Testing cold fusion of phyla: maternity in a tunicate x sea urchin hybrid determined from DNA comparisons.
Received September 11, 1995. Accepted September 19, 1995.
Many marine invertebrate groups with different adult forms have similar larvae (such as the trochophore larva of mollusks and annelids, or the pluteus larva of sea urchins and brittle stars), and these similar larval forms are usually considered to be examples of either convergent evolution by natural selection or functional constraints on the evolution of early stages in development (Jagersten 1972; Strathmann 1988; Raff and Kaufman 1991; Wray 1992, 1995). However, Williamson (1992) recently proposed that many of these similarities are due to horizontal genetic transfer of larval forms between members of different taxonomic orders, classes, or phyla by rare cross-fertilizations. This improbable hypothesis implies that the evolutionary history of major taxa is an anastomostic web of hybridizations and divergences, resulting in the insertion of larval forms (or new complex life cycles) into lineages that previously lacked them. The hypothesis raises fundamental questions about current understanding of genetics, systematics, development. and evolution.
Williamson's proposed mechanism of horizontal genetic transfer of larval forms is supported by only one set of experiments. He described laboratory crosses in which eggs of a tunicate (Ascidia mentula, phylum Urochordata) were inseminated with sperm of a sea urchin (Echinus esculentus, phylum Echinodermata) in seawater (Williamson 1992). Swimming blastula embryos were later collected and cultured. In several experiments. some of these embryos developed into functional pluteus larvae like those of sea urchins (Okazaki 1975). In one experiment, 20 larvae metamorphosed into juvenile sea urchins, three of which survived as adults for four years.
Williamson's heretical view of invertebrate phylogenetic relationships was widely dismissed, though some have advocated testing his experimental results (Strathmann 1993). Testing the hypothesis of interphyletic hybridization is useful because (1) Williamson's book was widely read by nonspecialists, some of whom found his ideas interesting and credible; and (2) testing improbable or heretical hypotheses ensures that such hypotheses succeed or fail on their merits and not on their relation to orthodoxy.
Here I test Williamson's experimental result. If tunicate eggs can be fertilized by sea urchin sperm to produce viable sea urchin-like hybrids, then the hybrids should have tunicate mitochondrial DNA (mtDNA) because mitochondria are normally inherited from the female parent only. Similarly, the hybrids should have both tunicate and sea urchin nuclear DNA. I used sequence and restriction-fragment data to characterize both the mitochondrial and nuclear genomes of one of Williamson's putative tunicate-sea urchin hybrids, and compared these data to A. mentula and E. esculentus. All of these comparisons indicate that the hybrid contained sea urchin DNA and did not contain tunicate DNA.
DNA EXTRACTION. AMPLIFICATION, AND CHARACTERIZATION
I extracted total DNA from frozen (adult putative hybrid) or alcohol-preserved (A. mentula and E. esculentus) tissue samples. For the sea urchin and putative hybrid, I used three or four tube feet; for the tunicate, I used pieces of dissected body wall. The sea urchin and tunicate samples did not come from the parents of the putative hybrid individual. Small (~0.2 g) pieces of tissue were macerated in a protease buffer (50 mM Tris-HCI, ph 7.5; 100 mM EDTA; 0.5% SDS; 0.05% w/v proteinase K) on ice, heated to 65 [degrees] C for 1-4 h, then extracted in phenol and phenol-chloroform-isoamyl alcohol (25:24:1). The aqueous phase was then adjusted to 0.7 M NaCI and digested in 1% w/v CTAB (cetyl trimethyl ammonium bromide) for 20 min at 65 [degrees] C. The DNA was then extracted again in chloroform-isoamyl alcohol (24:1), precipitated in cold isopropanol, and pelleted in a microcentrifuge. The pellet was washed in cold 70% ethanol, spun, dried over vacuum, dissolved in water, and stored at 20 [degrees] C.
