Molecular Phylogeny Inferred from Sequences of Small Subunit Ribosomal DNA, Supports the Monophyly of the Metazoa

7 Pages·1993·2.5 MB·English

by  Mari Kobayashi

ZOOLOGICAL SCIENCE 10: 827-833 (1993) 1993 ZoologicalSocietyoflapan
Molecular Phylogeny Inferred from Sequences of Small
Subunit Ribosomal DNA, Supports the
Monophyly of the Metazoa
Mari Kobayashi1 Miyuki Takahashi2 Hiroshi Wada3
, ,
and Noriyuki Satoh3
1Department ofBiology, Faculty ofScience, Ochanomizu University,
Bunkyo-ku, Tokyo 112,2Department ofBiology, Faculty of
Science, Kagoshima University, Kagoshima 890, and
^Department ofZoology, Faculty ofScience,
Kyoto University, Kyoto 606-01, Japan
—
ABSTRACT Recentstudiesofmolecularphylogenybaseduponcomparisonsofthepartialnucleotide
sequencesof18and28SrRNAshavesuggestedthatthemetazoaarepolyphyletic,i.e.,thatdiploblasts
(poriferans. cnidarians, ctenophores, andplacozoans) andtriploblastsformtwoseparatedmonophyle-
tic units. In orderto examine this hypothesis, we determined almost the complete sequencesofsmall
subunit (18S-like) rDNA for two poriferans and a ctenophore. Phylogenetic comparisons of the
sequences, together with those of a cnidarian, triploblasts and other eukaryotes, supported the
monophyly of the metazoa. Among the diploblasts, the ctenophore showed some similarities to the
poriferans.
tures (phylum Platyhelminthes), from which va-
INTRODUCTION
rious modern metazoans have diverged. Current
On the basis ofextensive studies ofcomparative debate concerns whether the metazoa are
anatomy, embryology, and paleontology, mul- monophyletic or polyphyletic and what the ances-
ticellular animals (metazoans) have been divided tral metazoa were like. Some investigators have
into three majortaxonomic units, namely; animals insisted that poriferans arose independently from
with three embryonic layers and bilateral symmet- colonial flagellates, whereas other metazoans
ry (triploblasts), those with two embryonic layers evolved from syncytial ciliates [8, 25].
and radial symmetry (diploblasts) except for bi- Advances in molecular biology have provided
lateral symmetry of anthozoans, and those with some answers to the problemsposed byevolution-
extremely loose tissue differentiation (poriferans), ary biologists. Comparisons based on molecular
the latter sometimes included within diploblasts data, such as the amino acid sequences of certain
[cf. 1. 2, 28]. The origin and evolution of these proteins and the nucleotide sequences of certain
multicellular animals have received considerable RNAs and DNAs, provide powerful tools with
attentionoverthe years [1, 2, 8-10, 15, 21,25,28]. which the phylogenetic relationships among ani-
It is generally accepted that the metazoa arose mal groups are deducible more objectively than
from protozoa, perhaps 700-1000 million years others [13, 20]. In particular, the small ribosomal
ago [3]. Models of the metazoan origin suggest subunit (16 or 18S) RNA or its gene (16 or 18S
that either colonial flagellates [9, 15] or syncytial rDNA) is ideallysuited forphylogeneticstudies of
ciliates [8. 10] became acoel flatworm-like crea- distantly related organisms, because it is rich in
