The Evolution of Nonrepetitive DNA in Sea Urchins*

Differentiation

Differentiation 10, 7-1 1 (1978)

0 by Springcr-Verlag 1978

The Evolution of Nonrepetidve DNA in Sea Urchins" MICHAEL M. HARPOLD' and SYDNEY P. CRAIG' Department of Biology, University of South Carolina, Columbia, S.C. 29208, USA and Department of Biology, Tulane University, New Orleans, La. 701 18, USA

Molecular hybridization of nuclear DNAs has been employed to study the evolution of nonrepetitive DNA sequences in four species of sea urchin. The data indicate that the extent of homology between the nonrepetitive DNA sequences of S. purpuratus, S.droebachiensis, S .franciscanus, and L. pictus confvms the phylogenetic relationship established through palaeontological evidence. The average rate of divergence of nonrepetitive DNA sequences was found to be approximately 0.22% per million years. In addition, a small fraction (approximately 14%) of the nonrepetitive DNA sequences is highly conserved between S . purpuratus and L. pictus after 120-200 million years divergence. This study may provide a basis for the use of interspecific hybrid embryos of these organisms to investigate the evolution and importance of certain DNA sequences in early developmental processes leading to cell differentiation. Introduction The use of DNA hybridization procedures to compare quantitatively all or any fraction, e.g., nonrepetitive and repetitive sequences [ll,of the genetic material of separate species, provides a powerful and comprehensive tool to study the evolution of the genome. Previous studies on the evolution of nonrepetitive DNA sequences in eukaryotic organisms have shown that these sequences diverge rapidly with high degrees of base substitution among closely related organisms [2, 3, 41. The lack of detailed genetic analysis in most higher organisms is a primary factor which makes interpretation of such results difficult. Echinoderms, especially sea urchins, have been used by developmental biologists for many years for interspecific hybrid experiments to investigate the Submitted by M. M. Harpold to the Department of Biology, Tulane University in partial fulfillment of the requirements for the doctor of philosophy degree Present address: The Rockefeller university, 1230 York Avenue, New York, N.Y. 10021 USA Offprint requests should be addressed to: S. P. Craig, Department of Biology, University of Carolina, Columbia, S.C. 29208, USA

role of embryonic genome in morphogenesis (see [5, 61 for reviews). The existence of information regarding the evolution of the genomes of any of these echinoderms might provide a basis for the use of interspecific hybrids to study the evolution and possible importance of some DNA sequences, at least in early developmental processes. As an initial approach to this problem, we have investigated the evolution of the repetitive and nonrepetitive DNA sequences of three morphologically distinct species of sea urchins in the family Strongylocentrotidae: Strongylocentrotuspurpuratus, Strongylocentrotus droebachiensis, and Strongylocentrotus franciscanus. Although rare, these three species have been observed to form interspecies hybrids which attain adulthood in nature [71. Laboratory-derived interspecies hybrids have also been studied [Sl. The divergence, relative to the DNA sequences of S.purpuratus, of the DNA of a fourth species, Lytechinuspictus, a member of the order Temnopleuroida, has also been investigated. The evolutionary divergence time between L. pfctus and S. purpuratus may be an order of magnitude greater than among the Strongylocentrotidae. Our findings relating to the evolution of the repetitive DNA sequences among these organisms appear elsewhere [8, 91. 0301-4 6 8 1/7 8 / 0 0 1O/OoO7/ $ 1 .OO

8

M. M. Harpold and S.P. Craig: The Evolution of Nonrepetitive DNA in Sea Urchins

The results presented here indicate that the nonrepetitive DNA sequences among these species diverge rapidly during evolution. In addition, the extent of homology between the nonrepetitive DNA sequences in these organisms confiims the phylogenetic relationships among these sea urchins established by palaeontological evidence [lo].

ments rendered entirely single-strandedat each temperature interval were eluted with three 2-ml rinses of 0.12 M phosphate buffer equilibrated at the same temperature. To increase accuracy as well as reproducibility, it was necessary to measure the temperature of the elution buffer within the column. The amount of DNA in each 0.12 M phosphate buffer rinse was determined by the methods described above.