I used the polymerase chain reaction (PCR) (Kocher et al. 1989; Palumbi et al. 1991) to amplify a 205 bp portion of the mitochondrial cytochrome oxidase subunit I (COI) gene. The choice of mitochondrial gene was unimportant: any relatively conserved mitochondrial sequence for which appropriate primers were available could be used. The PCR primers had the following 5'-3' sequences: CAGGAAACAGCTATG ACCTGCAGGAGGAGGAGAYCC (forward); GCTCGTG TRTCTACRTCCAT (reverse). The COI sequences were amplified in 50 [mu] l volumes containing whole genomic DNA (titrated for each sample); 0.5 [mu] M forward primer; 2.5 [mu] M reverse primer (more than the forward primer because the reverse primer is degenerate at more sites); 0.25 units UltraTherm DNA polymerase (BioCan); the manufacturer's recommended buffer and dNTP concentrations. The thermal cycle sequence was 94 [degrees] C (30 s), 50 [degrees] C (30 s), 72 [degrees] C (60 s) for 25 cycles, followed by a single cycle of 94 [degrees] C (30 s), 50 [degrees] C (30 s), 72 [degrees] C (10 min). These conditions produced single PCR products about 250 bp in length, but amplification of the A. mentula product was weak, so all three products were purified on a 1% agarose gel in TAE buffer (40 mM Tris-base; 5.7% glacial acetic acid; 2 mM EDTA), cut from the gel, spun through glass wool to a liquid phase, precipitated in isopropanol and ethanol, and dissolved in water. These purified PCR products were used as templates in a second PCR reaction, using the same conditions described above. Reamplification produced an abundant PCR product from all three samples.
I used similar methods to amplify a portion (~800 bp) of the nuclear 28S ribosomal RNA gene. The PCR primer sequences were: CTAACCAGGATTCCCCCAGTAACGG (forward); GACTCCTTGGTCCGTGTTTCAAGAC (reverse). I used 0.5 [mu] M primers; other amplification conditions were as described above. The thermal cycle sequence was 94 [degrees] C (30 s), 60 [degrees] C (30 s), 72 [degrees] C (60 s) for 30 cycles. Weak amplification products (or products associated with secondary, nontarget bands) were purified on an agarose gel and reamplified.
I sequenced the mitochondrial COI PCR products directly by the chain termination method (Sanger et al. 1977) using incorporated [alpha]-[sup.35]S dATP and the Sequenase 2.0 PCR Product Sequencing kit (Amersham). Mixtures of 5 [mu] l of enzymatically treated double-stranded PCR product, 10 pmol primer, and 1 [Micro] l DMSO were denatured by boiling and then snap-cooled in dry ice-ethanol for 5 min. The polymerase was diluted in a 50% glycerol buffer. Extension reactions (using the manufacturer's recommended enzyme and reagent concentrations plus 0.5 [Micro] l DMSO) were run for 1 min at 20 [degrees] C, followed by termination reactions at 37 [degrees] C for 5 min. Sequence-reaction products were run on 4.5% acrylamide gels in a glycerol-tolerant gel buffer (71 mM Tris-base; 23 mM taurine; 0.4 mM EDTA) with 8.3 M urea. The A. mentula (U21670) and E. esculentus (U21669) sequences have been deposited in Genbank.
I compared the 28S PCR products to each other using four restriction enzymes: Alu I, Ava I, Rsa I, and Hinf I (BRL). I combined 2 [Micro] l of each PCR product with 1 [Micro] l of 10 x restriction enzyme buffer (BRL REact 1 or 2), 2 units of enzyme, and water in a total volume of 10 [Micro] l. The PCR products were digested for 1 h at 37 [degrees] C, and the restriction fragments were run out on a 1.5% agarose gel in TAE stained with SYBR Green.
COI SEQUENCE COMPARISON
The aligned nucleotide sequences and the inferred amino acid sequences are shown in Figure 1. The 5' ends of the coding strand sequences in each PCR product corresponded to nucleotide 684 in the complete 1554 nucleotide COI sequence for another sea urchin (Strongylocentrotus purpuratus) (Jacobs et al. 1988), or nucleotide 639 in the partial 1263 nucleotide COI sequence for an other tunic ate (Halocynthia roretzi) (Yokobori et al. 1993). The nucleotide sequences of Echinus esculentus and the putative hybrid were identical, except for a single substitution (A-G) at position 103 (a third position in a serine codon), which did not affect the aminoacid sequence. In contrast, there were 63 nucleotide differences between A. mentula and the putative hybrid, of which 27 were differences at first or second codon positions. The two sea urchin species differed at 41 nucleotide positions, of which 35 were substitutions at third codon positions that did not affect the amino acid sequence.