information and the sequencing methodology per-
Accepted July 30, 1993 mits the rapid accumulation of large databases [5,
Received April 16. 1993 19]. A recent study ofthe molecular phylogeny of
828 M. Kobayashi, M. Takahashi et al.
metazoans based upon comparisons of sequences described previously [26]. In brief, frozen and
of 18S rRNAs has suggested that the metazoa are powderedsampleswerelysedinTEbuffer(10mM
polyphyletic; Cnidarians (diploblasts) arosefrom a Tris-HCl, 0.1 M EDTA, pH8.0) that contained
protist ancestry different from the Bilateria (tri- 0.5% sodium dodecyl sulfate. After digesting
ploblasts) [7]. In addition, analysis ofthe origin of sampleswith proteinase K (100,«g/ml) at 50°Cfor
DNA
metazoans, using comparisons ofsequences of28S 3 hr, wasextractedwithphenol, andprecipi-
M
rRNAs, has demonstrated that triploblasts and tated in ethanol and an equal volume of 5.0
diploblasts form two clearly separated units [4]. ammonium acetate. Samples resuspended in TE
However, the suggestion of polyphyletic origin of bufferwere further purified by RNase A digestion
metazoans from the above-mentioned studies [4, (20,ug/ml) at 37°C for 1 hr followed by ethanol
7], was based upon comparisons of partial nuc- precipitation.
leotide sequences; about 840basesof18and about
290 bases of 28S rRNAs, respectively. Amplification ofsmall-subunitrDNA andsequenc-
DNA
Very recently, Wainright etal. [27] reported the ing ofthe amplified
evolutionary origin and early branching pattern of Almost the entire length of the 18S rDNA was
the animal kingdom. Theyinferredthephylogene- amplified by the polymerase chain reaction (PCR)
tic framework by comparing small subunit ribo- [22] in a thermal cycler (Perkin-Elmer Corp.,
somal RNA sequences fortwoporiferans, acnida- Norwalk, CT, USA) usingprimers (5'CTGGTT-
rian, a ctenophore, and a placozoan. The authors GATCCTGCCAG3') and 10 [5'CACCTAC-
reported that the animal lineage is monophyletic GGA(AT)ACCTTG3']. These primers were de-
and that animals and fungi share acommon evolu- signed referring to conserved regions of aligned
tionary history. We also examined the phylogene- eukaryote 18S rDNA sequences, with which 18S
tic position ofdiploblastic animals (two poriferans rDNA ofdeuterostomes could be amplified. Am-
mM
and a ctenophore). The species we examined plification proceeded in 50/A of 50 KC1, 10
mM
however differed from those studied by Wainright Tris-HCl (pH9.0), 0.1% Triton X-100, with
mM
et al. [27]. 0.2 each dNTP, 50pmol primers, template
DNA (5-50fj.g) and 1 U Taq DNA polymerase
(Toyobo Co. Ltd., Osaka, Japan). One of the
MATERIALS AND METHODS
primers was kinased prior to PCR at the 5' termi-
nal phosphate. The temperature regimen for 30
Animals
cycleswas 1 min at 94°C, 2min at42°C, and 3min
Diploblasts examined in the present study were at 72°C.
two poriferans Sycon calcaravis Hozawa (class Amplified DNA was purified by electrophoresis
Calcarea, order Sycettida) and Tetilla japonica in a 0.8% agarose gel and treated with lambda
Lampe (class Demospongiae, subclass Tetractino- exonuclease (BRL, Bethesda, MD, USA) to
DNA
morpha, order Spirophorida), and a ctenophore obtainsingle-stranded [11]. With thesingle-
DNA
BeroecucumisFabricius(classAtentaculata, order stranded as the template, the nucleotide
Beroidea). The poriferans were collected in the sequence was directly determined by dideoxy
vicinity of the Tateyama Marine Laboratory, chain-termination [24] using Sequenase ver 2.0
Ochanomizu University, and the ctenophore in (United States Biochemical Corp., Cleveland,
Kagoshima Bay. Eggs of T. japonica and whole OH, USA) and [35S]-dATP (Amersham Japan,
animals of S. calcaravis and B. cucumis were Tokyo, Japan). In addition to primers and 10,
frozen quickly in liquid nitrogen and kept at primers 8 (5'CCGGAGAGGGAGCCTGA3'), 7
—80°C until use. (antisense analogofprimer8), 1 (5'CAGCAGCC-
GCGGTAATT3'), 9 (antisense analog of primer
Isolation ofDNA 1), 3 (5'GCGAAAGCATTTGCCAA3'), 4 (anti-
High-molecular-weight genomic DNA was ex- sense analog of primer 3), 5 [5'GAAACT(TQ-
tracted from the frozen samples by the method AAAGGAAT3'], 6 (antisense analog of primer
Molecular Phylogeny of Diploblasts 829
5), and2 [5'ACGGGCGGTGTGT(AG)C3'] were bootstrapping value (83%).