Results Methods DNA Zsolation and Reassociation Procedures. Unlabelled DNAs were isolated from sperm of each species as described in Table 2 of Britten et al. I 1 1 I. Radioactive DNA was isolated according to previously published procedures [ 111 from blastula stage S. purpuratus embryos grown for 10 h in Millipore-fdtered sea water containing 100 units/ml Penicillin-G in the presence of 25 pCi/ml 'H-methyl thymidine (Schwan-Mann, sp. act. 56 Ci/rnmol). The specific activity of the radioactive DNA was 4 x 10' CPM/pg. The DNAs were sonicated to average lengths of 400 to 420 nuclmtides as determined in alkaline isokmetic sucrose gradients 1121. All DNA preparations and buffers were passed over BioRad Chelex 100 before use. The labelled nonrepetitive S. purpuratus DNA was prepared by denaturation of the sheared radioactive DNA for 5 min at loo0 C, and incubation to Cot 300 in 0.41 M phosphate buffer (equimolar mixture of Na,HPO, and NaH2P0,, pH 6.8) at 67O C. This DNA was adjusted to 0.12 M phosphate buffer and fractionated into double stranded and single-strandedmolecules on hydroxyapatite at 600 C as described elsewhere [ 111. The single-stranded DNA molecules were again reassociated to Cat 300 and fractionated as described above. The nonreassociated fraction is termed nonrepetitive 'H-S. purpuratus DNA. Tracer amounts of nonrepetitive 3H-S.purpuratus DNA (800-2000 CPM) were hybridized with a vast excess (ratios ranging from SO00 to l0,OOO) of unlabelled, unfractionated DNA from each of the four species in a buffer containing 0.01 mM EDTA and 0.05%SDS in 0.12 M, 0.41 M, or 1.0 M phosphate buffer at 53", 600, or 700, respectively. In each case the hybridization temperature was 3 1' C below the Tm of long native S. purpuratus DNA as calculated by the expression previously established by Schildkraut and Lfson 1131. It was previously shown that the reassociation of S . purpuratus DNA at 33 to 35O C below the Tm of native DNA does not increase the amount of DNA sequences which reassociate as repetitive sequences 141. At appropriate Cat values, the reactions were terminated by freezing and were stored at -200 C until further analysis. The samples were diluted to 2.5 ml and to a final phosphate buffer concentration of 0.12 M, and were fractionated on hydroxyapatite at 53" C as described elsewhere [ill. The reassociation of the unlabelled driver DNAs was followed from the hydroxyapatite columns. Reassociation of the labelled DNAs were monitored by counting each fraction in Aquasol (NEN)with a scintillation spectrometer.

Hydroxyapatite Thermal Chromatography. Selected reactions were diluted to 2.5 ml and a final phosphate buffer concentration of 0.12 M. Two ml of the sample were applied to a I-mlcolumn of hydroxyapatite equilibrated in 0.12 M phosphate buffer at 53" C. Nonreassociated DNA fragments were eluted from the column by washing the hydroxyapatite with three 2-ml rinses of 0.12 M phosphate buffer. The temperature was then increased in 5 O C increments and the column allowed to equilibrate approximately 5 min. The DNA frag-

Extent of Reassociation of Homologous and Heterologous Nonrepetitive DNA Sequences The reassociation kinetics of nonrepetitive 3H-S.purpuratus DNA with an excess of unlabelled, unfractionated S. purpuratus DNA fragments shows the extent of final reassociation of the labelled DNA was 76.9% (Fig. 1A). Previous studies with S . purpuratus demonstrated that the repetitive DNA sequences in sheared (300-450 nucleotide) DNA reassociate by Cot 200 [141. Since the nonrepetitive preparation used here was derived from unreassociated sequences following two successive reassociations to Cot 300, repetitive sequences were estimated to comprise less than 5% of this nonrepetitive DNA sequence preparation. To determine the extent of relatedness between the nonrepetitive DNA sequences of S. purpuratus and the DNA of S. droebachiensis, S. franciscanus, or L. pictus, the nonrepetitive 'H-S. purpuratus DNA was hybridized with an excess of the unfractionated heterologous DNAs as shown in Figure 1B-D, respectively. The hybridization of the labelled S.purpuratus nonrepetitive DNA sequences with excess S. droebachiensis DNA was 70.2% of the homologous reaction (0.54 x 100/0.769), with excess S. franciscanus DNA was 53.1% (0.408 x 100/0.769), and with L. pictus DNA the reaction was 14.3% of the homologous reaction (0.11 x 100/0.769).