[Figure 1 ILLUSTRATION OMITTED]
Amino-acid sequences were inferred from published mitochondrial codon tables for sea urchins (Jacobs et al. 1988) and tunicates (Yokobori et al. 1993). There were 25 aminoacid substitutions in comparisons among the inferred COI amino-acid sequences from A. mentula, E. esculentus, the putative hybrid, and the published S. purpuratus or H. roretzi sequences. Only two of these substitutions were differences between S. purpuratus and the E. esculentus or putative hybrid sequences. Six of these substitutions were shared differences between the tunicates and sea urchins. The remaining 17 substitutions distinguished either the H. roretzi or A. mentula sequences from the others.
28S RESTRICTION FRAGMENT COMPARISON
Three of the four restriction enzymes (Alu I, Ava I, Rsa I) cut the Echinus esculentus and hybrid 28S PCR products into identical-sized fragments that were different from the fragments produced by cutting the Ascidia mentula PCR product (Fig. 2). Hinf I also cut these three PCR products into fragments, but the Hinf I fragments were the same for all three PCR products. This comparison indicates that the Echinus and hybrid 28S genes shared some Alu I (AG/CT) and Ava I (C/YCGRG) restriction site(s) not found in the Ascidia 28S gene, and lacked some Rsa I (GT/AC) restriction site(s) found in Ascidia. The fragment patterns were difficult to interpret in some cases where the sum of sizes of all fragments was larger than the uncut PCR product (e. g., Alu I fragments for E. esculentus and the hybrid; Rsa I fragments for A. mentula). These patterns may be due to incomplete digestion, comigration of multiple bands of similar size, or the presence of multiple 28S alleles (Hillis and Moritz, 1990). However, there is no indication that the hybrid PCR product included both A. mentula and E. esculentus 28S genes. The comparison of PCR products for a nuclear gene indicates that the hybrid contained sea urchin nuclear DNA (as expected), and did not contain tunicate nuclear DNA in amounts detectable by this method.
[Figure 2 ILLUSTRATION OMITTED]
The near identity of mtDNA sequences from the putative hybrid and E. esculentus, and the strong similarity between these sequences and S. purpuratus, demonstrates that the putative hybrid had sea urchin mtDNA and probably lacked tunicate mtDNA. If the putative hybrid had also contained tunicate mtDNA, then direct sequencing of reamplified PCR products should have produced numerous ambiguities in the sequence ladder caused by the superposition of two sequences in the hybrid sample. Similarly, if the putative hybrid had contained both tunicate and sea urchin nuclear DNA, restriction-enzyme digestion of the putative hybrid nuclear PCR product should produce some fragments of the same size as the sea urchin and others of the same size as the tunicate fragments. Instead, restriction fragments of the putative hybrid nuclear PCR product were always the same size as those of the sea urchin. These results indicate that the putative hybrid developed from a sea urchin egg fertilized by a sea urchin sperm, and not from a tunicate egg as described by Williamson (1992).
What alternative explanations could account for these results if Williamson's identification of the putative hybrid is correct? The putative hybrid could have contained maternal mitochondrial and nuclear genomes inherited from the tunicate that could not be detected by relatively crude PCR methods, or the sea urchin mitochondrial and nuclear genomes (both inherited through the sperm) could have replaced the tunicate genomes completely (requiring duplication of the nuclear genome).
My interpretation of the mtDNA comparison is based in part on the predominantly maternal inheritance of mitochondria in animals (see Avise 1994). There is good evidence in flies, mice, and mussels for regular transmission and inheritance of the sperm mitochondrion (Kondo et al. 1990; Gyllensten et al. 1991; Skibinski et al. 1994), and there may be competition and replacement of one mitochondrial genome by another in heteroplasmic flies (Stordeur et al. 1989). It is remotely possible that the putative hybrid contained some tunicate mtDNA that I did not detect. If the COI PCR primers preferentially amplify the sea urchin gene over the tunicate gene when both are present, then this method would fail to identify a small tunicate component. If the tunicate genome was originally present in large numbers and was subsequently replaced by the sea urchin genome, then no method would successfully identify the original maternal mitochondria in the egg. However, there is no evidence in tunicates or sea urchins for either replacement of maternal mitochondria by paternal mitochondria or recombination between the maternal and paternal organelles. In the best known example, male mussels inherit both the maternal and paternal mitochondrial genomes from their parents but sequester only the paternal genome in sperm (Skibinski et al. 1994; Zouros et al. 1994). Published analyses of mtDNA sequences from tunicates and sea urchins have not reported mitochondrial heteroplasmy, and heteroplasmy is generally rare (a major reason for the widespread use of mtDNA in demographic and phylogenetic studies; Avise 1994).