used for sequencing. As shown previously, se- Because the monophyly of the metazoans was
quences of 18S rDNA can be determined more demonstrated in Fig. 1, we further analyzed the
accurately than those of 18S rRNA [26]. phylogeneticrelationshipsbetweendiploblastsand
triploblasts by adding data from the flatworm
Comparison ofsequences and inferences regarding Schistosoma mansoli [5] and by taking the yeast
phylogeny Saccharomyces cerevisiae as an outgroup. About
Sequences were aligned on the basis of max- 60 confidently aligned sites were added in this
imum nucleotide similarity. Using the aligned analysis. These sites were relatively variable and
sequences, we calculatedthe evolutionarydistance contained information usefulto increase the preci-
valuesinapair-wisemanner, asdescribedbyJukes sion of analysis. The structural similarity and
and Cantor [16]. The phylogenetic tree was con- evolutionary distance values of these organisms
structed from an analysis of results by the neigh- are shown in Table 1. The phylogenetic tree
bor-joining method of Saitou and Nei [23]. The constructed using the distance values, is shown in
degree ofsupport for internal branches ofthe tree Fig. 2. The tree suggested that the two poriferan
was further assessed by bootstrapping [6]. and a ctenophore species form one unit, while the
cnidarian species and triploblasts form another.
The branching ofthe former group was supported
RESULTS
by the relatively higher value of bootstrapping
Almost the entire length of the 18S rDNA of (62%).
two species of poriferan and a species of cte-
nophore was amplified by PCR, and the complete
DISCUSSION
nucleotide (more than 1600 nucleotide) sequences
ofthe amplifiedproducts ofPCR, exceptforthe5' The present molecularphylogenetic study based
and 3' termini, were determined directly without upon comparisonsofvirtuallycomplete nucleotide
cloning. The nucleotide sequence data reportedin sequences of 18S rDNAs showed that the metazoa
this study will appear in the DDBJ, EMBL and are of monophyletic origin. Although this study
GenBank Nucleotide Sequence Database under did not include placozoa, the grouping of the
the following accession number; D15066forSycon metazoans as a discrete taxonomic unit was sup-
calcaravis, D15067 for Tetilla japonica, and ported by the high value obtained by bootstrap-
D15068 for Beroe cucumis. ping. This result corroborates the results ofWain-
The nucleotide sequences of the three diplo- rightetal. [27] butnottheprevioussuggestion that
A
blastswerealignedwiththoseofAnemoniasulcata, the metazoa are polyphyletic. study of the
Anemia salina.. Homo sapiens and other eukary- molecular phylogeny of the animal kingdom
otes [5, 19] on the basis of maximum nucleotide pioneered by Field et al. [7] suggested that cnida-
similarity; the alignment of 1269 nucleotides was rians arise from a protist ancestry different from
analysed. Figure 1 shows a phylogenetic tree of a the bilaterians. However, Lake [18] has studied
variety of organisms, constructed by neighbor- the animal phylogeny by another systematic
joining [23]. The tree clearly indicated that the method, usingthesequence datareportedbyField
metazoans form an independent, monophyletic et al., and he concluded that the metazoa are a
unit. In other words, the four species of diplo- monophyletictaxon. An analysisbyChristenetal.
blasts and two species of triploblasts formed a [4] suggested that triploblasts and diploblasts form
discrete unit of monophyletic origin. This group- two clearlyseparatedmonophyleticunits. Howev-
ingofthe metazoawassupported bythehighvalue er, their main conclusion was that the origin of
obtained by bootstrapping (73.8%) [6]. The tree triploblasts is much more ancient, with respect to
also suggested that animals and fungi share an diploblasts, than was classically assumed.
