1–3, nad1–6, and nad4L), two ribosomal RNA genes (rnL and rnS), and a full complement of 24 tRNA genes (having one extra trnL, trnM, trnR, and trnS). Figure 1 shows the linearized annotated maps of each mitochondrial genome sequenced in this study in addition to that of Trichoplax . Many common features were found in all four placozoan strains. cox1 was found to be distributed across at least six exons on both strands, while the large ribosomal RNA (rnL) was split into at least two segments, on the same strand. Another common feature found across all four placozoan mtDNAs was the presence of a large 501– to 677–amino acid ORF containing an intron group I LAGLIDADG endonuclease domain (LAG). Its position is conserved between cox1–2 and nad4 in all four placozoan mitochondrial genomes.
Each of the four heavy black lines represents the nucleotide sequence of a genome. Genes and RNAs are indicated by their names and color-coded rectangles. Genes transcribed in opposing directions are positioned to the right or left of the sequence line. Introns of cox1 and nad5 are denoted by thin black lines connecting each exon. The four mitochondrial genomes were divided into two groups based on their structural and phylogenetic similarity: BZ10101 and BZ49 (A) and Trichoplax and BZ2423 (B). Within each group, red and gray lines over the sequence indicate segmental translocations and insertions, respectively. The gray dashed lines across groups (A) and (B) indicate an inversion of the delineated region.
Each placozoan strain sequenced in this study was unique in its mitochondrial genome content and structure. Table 1 provides a general summary of major features with T. adhaerens included for comparison. Specifically, the largest placozoan mitochondrial genome belongs to Trichoplax, at just over 43 kb, followed by BZ2423 (36.7 kb), BZ49 (37.2 kb), and, finally, BZ10101, at 32.7 kb. The percentage of coding plus structural RNA sequences ranged from 55% (Trichoplax) to 67% (BZ49), and the median respiratory chain gene length varied from 949.5 bp (BZ10101) to 979.5 bp (Trichoplax). BZ10101 and BZ49 had the shortest median intergenic spacer length, at 101 and 105 bp, respectively, while BZ2423 and Trichoplax had the longest, at 154 and 209 bp, respectively. BZ10101 presented the lowest G + C content at both genic (coding and RNAs) and nongenic regions—36.4% and 44.4%, respectively, while Trichoplax had the highest—39.6% and 56.2%.
Summary of Mitochondrial Genome Features across Various Metazoa and the Choanoflagellate Monosiga
Structural and compositional differences between these four mitochondrial genomes are summarized in Figure 1. A major inversion, depicted by the dotted lines, between nad1 (or nad4L) and cox2 distinguished BZ10101 and BZ49 (group A) from Trichoplax and BZ2423 (group B). Within each group, A or B, two minor segmental inversions or translocations, depicted by the red lines, were also seen. A previously reported large ORF of unknown function containing a reverse transcriptase domain and an intron group II maturase domain (RVT-IM) found in Trichoplax  was also observed in the mtDNA of BZ49 and BZ2423. Unlike LAG, the position of RVT-IM was not conserved, being found between cox1-2a and cox1-2b in BZ2423 and Trichoplax (group B) and between cox1-5a and cox1-5b in BZ49 (group A). Amino acid sequence alignment of these RVT-IM genes indicated very poor similarity between BZ49 and group B strains (23% to 24% identity, excluding indel sites), while the similarity within group B, excluding indel sites, was 63%. This observation suggested that group A and B RVT-IM genes might have originated independently within these placozoans. We also found two positionally conserved exons for nad5 in the genomes of BZ49, BZ10101, and Trichoplax but an intronless nad5 in BZ2423.
A 491–amino acid ORF similar (BLASTX best match to the fungus, Hebeloma circinans, E = 6 × 10−55, see Dataset S3 for multiple sequence alignment) to the fungal DNA-directed DNA polymerase type B (polB) was identified in the mtDNA of BZ49. To further investigate the origins of this gene, primers designed to conserved regions of polB were used to attempt to amplify this gene in BZ10101, BZ2423, and Trichoplax, as well as in four additional placozoan isolates belonging to Clade V, to which BZ49 is also a member . The polB gene was detected in all members of Clade V but not in other isolates by this method. To determine whether or not the mitochondrial position of polB in BZ49 is conserved in this clade, we designed and tested primers targeted at the flanking regions of BZ49 polB. Results showed that one other strain, BZ931, contained polB in the same position as in BZ49, indicating this gene is located elsewhere in some placozoans from Clade V. This analysis could not distinguish whether the alternate polB location or locations were nuclear or mitochondrial.