Similar difficulties might confound the interpretation of restriction fragments from the nuclear 28S ribosomal RNA PCR products. Differential amplification of the tunicate and sea urchin genes could prevent this method from identifying both genomes if they are present in the putative hybrid, and no method could identify the tunicate nuclear genome if (unlikely as this seems) it was replaced during development by the sea urchin genome.
All of my results are negative, but a single positive result would bolster Williamson's hypothesis. If tunicate genes are present, but in very small numbers, in the putative hybrid sample (relative to sea urchin genes), this would be sufficient evidence to confirm the identification of the hybrid. Other methods, such as Southern blotting of mtDNA preparations or screening of a genomic DNA library, could be used in a more sensitive search for tunicate genes in the hybrid. However, there is no indication from my crude PCR-based comparisons that the putative A. mentula X E. esculentus hybrid was anything but a sea urchin.
It is not known how laboratory cultures of A. mentula eggs inseminated with E. esculentus sperm produced Echinus-like offspring. One possible explanation suggested by Strathmann (1993) is the occurrence of rare hermaphrodites in the E. esculentus population from which the sea urchins were collected (Moore 1930). A hermaphrodite used in the crossfertilization experiment to provide sperm may have provided eggs as well. These hermaphrodites are known to be self-fertile (Moore 1930). A similar, alternative explanation is that the hybrids are outcrossed offspring from two different sea urchins, produced as the result of some major laboratory artifact. Williamson (pers. comm. 1994) took care to prevent the contamination of his experiments by sea urchin eggs (by collecting sea urchin sperm and tunicate eggs on different days, and by counting the eggs before and after insemination), but the apparent genetic identity of the hybrid suggests that contamination may have occurred.
Williamson's hypothesis and experimental evidence were widely dismissed, but similarly controversial hypotheses (such as the endosymbiotic theory of eukaryotic cell origins; Margulis 1970) have in time been supported (see Margulis 1981). Horizontal gene transfer may affect some evolutionary processes (Benveniste 1985; Daniels et al. 1990), and there is evidence for horizontal movement of transposable elements between members of different phyla (Garcia-Fernandez et al. 1995). However, this transfer is probably not responsible for the distributions of larval forms among higher taxa of marine invertebrates.
D. Williamson provided tissue samples and encouraged the testing of his ideas. Though these results contradict his identification of the putative hybrids, I commend his willingness to hazard an unpopular idea and his good humor in response to skepticism.
A. Arndt, K. Beckenbach, and S. Cohen helped with DNA extraction, PCR, and sequencing. A. Arndt and K. Beckenbach provided primers. J. Bishop, T. Crompton, and T. Wakeford helped in collecting, storing, and transporting tissue samples. R. Grosberg and S. Palumbi provided thoughtful criticisms of an earlier manuscript, and S. Palumbi directed me to more examples of paternal inheritance of mitochondria. I was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) postdoctoral fellowship, a NSERC grant to M. Smith, and a NSF grant to R. Grosberg.
AVISE, J. C. 1994. Molecular markers, natural history, and evolution. Chapman and Hall, New York.
BENVENISTE, R. E. 1985. The contributions of retroviruses to the study of mammalian evolution. Pp. 359-417 in R. I. MacIntyre, ed. Molecular evolutionary genetics. Plenum, New York.
DANIELS, S. B., K. R. PETERSON, L. D. STRAUSBAUGH, M. G. KIDWELL, and A. CHOVNICK. 1990. Evidence for horizontal transmission of the P transposable element between Drosophila species. Genetics 124:339-355.
GARCIA-FERNANDEZ, J., J. R. BAYASCAS-RAMIREZ, G. MARFANY, A. M. MUNOZ, A. CASALI, J. BAGUNA, and E. SALO. 1995. High copy number of highly similar mariner-like transposons in planarian (Platyhelminthe): Evidence for a trans-phyla horizontal transfer. Mol. Biol. Evol. 12:421-431.