evolutionary history; the branching of the two The evolutionary pathway from unicellular
groups from the others was supported by the high organisms to multicellular animals has been a
830 M. Kobayashi, M. Takahashi et al.
Oxytricha nova
100 —
Paramecium tetraurelia
Sarcocystismuris PROTOZOA
95.2 —
Theileria annulata
100
Prorocentrum micans
- Ochromonasdanica
r— 91.8
Tribonema aequale
100
55.8 Achlya bisexualis
Arabidopsisthaliana PLANTS
100
Zamiapumila
100
94.6 Chlamydomonasreinhardtii
74.4
Cryptomonasphi
Saccharomyces cerevisiae
i— 93.4
Schizosaccharomycespombe
98.8 99.2 FUNGI
Filobasidiella neoformans
r 61.8
Blastocladiella emersonii
Homosapiens
100
83.0
Artemia salina
I— 55.0 TRIPLOBLASTS
Anemonia sulcata
METAZOA
73.8
Beroe cucumis
DIPLOBLASTS
53.2
Sycon calcaravis
49.6
Tetillajaponica
0.05unit
Fig. 1. General phylogenetic tree of eukaryotes, as deduced by neighbor-joining using the fraction of observed
substitution differences of confidently aligned sites. The scale bar indicates the evolutionary distance of 0.05
nucleotide substitutionspersequenceposition. Numbersateachbranchindicate thepercentageoftimesanode
wassupportedin500bootstrappseudoreplicationsbyneighbor-joining. Thetreedemonstratedthemonophyletic
origin ofmetazoans, including both diploblasts and triploblasts.
This is an unrooted tree, because we could not include any outgroup organisms. Therefore, we cannot
determine where the root ofthis tree is and what the ancestor ofthe organisms is.
Molecular Phylogeny of Diploblasts 831
Table 1. Structuralsimilarityandevolutionarydistancedatafor 18SrDNAsequencesofmetazoansand
veast
Species Sy.ca. Te.ja. Be.cu. An.su. Sc.ma. Ar.sa. Ho.sa. Sa.ce.
Sycon calcaravis 0.0675 0.0749 0.0741 0.1722 0.1527 0.1554 0.0917
Tetilla japonica 86 0.0900 0.0858 0.1828 0.16865 0.1610 0.0951
Beroe cucumis 95 113 0.0875 0.1894 0.1789 0.1875 0.1071
Anemonia sulcata 94 108 110 0.1713 0.1564 0.1564 0.0917
Schistoma mansoni 205 215 223 204 0.1408 0.1345 0.1846
Anemia salina 184 201 212 188 171 0.1203 0.1808
Homo sapiens 187 193 221 188 164 148 0.1837
Saccharomyces cerevisiae 115 119 133 155 218 214 217
The lower-left half of the table gives the number of substitutions in which gaps are not included. The
upper-righthalfofthetablegivestheevolutionarydistancevalues(averagenumberofnucleotidesubstitutions
per sequence position) determined by the Jukes and Cantor (1969) [16] formula. The sequence data forA.
sulucata, S. mansoni, A. salina, H. sapiens and S. cerevisiae were obtained from references 5 and 19.
Saccharomyces cerevisiae
Schistoma mansoni
100
Homosapiens
L 51.0
52.0 Artemia salina
Anemonia sulcata
98.6
Beroe cucumis
47.2
Sycon calcaravis
62.0
Tetilla japonica
0.05unit
Fig. 2. Phylogenetic relationships ofdiploblastswith triploblasts (Schistosoma mansoni, Artemiasalina, and Homo
sapiens) asdeduced by reference to yeast as the out group. The tree wasconstructed by neighbor-joining. The
scale bar indicates an evolutionary distance of0.05 nucleotide substitutions persequence position. Numbers at
each branch indicate the percentage of times a node was supported in 500 bootstrap pseudoreplications by
neighbor-joining. The ctenophore Beroe forms a unit with poriferans.
subject of extensive investigation and vigorous and 2 implies two divergences of the metazoans;
discussion for more than a century [1. 2, 8-10, 15, one evolved a variety of diploblasts and the other
21, 25, 28]. The emergence ofmetazoanshasbeen that occurred later, evolved a variety of triplo-
explained by two major theories; the syncytial blasts. These two divergences are also suggested
theory (from a multinucleated ciliate) [8, 10] and from paleontology [3]. The former is the Vendian
the colonial theory (from a colonial flagellate) [9, radiationwhichgaverisetoEdiacaranassemblage,
15]. Although the present molecular data as well and the latter is the Cambrian explosion. The first
as those shown byFieldetal. [4] andChristen etal. Vendian radiation may have occurred soon after
[7] do not allow clarification of this point, the multicellular animals derived from other protists.