Mitochondrial Genome Comparisons and Phylogeny of the Metazoa
Comparisons of certain features across metazoans and the choanoflagellate mitochondria are presented in Table 1. Placozoans have on average the greatest number of ORFs and introns per genome compared to cnidarians, which have up to two introns and ORFs, and poriferans and bilaterians, which have none. The choanoflagellate Monosiga has six ORFs and four introns: one in nad5 and three in cox1. Strains BZ49, BZ10101, and Trichoplax and all three cnidarians belonging to the subclass Hexacorallia (Acropora, Metridium, and Ricordea), listed in the table share a conserved intron position with Monosiga at nad5. Furthermore, cox1-(1,2,3) exons in group A animals (BZ49 and BZ10101) and LAG in all placozoan strains share conserved positions with Monosiga. Placozoan intergenic spacer regions also tended to be greater in number and longer than those of cnidarians, poriferans, and bilaterians. But, like all metazoans, placozoans lack ribosomal proteins.
Maximum likelihood (ML) and Bayesian analyses were performed on a dataset containing 12 concatenated protein sequences (atp6, cob, cox1–3, nad1–6, and nad4L) totaling 2,553 amino acids from 18 taxa (see Dataset S1). Both phylogenetic inference procedures returned the same well-supported tree topology for these taxa (Figure 2). Specifically, the data support a major split in the metazoan tree, with one group containing the lower metazoans and the other containing the bilaterians. These phylogenetic analyses also supported phylum Placozoa as the basal group within the lower Metazoa and, in addition, the two placozoan groups, A and B, that emerged from structural features described.
This phylogenetic tree is based on 2,553 amino acids from 12 concatenated respiratory chain genes (atp6, cob, cox1–3, nad1–6, and nad4L). Values above internal nodes represent Bayesian posterior probabilities, and those below represent bootstrap percentages under ML. The tree was rooted with the chytrid fungus Monoblepharella.
To determine whether or not the rate of evolution of the placozoan lineage differed significantly from that of the other lower metazoans used in this study, relative-rate tests  on each of the 12 respiratory chain genes were performed. Three of the 12 genes produced a significant p-value (<0.05), nad1, nad4, and nad5, indicating that these three genes are likely evolving at different rates between placozoans and the other lower metazoans. However, none of the rates were significantly different after controlling the overall significance level at 5% using the Bonferroni correction.
Sequencing three additional placozoan mtDNAs (BZ49, BZ10101, and BZ2423) allowed us to determine whether the unprecedented features uncovered by the first sequenced placozoan mitochondrial genome , that of T. adhaerens, were general phenomena within this phylum. A phylogenetic comparison of these complete placozoan mitochondrial genome sequences to other phyla gave further support for the placement of Placozoa as a basal lower metazoan phylum and provided evidence of the ancestral animal mtDNA condition.
The complete sequences of three additional placozoan mitochondrial genomes revealed that large, noncompact, circular molecules are indeed shared features among members of this phylum. All placozoan mtDNA encoded a common set of 12 respiratory chain genes, 24 tRNAs, two rRNAs, and a large 501– to 677–amino acid ORF containing a group I intron LAGLIDADG endonuclease domain. No atp8 gene was detected in any of the placozoan mtDNAs sequenced in this study, but because atp8 is known to be highly variable, we cannot conclusively rule out its presence. Other notable features common to all placozoan mitochondria were the lack of atp9 and the presence of multiple introns, several unknown ORFs, and relatively large intergenic spacer regions for metazoan genomes. Although their genome size varies considerably (e.g., Trichoplax has a mitochondrial genome more than 11 kb larger than that of BZ10101), the gene content, including ORFs of unknown function, across mitochondrial genomes did not show as much variation, ranging between 39 and 42 genes. Even accounting for extra genes, the mitochondrial genome of Trichoplax is still larger than any other placozoan mtDNA by at least 3.5 kb. The genome size variation among the placozoan mitochondria was, thus, mainly attributed to differences in intergenic spacer length.