GYLLENSTEN, U., D. WHARTON, A. JOSEFSSON, and A. C. WILSON. 1991. Paternal inheritance of mitochondrial DNA in mice. Nature 352:255-257
HILLIS, D. M., and C. MORITZ, EDS. 1990. Molecular systematics. Sinauer, Sunderland MA.
JACOBS, H. T., D. J. ELLIOT, V. B. MATH, and A. FARQUHARSON. 1988. Nucleotide sequence and gene organization of sea urchin mitochondrial DNA. J. Mol. Biol. 202:185-217.
JAGERSTEN, G. 1972. Evolution of the metazoan life cycle. Academic Press, London.
KOCHER, T. D., W. K. THOMAS, A. MEYER, S. V. EDWARDS, S. PAABO, F. X. VILLABLANCA, and A. C. WILSON. 1989. Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers. Proc. Nat. Acad. Sci. USA 86:6196-6200.
KONDO, R., Y. SATTA, E. T. MATSUURA, H. ISHIWA, N. TAKAHATA, and S. I. CHIGUSA. 1990. Incomplete maternal transmission of mitochondrial DNA in Drosophila. Genetics 126:657-663.
MARGULIS, L. 1970. Origin of eukaryotic cells. Yale Univ. Press, New Haven, CT.
-- . 1981. Symbiosis in cell evolution. Freeman, San Francisco, CA.
MOORE, H. B. 1930. A hermaphrodite sea-urchin. Nature 130:59.
OKAZAKI, K. 1975. Normal development to metamorphosis. Pp. 177-232 in G. Czihak, ed. The sea urchin emhryo. Springer, Berlin.
PALUMBI, S., A. MARTIN, S. ROMANO, W. O. MCMILLAN, L. STICE, and G. GRABOWSKI. 1991. The simple fool s guide to PCR. Vers. 2.0. Univ. of Hawaii, Honolulu.
RAFF, R. A., and T. C. KAUFMANN. 1983. Embryos, genes, and evolution. Indiana Univ. Press, Bloomington.
SANGER, F., S. NICKLEN, and A. R. COULSON. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Nat. Acad. Sci. USA 74:5463-5467.
SKIBINSKI, D. O. E, C. GALLAGHER, and C. M. BEYNON. 1994. Mitochondrial DNA inheritance. Nature 368:817-818.
STORDEUR, E. DE, M. SOLIGNAC, M. MONNEROT, and J.-C. MOUNOLOU. 1989. The generation of transplasmic Drosophila simulans by cytoplasmic injection: Effects of segregation and selection on the perpetuation of mitochondrial DNA heteroplasmy. Mol. Gen. Genet. 220:127-132.
STRATHMANN, R. R. 1988. Functional requirements and the evolution of developmental patterns. Pp. 53-68 in R. D. Burke, R V Mladenov, P. Lambert, and R. L. Parsley, eds. Echinoderm phylogeny and evolutionary biology. A. A. Balkema, Rotterdam.
-- . 1993. Larvae and evolution: toward a new zoology [book review]. Quar. Rev. Biol. 68:280-282.
WILLIAMSON, D. I. 1992. Larvae and evolution: Toward a new zoology. Chapman and Hall, London.
WRAY, G. A. 1992. The evolution of larval morphology during the post-Paleozoic radiation of echinoids. Paleobiology 18:258-287. .
--. 1995. Punctuated evolution of embryos. Science 267: 1115-1116.
YOKOBORI, S.-I., T. UEDA, and K. WATANABE. 1993. Codons AGA and AGG are read as glycine in ascidian mitochondria. J. Mol. Evol. 36:1-8.
ZOUROS, E., A. O. BALL, C. SAAVEDRA, and K. R. FREEMAN. 1994. Mitochondrial DNA inheritance. Nature 368:818.
Michael W. Hart, Institute of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, British Columbia V5A IS6, Canada; Present address: Section of Evolution and Ecology, Division of Biological Sciences, University of California, Davis, California 95616-8755.
|Printer friendly Cite/link Email Feedback|
|Author:||Hart, Michael W.|
|Date:||Aug 1, 1996|
|Previous Article:||The effect of sex on the variance in fitness and mean fitness.|
|Next Article:||Female mate discrimination or male responses to female stimulation?|