branching among the metazoans shown in Figs. 1 The rapid explosion, together with the relatively
1
832 M. Kobayashi, M. Takahashi et al.
higher rate of substitutions observed in the se- thank an anonymous reviewer for his (or her) construc-
quences of 18S rDNA of triploblasts, made it tivesuggestioninpreparationofthemanuscript. H.W. is
difficult to elucidate the affinity between triplo- supportedbyaJSPSPredoctoralFellowshipforJapanese
Junior Scientists with a research grant (2447).
blasts and diploblasts. These may be the reasons
why previous studies of partial 18 and 28S rRNA
sequences could not show the monophyly of the REFERENCES
metazoa.
A phylogenetic tree constructed from the 5S 1 Barnes RD (1987) Invertebrate Zoology. 5th ed,
Saunders College Publ, Fort Worth, pp58-163
rRNA sequences suggested a slightly earlier e- 2 Brusca RC, Brusca CJ (1990) Invertebrates.
mergence of platyhelminthes (Dugesia japonica; Sinauer Assoc Inc Publ, Sunderland, pp97-277
Tricladida) and nematodes (eg, Caenorhabditis 3 Conway Morris S (1993) The fossil record and the
elegans) than those of other metazoans, including earlyevolutionoftheMetazoa.Nature361:219-225
diploblasts and other bilateralia [14]. However, 4 Christen R, Ratto R, Baroin A, Perasso R, Grell
that view was not supported by a recent compari- KG, Adoutte A (1991) An analysis ofthe originof
sonofpartial 18SrDNAsequencesbyKatayamaet tmheeta2z8oSa,RNuAsi,ngrecvoemaplasrainsoenarslyofempaerrtgiaelncseeqoufetnrciepslobo-f
al. [17]. Instead, Katayama et al. [17] suggested lasts. EMBO J 10: 499-503
the early emergence of the acoel turbellarians in 5 De RijkP, NeefsJ-M, Vande PeerY, DeWachter
metazoic evolution; they might be some of the R (1992) Complication of small ribosomal subunit
closest multicellular animals to the metazoan RNA sequences. Nucl Acids Res 20: 2075-2089
ancestors. Although Fig. 2 indicates agroupingof 6 Felsenstein J (1985) Confidence limits on phy-
logenesis: an approach using the bootstrap. Evolu-
Schistoma mansoni (Platyhelminthes, Trematoda)
tion 39: 783-791
and the two other triploblasts, the phylogenetic 7 Field KG, Olsen GJ, Lane DJ, Giovannoni ST,
relationship between the diploblasts and the acoel Gheselin MT, Raff EC, Pace NR, Raff RA (1988)
turbellarians should be subject of further study Molecular phylogeny of the animal kingdom. Sci-
based upon the comparison of almost complete ence 239: 748-753
sequences of 18S rDNA. 8 Hadzi J (1963) The Evolution of Metazoa. Perga-
mon Press, Oxford, pp1-499
Anothersuggestionofthepresentstudy,wasthe
9 Haeckel E (1874) The gastraea-theory, the phy-
phylogeneticpositionofthectenophore. Classical- logenetic classification of the animal kingdom and
ly ctenophores were positioned closely to cnida- thehomologyofthegerm-lamellae. QJMicroscSci
rians [15]. However, the presence of true mesen- 14: 142-165
chymal muscle cells and gonoducts no longer sup- 10 Hanson ED (1977) Origin and Early Evolution of
Animals. Wesleyan Univ Press, Middletown, pp1-
ports the supposed affinity but suggests its closer
670
Friegl.at2i,onhsohwipesvetro,pltahteyhperlemsienntthsan[a1l2y]s.isAssuegvgiedsetnetdian 1 sHtirgauncdheidRDGN,AOcthemmaplnatHes(1b9y89e)xoPnruocdluecaisnegodfigseisntgiloen-
close molecular similarity between the poriferans followingthepolymerasechainreaction. NuclAcids
and the ctenophore. This will be examined in Res 17: 5865
further studies. 12 Harbison GR (1985) On the classification and
evolution of the Ctenophora. In "The Origin and
In conclusion, together with a study of Wain-
Relationships of Lower Invertebrates" Ed by S
right et al. [27], the present study supports the Conway Morris, JD George, R Gibson, HM Piatt,