The cox1 gene of Trichoplax was previously found to have an unusual fragmented structure, with exons encoded on different strands of the genome and a large number (at least five) of introns . This fragmentation appears to be a shared feature among all three placozoan strains examined in this study. All genomes contained up to seven cox1 exons arranged on both strands (Figure 1). Complex gene arrangements have also been observed in the ciliate Tetrahymena pyriformis mitochondrial genome , where nad1 is split into two fragments, one on each strand. Another unusual feature seen in the Trichoplax mitochondrial genome, the split in the large subunit ribosomal RNA, rnL, into at least two segments on one strand, was also confirmed to be present in other placozoans (Figure 1). Interestingly, the T. pyriformis mitochondrial genome contains a small subunit ribosomal RNA split into two segments . It should be noted, however, that further experimental evidence will be needed to determine whether or not these rnL segments in placozoans are spliced together or exist as separate gene segments.
Despite having a common set of respiratory chain genes and structural RNAs, these four placozoan mitochondrial genomes show substantial structural and molecular polymorphisms. While the gene order between BZ49 and BZ10101 and that between BZ2423 and Trichoplax were nearly identical, a major inversion between nad1 (or nad4L in BZ2423) and cox2 distinguished BZ49 and BZ10101 as a separate group from BZ2423 and Trichoplax. Molecular phylogenetics based on the 12 respiratory chain genes further supported this division (Figure 2). In addition, the large ORF containing a reverse transcriptase domain and a group II intron maturase domain (RVT-IM) was located in the same position in BZ2423 and Trichoplax but at a different mitochondrial genomic location in BZ49. The alignment of these genes revealed that the BZ49 RVT-IM was highly divergent from the other two (see Dataset S2). Because of this low sequence similarity, the mobile nature of group II introns, and the fact that BZ10101 lacks RVT-IM altogether, the origin of RVT-IM in BZ49 may have been independent of that of the RVT-IM present in BZ2423 and Trichoplax.
Phylogenetic analyses were performed using a concatenated amino acid dataset of 12 common respiratory chain genes spanning 18 taxa from the phyla Choanoflagellata, Porifera, Cnidaria, and Placozoa and four bilaterians. A split in the metazoan phylogeny (Figure 2) between two major clades, the Bilateria and lower Metazoa, was observed and may be a result of long-branch attraction due to faster rates of evolution in the Bilateria . Moreover, better resolution of the metazoan phylogeny may be obtained with the eventual addition of other lower metazoan taxa, such as members of the phylum Ctenophora. Nonetheless, this metazoan phylogeny is in agreement with our previous findings , supporting Placozoa as a basal lower metazoan phylum. The monophyly of Placozoa was strongly supported by both our comparative mtDNA and phylogenetic analyses. The relatively long branch leading to the phylum Placozoa (Figure 2) additionally suggests an ancient split from other lower metazoans. Relative-rate tests  did not show a significant difference in the rate of evolution between the placozoan lineage and the combined sponge and cnidarian lineage compared to the Monoblepharella and Monosiga outgroups. Furthermore, the branches leading to all placozoan strains are relatively short, indicating that diversification within Placozoa likely occurred in the recent past.
Now that some of the common features among placozoan mtDNA genomes have been described, we can begin to deduce the ancestral condition of the mitochondrial genome of all animals. For example, our data suggest that, contrary to conventional wisdom, the common ancestor of all animals actually possessed large, noncompact mitochondria, owing to the fact that both Placozoa and its closest nonanimal relative, Monosiga , have distinctively large mitochondrial genomes. Compaction of the mitochondrial genome likely occurred secondarily after the emergence of Metazoa. Our data support the hypothesis that loss of ribosomal protein genes from the mitochondrial genome is a metazoan synapomorphy, as no animal mitochondrial genome sequenced so far, including the large mtDNAs of placozoans described here, has identified sequences coding for ribosomal proteins.