Determination of Base Sequence Mismatch in the Nonrepetitive DNA Hybrids An approximation of the base pair mismatch or nucleotide substitution in hybrid molecules can be obtained by measuring the difference in the melting temperature (Tm) of the heterologous nonrepetitive 3H-S.purpuratus DNA hybrids relative to the Tm of the homologous DNA duplexes [la, 171. For homopolymers and linear DNA fragments there is 1.0% base pair mismatch for every lo C reduction to the Tm. This relationship was derived from Tm reductions in hybrid prokaryotic DNA

,

M.M. Harpold and S.P. Craig: The Evolution of Nonrepetitive DNA in Sea Urchins

9

EQUIVALENT Cot Fig. 1. The reassociation of nonrepCtitive (open triangles, dotted lines) 'H-S. purpuratus DNA with a large excess of unlabeled S. purpuratus,S. droebachiensis,S.franciscanus, and L.pictus DNAs are shown in (A), (B), (C), and @), respectively. The protile of the reassociationof the unlabeled (driver) DNAs for each of the four species is also shown in each of the respectivefigures (closed triangles. solid lines) and are essentially indistinguishable from each other. This is as expected since the genome sizes of these species are very similar: S. purpuratus 0.89 pg, S. droebachiensis 0.9 pg, S. franciscanus 0.83 pg, and L. pictus 0.97 pg [151

Fig. 2. The hydroxyapatite thermal elution profiles of the heterologous hybrids, formed at Cot 100,OOO between 'H-S. purpuratus nonrepetitive DNA and unlabeled S. droebachiensis, S. franciscanus. and L. pictus DNAs arc shown in (A), (B), and (C) respectively (closed circles, solid lines). The t h m a l elution protile of the homologous duplexes formed at Cot l00,OOO between the labeled nonrepetitive S. purpuratus DNA and unlabeled S. purpuratus DNA is also shown in each figure (open circles, darhed lines)

Temperature ("C) fragments measured by optical hyperchromicity and hydroxyapatite thermal chromatography employing 5 O C increments [171. The melting temperature of the hybrid DNA molecules formed between the repetitive sequences of heterologous DNAs by Cot 100,OOO was determined by thermal chromatography on hydroxyapatite. A Cot 100,OOO is, in practice, the highest attainable with present techniques. In theory, these hybrids

should also contain those sequences whose rate of hybridization may be very slow due to severe base pair mismatch. To determine the extent of base pair mismatch in nonrepetitive hybrid molecules, hydroxyapatite thermal chromatography was performed, as described above, on hybrids formed at Cot 100,OOO between nonrepetitive 'H-S. purpuratus DNA and excess S. droebachiensis

M.M.Harpold and S. P. Craig: The Evolution of Nonrepetitive DNA in Sea Urchins

10

DNA: Tm = 73.5' C, S.franciscanus DNA: Tm = 72" C, and L. pictus DNA: Tm = 70° C as shown in Figure 2A-C, respectively. The thermal elution profile of Cot 100,OOO nonrepetitive 3H-S. purpuratus DNA-unlabelled S. purpuratus DNA duplexes, Tm = 78.5' C, is also included in each figure for comparison. Calculation of the base pair mismatch in the heterologous hybrids shows that there was 5% base pair mismatch in nonrepetitive S . purpuratus DNA-S. droebachiensis DNA hybrids, 6.5% in the hybrid nonrepetitive S . purpuratus DNA-S. franciscanus DNA molecules, and 8.5% mismatch in the nonrepetitive S. purpuratus DNA-L. pictus DNA hybrids. Discussion