monophyly of the metazoa. Clarendon Press, Oxford, pp78-100
13 Hills DM, Moritz C (1990) Molecular Systematics.
Sinauer Assoc Inc Publ, Sunderland, pp1-588
ACKNOWLEDGMENTS
14 Hori H, Osawa S (1987) Origin and evolution of
RNA
We are very grateful to Dr. Takashi Miyata and Mr. organisms as deduced from 5S ribosomal
Naruo Nikoh, of the Department of Biophysics, Kyoto sequences. Mol Biol Evol 4: 445-472
Universityfortheirhelpinconstructingthephylogenetic 15 Hyman LH (1940) The Invertebrates: Protozoa
trees, and to Dr. Yoko Watanabe of Ochanomizu Uni- through Ctenophora, Vol 1. McGraw-Hill, New
versity and Dr. Yoshiko Kakinuma of Kagoshima Uni- York, pp248-696
versity for suggestions and encouragements. We also 16 Jukes TH, Cantor RC (1969) Evolution ofprotein
Molecular Phylogeny of Diploblasts 833
molecules. In"MammalianProteinMetabolism"Ed directed enzymatic amplification of DNA with a
byHN Munro, Academic Press, New York, pp21- thermostable DNA polymerase. Science 239: 487-
132 491
17 Katayama T, Yamamoto M, Wada H, Satoh N 23 Saitou N, Nei M (1987) The neighbor-joining
(1993) Phylogenetic position of acoel turbellarians method: a new method for reconstructing phy-
inferredfrompartial 18SrDNAsequences. ZoolSci logenetic trees. Mol Biol Evol 4: 406-425
10: 529-536 24 Sanger F, Nicklen S, Coulson AR (1977) DNA
18 Lake JA (1990) Origin of the metazoa. Proc Natl sequencing with chain-terminating inhibitors. Proc
Acad Sci USA 87: 763-766 Natl Acad Sci USA74: 5463-5467
19 NeefsJ-M,VandePeerY,DeRijkP,GorisA, De 25 Steinbock O (1963) Origin and affinities of the
Wachter R (1991) Complicationofsmall ribosomal lowerMetazoa. In"TheLowerMetazoa"EdbyEC
subunit RNA sequences. Nucl Acids Res 19: 1987- Dougherty, University of California Press, Ber-
2015 keley, pp45-54.
20 Pace NR, Stahl DA. Lane DJ, Olsen GJ (1985) 26 WadaH,MakabeKW,NakauchiM,SatohN(1992)
Analyzing natural microbial populations by rRNA Phylogeneticrelationshipsbetweensolitary andcol-
sequences. Am Soc Microbiol News 51: 4-12 onialascidians, asinferredfromthesequenceofthe
21 Rieger RM (1985) The phylogenetic status of the central region of their respective 18S rDNAs. Biol
acoelomate organization within the Bilateria. In Bull 183: 448-455
"The Origins and Relationships of Lower Inverte- 27 Wainright PO, Hinkle G, Sogin ML, Stickel SK
brates" Ed by S Conway Morris, JD George, R (1993) Monophyletic origins of the Metazoa: An
HM
Gibson, Piatt, Clarendon Press, Oxford, pp evolutionary link with fungi. Science 260: 340-342
101-122 28 WillmerP (1990) Invertebrate Relationships. Cam-
22 SaikiRK,GelfandDH,StoffelS,ScharfSJ,Higuchi bridge Univ Press, Cambridge, pp 163-197
R, HornGT, MullisKB, Erlich HA (1988) Primer-