In terms of mitochondrial genome size, structure, and composition, placozoan mitochondrial genomes appear to be intermediates between the very large protist and fungal mtDNAs and the compact animal mtDNAs. Like choanoflagellates and fungi mtDNAs, placozoan mitochondrial genomes have larger protein coding sequences than other lower metazoans and many large intergenic spacers and introns (Table 1). Not only is the nad5 intron position in placozoans conserved in Monosiga and cnidarians, but the positions of the first three cox1 exons in BZ49 and BZ10101 and the position of LAG in all four placozoan genomes are conserved in Monosiga. Placozoan mitochondrial genomes share other features similar to most animal mitochondria—predominantly, the lack of all ribosomal protein sequences and the atp9 gene only found in Porifera [5,23]. Placozoan mtDNAs have some unique features as well. The absence of atp8 in all four placozoans is likely a synapomorphy for this phylum given its presence in all other taxa sampled in this study (atp8 is also absent in nematodes and some mollusks; reviewed in ). An unusual feature discovered in the BZ49 genome was the presence of a DNA-directed DNA polymerase type B, polB. This gene is found in plant and fungal  mtDNAs but has only once  been previously observed in animal mitochondrial sequences, in the linear mtDNA of the moon jelly Aurelia aurita. However, amino acid sequence alignment of these two animal mitochondrial polB genes shows highly diverged sequences, with only 21% identity. Furthermore, the fragmentation and distribution of cox1 and rnL on both placozoans mtDNA strands are features common only to placozoans and no other animal mitochondrial genome sequenced thus far.
Since the Placozoa is a phylum of at least five highly divergent mitochondrial clades [17,18] and our work represents only four of the eight identified 16S rRNA haplotypes, it is likely that by sequencing other mitochondrial genomes we will gain additional insights into both placozoan and metazoan evolution. In particular, with more placozoan taxa sampled, we may be able to determine which taxon belongs to the basal-most placozoan lineage and thus be able to better reconstruct the order of evolutionary events that lead to the diversity of placozoan mitochondrial genomes we observe today. The findings presented here used bioinformatically inferred genes to infer structural and phylogenetic features and, as such, will need experimental confirmation. The resources generated through this study will facilitate experimental studies of placozoan mitochondria. One gene of special interest is cox1, which was found arranged on both strands. If cox1 is experimentally proved to be a viable gene, then the mechanism allowing this protein or proteins to be functional will need to be addressed.
The comparative work presented here demonstrated that animal mitochondrial genomes can be large, not due to repetitive sequences or duplicated regions but due to numerous intergenic spacers, introns, and ORFs of unknown function. The phylum Placozoa, as far as it has been sampled, represents a group of animals with uniquely large mitochondrial genomes. Comparative mitochondrial genomics, both among lower metazoans and within the Placozoa, has proved to be extremely useful in resolving deep phylogenetic relationships. This shared large mitochondrial genome size and basal phylogenetic position among placozoans raises the question of what has allowed the placozoan mitochondrial genomes to remain so large when all other animal mitochondrial genomes have been drastically compacted.
Materials and Methods
Placozoan strain selection, cloning, and sequencing.
Three placozoan strains originally isolated from Twin Cays, Belize , were chosen for whole mitochondrial genome sequencing: BZ2423, BZ10101, and BZ49. These strains were selected because each represents a different and highly divergent mitochondrial clade [17,18]. Strain BZ2423 belongs to Clade II, BZ10101 belongs to Clade III, and BZ49 belongs to Clade V. Because no formal species description exists for any placozoan species other than T. adhaerens, here we referred to the three strains used in our study by their laboratory identification numbers: BZ2423, BZ10101, and BZ49. Each strain was clonally maintained in laboratory cultures, and 2 to 5 μg of total genomic DNA was isolated according to protocols in Signorovitch et al. .