The degrees of hybridization of the nonrepetitive DNA sequences are consistent with the evolutionary relationships among these sea urchins previously established through palaeontological evidence [ 101. We have shown that at least 50-70% of the nonrepetitive DNA sequences are conserved among the three sea urchins of the family Strongylocentrotidae. We have also shown that there is a high degree of nonrepetitive DNA sequence divergence between S.purpuratus and L. pictus, although a small fraction of these sequences appears to be highly conserved. The results of nonrepetitive DNA sequence divergence among these sea urchins is shown in Table 1, along with estimates of the rate of change in these sequences. The values for the rate of nonrepetitive DNA sequence divergence average approximately 0.22% per million years. The low degree of cross-reaction for the nonrepetitive DNA sequences of S. purpuratus with L. pictus precludes determining a valid divergence rate in this case. It is also probable (based on the longer divergence time) that two or more successive mutations may have occurred at the same loci, and would not be scored

by the present techniques. The reaction criterion employed in this study is of moderate stringency. It is possible that lowering the criterion still further might allow more cross-reaction, while at the same time lowering the Tm of the resulting hybrids. This would not alter the relationships established in this study, but would effect the calculated rate of nonrepetitive sequence divergence, thus the rates in Table 1 must be considered minimal values. The majority of the DNA sequences producing messenger RNA have been shown to be nonrepetitive sequences, albeit only a small proportion of the nonrepetitive DNA sequences in most higher eukaryotes could serve such a function due to genetic load arguments (see [18, 191 for reviews). High degrees of structural gene sequence conservation are not necessarily expected since "wobble" in the third position of codons can occur without an amino acid change [201. Other roles have also been postulated for nonrepetitive DNA sequences such as: nonfunctional, "spacer", or regulatory sequences [211. However, at present the existence of such sequences has not been determined. The 14% of highly conserved nonrepetitive DNA sequences between S. purpuratus and L. pictus may be an example of strong selective pressure to maintain a small subset of specific nonrepetitive DNA sequences for some, as yet, unknown role. It is of interest to note that the rate of sequence divergence in the nonrepetitive DNA sequences among these organisms appears to be consistently greater than for the repetitive DNA sequences, measured in an analogous manner,by a factor of 2 or 3 [8, 91. Due to uncertainty in interpretation of differences between repetitive and nonrepetitive sequence divergence rates, it is only possible to conclude tentatively that on average the repetitive DNA sequences diverge more slowly than the nonrepetitive DNA sequences among the same organisms. Similar conclusions can also be inferred from the Xenopus 141, rodent [31, and primate [221 data. Whether

Table 1. Evolution of nonrepetitive DNA sequences Organisms compared

Total divergence time x iO'years

Extent of shared nonrepetitive DNA sequences

Percent nonrepetitive sequence divergence in shared sequences

Rate of sequence divergence per 1V years

S. purpuratusS. droebachiensis

2

70.2%

5%

0.25%

S . purpuratusS. franciscanus

2.5-4

53.1%

6.5%

0.16-0.26%

S. purpuratusL. pictus

12-20

14.3%

8.5%

(see text)

M. M. Harpold and S.P. Craig: The Evolution of Nonrepetitive DNA in Sea Urchins

11

this might be due to selective pressure to conserve certain repetitive DNA sequences to a greater extent than nonrepetitive DNA sequences is unknown. After this work was completed a report appeared on the evolution of nonrepetitive DNA sequences among these sea urchins [231 which is basically in agreement with our results and strengthens our conclusions. We believe that this study provides a basis for studying the evolution of DNA sequences, possibly of sequences functional in early development, through the use of interspecific hybrids among these sea urchins. Such an analytical approach would furnish greater insight into the role and evolution of various components of the eukaryotic genome.