Cloning and subcloning of the three placozoan mitochondrial genomes were each carried out as described in the protocols of Dellaporta et al.  using pCC1FOS and pSMART LC-Kan (Lucigen, http://www.lucigen.com) vectors, respectively. For each strain, we amplified approximately 384 random subclones by PCR in a 25-μl reaction volume, using the manufacturer's forward and reverse primers, Taq polymerase (Qiagen, http://www.qiagen.com), and the following conditions: 95 °C denaturation for 10 min; 30 cycles of 95 °C for 30 s, 57 °C for 30 s, and 72 °C for 4 min; and a 72 °C final extension for 10 min. The amplified products were purified by precipitation using an equal volume of 20% PEG-8000/2.5 M NaCl and then resuspended in 25 μl of 10 mM Tris-Cl (pH 8.0). Purified products were sequenced using PCR primers and TaqFS dye-terminator cycle-sequencing reactions on Prism 3730 DNA sequencers (Applied Biosystems, http://www.appliedbiosystems.com) at the W. M. Keck DNA Sequencing Facility (Yale University) and Genaissance Pharmaceuticals (New Haven, Connecticut, United States).
Certain regions of the mitochondrial genomes of strains BZ10101 and BZ49 contained secondary structures, such as GC-rich hairpin loops, that proved difficult to sequence using the regular methods outlined above. Subclones spanning these regions were selected and plasmid purified from each using the Qiagen Plasmid Purification kit. We used a two-pronged approach to resolving these poor-quality regions. We first attempted sequencing of the purified plasmid using dGTP BigDye polymerase (Applied Biosystems) and the manufacturer's primers. If this first alternative method did not succeed in producing good-quality sequences, we amplified the problematic insert (usually 2 to 4 kb) using Herculase II Fusion (Stratagene, http://www.stratagene.com) polymerase and then sheared the PCR product (approximately 10 μg) by sonication to fragments between 0.5 and 1 kb. The sheared DNA was then end-repaired and cloned into pSMART LC-Kan vector as described above. Approximately 48 random clones were sequenced using the dGTP Big Dye polymerase.
The DNA sequences of each strain were assembled separately using the Phred, Phrap, and Consed package, release 15.0 [27–29]. Potential genes were identified using the National Center for Biotechnology Information's ORF FINDER, using the Mold, Protozoan, and Coelenterate Mitochondria genetic code. For ORFs of unknown function, only those greater than 100 bp and not overlapping other known genes (i.e., respiratory chain subunits or structural RNAs) were annotated. tRNAs were inferred using the program tRNAscan-SE 1.21 (http://lowelab.ucsc.edu/tRNAscan-SE). Each of the 12 placozoan inferred gene sequences (atp6, cob, cox1–3, nad1–6, and nad4L) was aligned to its homologous sequence in sponges (Axinella corrugata, Geodia neptuni, and Tethya actinia), cnidarians (Metridium senile, Acropora tenuis, Aurelia aurita, Briareum asbestinum, and Ricordea florida), bilaterians (Artemia franciscana, Katharina tunicata, Saccoglossus kowalevskii, and Strongylocentratus purpuratus), the choanoflagellate Monosiga brevicollis, and the chytrid fungus Monoblepharella using the program CLUSTALW  and edited manually, in order to predict translational start sites and intron-exon boundaries. These alignments were edited using Gblocks, version 0.91b , so as to exclude gaps and large nonconserved regions, and the output alignments were concatenated for phylogenetic analyses. Two likelihood-based phylogenetic inference procedures were employed: ML and Bayesian, and both used the mtREV model of amino acid substitution. The program PHYML v2.4.4 [32,33] was used to run the ML analysis and to obtain bootstrap support values (4,000 replicates). The Bayesian analysis was run in MrBayes [34,35], and posterior probabilities were obtained after 500,000 generations with a burn-in of 25%. All parameter values were the same as in Dellaporta et al. . Relative-rate tests were performed on each of the 12 respiratory chain genes individually using the program RRTree . Two lineages were defined: one composed of only placozoan sequences and the other composed of sponge and cnidarian sequences. The outgroup contained Monoblepharella and Monosiga sequences. A guide tree topology obtained from MrBayes was used in RRTree.
Amplification of polB.
Using strain BZ49 as the reference in designing primer pairs for PCR, we checked for the presence of polB in BZ10101, BZ2423, and T. adhaerens  and four other Clade V strains (BZ931, BZ322, BZ42, and JM614). Two types of amplification reactions were performed. The first reaction used primers designed to amplify a flanking segment of mitochondrial DNA that included polB in BZ49. These reactions used PCR amplification with the use of Herculase II Fusion polymerase and 2% DMSO with the forward 5′-GCTGCAATGGAGGTTGTTTT-3′ and the reverse 5′-ACACCATTTTAAACCCCACCAATC-3′ primer pair and the following PCR conditions: 98 °C for 4 min; 30 cycles at 98 °C for 20 s, 57 °C for 20 s, and 72 °C for 1.5 min; and 72 °C for 3 min. The second reaction was designed to amplify an internal conserved region of polB using forward 5′-TCTAAAGATGTTTGATGTGCACTTTT-3′ and the reverse 5′-TTTTGGGCGTTTTTCAACTCTTCT-3′ primer pair and used the same PCR conditions as above.
Dataset S1. Concatenated Amino Acid Sequence Data of 12 Respiratory Chain Genes
(5.5 MB PDF)
Dataset S2. Amino Acid Sequence Alignment of RVT-IM
(68 KB PDF)
Dataset S3. Amino Acid Sequence Alignment of polB
(93 KB PDF)
The GenBank (http://www.ncbi.nlm.nih.gov/Genbank) accession numbers for the genes and gene products discussed in this paper are BZ2423, BZ10101, and BZ49, belonging to Clades II, III, and V, respectively (DQ889458, DQ889456, and DQ889457, respectively), A. corrugata (NC_006894), G. neptuni (NC_006990), T. actinia (NC_006991), M. senile (NC_000933), A. tenuis (NC_003522), A. aurita (NC_008446), B. asbestinum (NC_008073), R. florida (NC_008159), A. franciscana (NC_001620), K. tunicata (NC_001636), S. kowalevskii (NC_007438), S. purpuratus (X12631), M. brevicollis (NC_004309), and Monoblepharella (NC_004624).
We are grateful to Maria Moreno for critical discussions on this manuscript and various laboratory techniques. We thank Rafael Rosengarten, James Signorovitch, Anthony Xu, Ivan Acosta, Matthew Nicotra, Anahid Powell, Sabrina Rosa, and Christina Glastris for helpful suggestions and critical reading of this manuscript. Chad Kirtzberger, Zack Snable, and Aaron Thier helped with animal care.
LWB and SLD conceived and designed the experiments. AYS performed the experiments. AYS, LWB, and SLD analyzed the data and wrote the paper.
- 1. Bridge D, Cunningham CW, DeSalle R, Buss LW (1995) Class-level relationships in the phylum Cnidaria: Molecular and morphological evidence. Mol Biol Evol 12: 679–689.
- 2. Paquin B, Laforest MJ, Forget L, Roewer I, Wang Z, et al. (1997) The fungal mitochondrial genome project: Evolution of fungal mitochondrial genomes and their gene expression. Curr Genet 31: 380–395.
- 3. Lang BF, O'Kelly C, Nerad T, Gray MW, Burger G (2002) The closest unicellular relatives of animals. Curr Biol 12: 1773–1778.
- 4. Bullerwell CE, Forget L, Lang BF (2003) Evolution of monoblepharidalean fungi based on complete mitochondrial genome sequences. Nucl Acids Res 31: 1614–1623.
- 5. Lavrov DV, Forget L, Kelly M, Lang BF (2005) Mitochondrial genomes of two demosponges provide insights into an early stage of animal evolution. Mol Biol Evol 22: 1231–1239.
- 6. Dellaporta SL, Xu A, Sagasser S, Jakob W, Moreno MA, et al. (2006) Mitochondrial genome of Trichoplax adhaerens supports Placozoa as the basal lower metazoan phylum. Proc Natl Acad Sci U S A 103: 8751–8756.
- 7. Medina M, Collins AG, Takaoka TL, Kuehl JV, Boore JL (2006) Naked corals: Skeleton loss in Scleractinia. Proc Natl Acad Sci U S A 103: 9096–9100.
- 8. Boore JL (1999) Animal mitochondrial genomes. Nucl Acids Res 27: 1767–1780.
- 9. Gray MW, Lang BF, Burger G (2004) Mitochondria of protists. Annu Rev Genet 38: 477–524.
- 10. Ruthmann A, Wenderoth H (1975) DNA measurements on cells of primitive metazoon Trichoplax adhaerens F. E. Schulze. Cytobiologie 10: 421–431.