8. Harpold, M. M.: The evolution of repetitive and nonrepetitive DNA sequences in sea urchins. Ph.D. dissertation, Tulane University. AM Arbor, Mich.: University Microfilms, Inc. 1976 9. Harpold, M.M., Craig, S. P.: The evolution of repetitive DNA sequences in sea urchins. Nuc. Acids Res. (in press) 10. Durham, 1. W.:Echinoids, classification. In: Treatise on Invertebrate Paleontology. Moore, R. C. (ed.). Part U, Echinodermata 3, Vol. I, p. 280. Geological Society of America and University of Kansas Press 1966 11. Britten, R. J., Graham, D. E., Neufeld, B. R.: Analysis of repeated DNA sequences by reassociation. In: Methods in Enzymology. Grossman, L., Moldave, K. (eds.), Vol. 29, part E, p. 363. New York: Academic Press 1974 12. Noll, H.: Characterization of macromolecules by constant velocity sedimentation. Nature (London) 215, 360 (1967) 13. Schildkraut, C., Lifson, S.: Dependence of the melting temperature of DNA on salt concentration. Biopolymers 3, 195

Acknowledgements: We thank Professor H. B. Fell of Harvard University for helpful discussions on the paleontological evidence concerning the evolution of sea urchins. A preliminary account of this work was presented at the 18th S. E. Developmental Biology Conference. This project was initiated while one of us (M.M.H.) was supported by an NIH Fellowship at the Marine Biological Laboratory, Woods Hole, Mass., during the Summer, 1974. This project was also supported by the Department of Biology at the University of South Carolina.

(1 965) 14. Graham, D. E., Neufeld, B.

References R. J., Kohne, D. E.: Repeated sequences in DNA. Science 161, 529 (1968) Kohne, D. E.: Evolution of higher organism DNA. Quart. Rev. Biophysics 3, 327 (1970) Rice, N. R.: Change in repeated DNA in evolution. Brookhaven Symp. 23, 44 (1972) Galau, G. A., Chamberlin, M. E., Hough, B. R., Britten, R. J., Davidson, E. H.: Evolution of repetitive and nonrepetitiveDNA in two species of Xenopus. In: Molecular Evolution. Ayala, F. J. (ed.), p. 200. Sunderland, Mass.: Smauer Press 1976 Harvey, E. B.: The American Arbacia and Other Sea Urchins. Princeton, N. J.: Princeton Univ. Press 1956 Davidson, E. H.: Gene Activity in Early Development, 2nd ed. New York: Academic Press 1976 Swan, E. F.: The strongylocentrotidae(echinoidea)of the northwest pacific. Evolution 7, 269 (1953)

15. 16.

17.

18. 19.

1. Britten. 2. 3. 4.

5. 6. 7.

20. 21.

22.

R.,Davidson, E. H., Britten, R. J.: Interspersion of repetitive and nonrepetitive DNA sequences in the sea urchin genome. Cell 1, 127 (1974) Hinegardner, R.: Cellular DNA content of Echinodermata. Comp. Biochem. Physiol. 49B, 219 (1974) Laird, C. D., McConaughy, B. L., McCarthy, B. J.: Rate of furation of nucleotide substitutions in evolution. Nature (London) 224, 149 (1969) Bonner, T. I., Brenner, D. J., Neufeld, B. R.,Britten, R. J.: Reduction in the rate of DNA reassociation by sequence divergence. J. Mol. Biol. 81, 123 (1973) Bishop, J. 0.: The gene numbers game. Cell 2, 81 (1974) Lewin. B.: Units of transcription and translation: sequence components of heterogeneous nuclear RNA and messenger RNA. Cell 4, 77 (1975) Crick, F. H. C.: Condon-anticodonpairing: the wobble hypothesis. J. Mol. Biol. 19, 548 (1966) Davidson, E. H., Klein, W. H., Brittcn, R. J.: Sequence organization in animal DNA and a speculation on hnRNA as a coordinate regulatory transcript. Develop. Biol. 55, 69 (1977) Deininger, P. L., Schmid, C. W.: Thermal stability of human DNA and chimpanzee DNA hetcroduplexes. Science 194, 846

(1976) 23. Angerer, R. C., Davidson, E. H., Britten, R. J.: Single copy DNA and structural gene sequence relationships among four sea urchin species. Chromosoma 56, 213 (1976)

Received June 1977/Revised September 1977