- 11. Ruthmann A (1977) Cell-differentiation, DNA content and chromosomes of Trichoplax adhaerens Schulze, F. E. Cytobiologie 15: 58–64.
- 12. Birstein VJ (1989) On the karyotype of Trichoplax sp. (Placozoa). Biologisches Zentralblatt 108: 63–67.
- 13. Boyce TM, Zwick ME, Aquadro CF (1989) Mitochondrial-DNA in the bark weevils: Size, structure and heteroplasmy. Genetics 123: 825–836.
- 14. Azevedo JLB, Hyman BC (1993) Molecular characterization of lengthy mitochondrial-DNA duplications from the parasitic nematode Romanomermis culicivorax. Genetics 133: 933–942.
- 15. Fuller KM, Zouros E (1993) Dispersed discrete length polymorphism of mitochondrial-DNA in the scallop Placopecten magellanicus (Gmelin). Curr Genet 23: 365–369.
- 16. Raimond R, Marcade I, Bouchon D, Rigaud T, Bossy JP, et al. (1999) Organization of the large mitochondrial genome in the isopod Armadillidium vulgare. Genetics 151: 203–210.
- 17. Voigt O, Collins AG, Pearse VB, Pearse JS, Ender A, et al. (2004) Placozoa: No longer a phylum of one. Curr Biol 14: R944–R945.
- 18. Signorovitch AY, Dellaporta S, Buss L (2006) Caribbean placozoan phylogeography. Biol Bull 211: 149–156.
- 19. Robinson-Rechavi M, Huchon D (2000) RRTree: Relative-rate tests between groups of sequences on a phylogenetic tree. Bioinformatics 16: 296–297.
- 20. Edqvist J, Burger G, Gray MW (2000) Expression of mitochondrial protein-coding genes in Tetrahymena pyriformis. J Mol Biol 297: 381–393.
- 21. Schnare MN, Heinonen TYK, Young PG, Gray MW (1986) A discontinuous small subunit ribosomal RNA in Tetrahymena pyriformis mitochondria. J Biol Chem 261: 5187–5193.
- 22. Lang BF, O'Kelly C, Nerad T, Gray MW, Burger G (2002) The closest unicellular relatives of animals. Curr Biol 12: 1773–1778.
- 23. Lavrov DV, Lang BF (2005) Transfer RNA gene recruitment in mitochondrial DNA. Trends Genet 21: 129–133.
- 24. Mouhamadou B, Barroso G, Labarere J (2004) Molecular evolution of a mitochondrial polB gene, encoding a family B DNA polymerase, towards the elimination from Agrocybe mitochondrial genomes. Mol Genet Genomics 272: 257–263.
- 25. Shao ZY, Graf S, Chaga OY, Lavrov DV (2006) Mitochondrial genome of the moon jelly Aurelia aurita (Cnidaria, Scyphozoa): A linear DNA molecule encoding a putative DNA-dependent DNA polymerase. Gene 381: 92–101.
- 26. Signorovitch AY, Dellaporta SL, Buss LW (2005) Molecular signatures for sex in the Placozoa. Proc Natl Acad Sci U S A 102: 15518–15522.
- 27. Ewing B, Hillier L, Wendl M, Green P (1998) Basecalling of automated sequencer traces using Phred. I. Accuracy assessment. Genome Res 8: 175–185.
- 28. Ewing B, Green P (1998) Basecalling of automated sequencer traces using Phred. II. Error probabilities. Genome Res 8: 186–194.
- 29. Gordon D, Abajian C, Green P (1998) Consed: A graphical tool for sequence finishing. Genome Res 8: 195–202.
- 30. Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673–4680.
- 31. Castresana J (2000) Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol 17: 540–552.
- 32. Guidon S, Lethiec F, Duroux P, Gascuel O (2005) PHYML online: A Web server for fast maximum likelihood-based phylogenetic inference. Nucl Acids Res 33: W557–W559.
- 33. Guidon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52: 696–704.
- 34. Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17: 754–755.
- 35. Ronquist F, Huelsenbeck JP (2003) MRBAYES 3. Bioinformatics 19: 475–481.