- Email: [email protected]
The mitochondrial and ribosomal DNA components of oocytes ofUrechis caupo
DEVELOPMENTALBIOLOGY 22,
The Mitochondrial
1-14 (1970)
and Ribosomal
of Oocytes IGOR Deportment
B. DAWID
DNA Components
of Urechis AND
DONALD
caupo D. BROWN
of Embryology,
Carnegie Institution Baltimore, Maryland 21210 Accepted
of Washington,
October 6, 1969
INTRODUCTION
Among the substances stored by oocytes for use during embryogenesis are large numbers of mitochondria and ribosomes. The abundance of these structures is correlated with an increased content of two special types of DNA. These DNAs and their occurrence in oocytes have been studied most extensively in amphibians. The presence of DNA in mitochondria is a general phenomenon, and the preeminence of mitochondrial DNA (M-DNA)’ in frog (Dawid, 1965, 1966) and in sea urchin eggs (Pike et al., 1967) is a consequence of the large number of mitochondria in each egg. The nuclei of oocytes contain extra copies of genes for 28 S and 18 S ribosomal RNA which function in the synthesis of rRNA during oogenesis. Amplification of ribosomal DNA (rDNA) has been demonstrated in oocytes of amphibians (Brown and Dawid, 1968; Gall, 1968) and fish (Vincent et al., 1968) which have multiple nucleoli, and in some insect oocytes which contain a large DNA body surrounding the nucleolus (Lima-de-Faria et al., 1969; Gall et al., 1969). No cytological indication for excess rDNA has been found in oocytes of marine invertebrates which contain only a single large nucleolus in each nucleus. Evidence for rDNA amplification in oocytes of two such organisms, Spisula solidissima and Urechis caupo has been mentioned earlier (Brown and Dawid, 1968), but a recent study did not detect any rDNA amplification in starfish oocytes (Vincent et al., 1969). This paper presents evidence on the abundance and some properties of M-DNA, and on the amplification of rDNA in mature oocytes of Urechis caupo, an echiuroid worm. ’ Abbreviations: SLS, sodium lauryl sulfate; rRNA, ribosomal RNA, defined as 28 S and 18 S RNA; rDNA, ribosomal DNA, defined as the DNA component containing the sequences homologous to rRNA; M-DNA, mitochondrial DNA: GC content, content in deoxyguanylic acid and deoxycytidylic acid; SSC, 0.15 M sodium chloride, 0.015 M sodium citrate; MAK, methylated albumin adsorbed on kieselguhr.
2
DAWID
AND
BROWN
METHODS
Collection of Urechis eggs and sperm. Urechis caupo were collected by Mr. H. Little of Stanford University from the mud flats near San Francisco and shipped by air to Baltimore, where they were maintained at 11-14’C without feeding in an Instant Ocean Culture System (Aquarium Systems, Inc., Wickliffe, Ohio). Urechis oocytes develop free in the coelomic cavity without any follicle cell envelope and are collected into storage sacs when they are mature (Newby, 1940). Mature oocytes were obtained by probing through the external gynophores or by dissecting them from the storage sacs (Gould, 1967). After settling 3 times in sterile seawater, oocytes were suspended in about 4 volumes of 5% bovine serum albumin in seawater and overlaid on 10 ml of 12% albumin layered over 10 ml of 18% albumin in seawater. After centrifugation at 2000 rpm for 15 minutes in a horizontal rotor of the International HN centrifuge, the oocytes were dispersed through the 12% albumin layer and red blood cells and debris had sedimented. The oocytes were removed with a syringe and washed twice with sterile seawater by sedimentation. Such preparations contained less than one blood cell for each 100 oocytes. Purification of DNA from oocytes. Washed oocytes were homogenized in about 20 volumes of SSC (0.15 M NaCl, 0.015 M sodium citrate), sodium lauryl sulfate (SLS) was added (2%), and the homogenate was gently shaken with one volume of phenol overnight at 4“C. The aqueous phase was recovered after centrifugation, two volumes of ethanol were added, and the precipitate was collected and dissolved in about one-tenth of the original volume of 0.02 M Tris-HCl, pH 7.4. The solution was made 0.5% with respect to SLS and 1 M with NaCl and incubated at 4’C for several hours with gentle shaking. More than 80% of the RNA precipitates under those conditons with little if any loss of DNA. The nucleic acids in solution were precipitated with two volumes of ethanol and the precipitate was dissolved in one-tenth of the original volume of 0.02 M Tris, pH 7.4. Pancreatic RNase (0.1 mg/ml), and RNase Tl (50 units/ml), both heated to 80°C for 10 minutes to destroy DNase, were added, and the solution was incubated for 30 minutes at 37°C. Pronase was preincubated for 30 minutes at 37°C and then added at 1 mg/ml; 0.2% SLS was also added and incubation continued for 1 hour. The solution was extracted with phenol overnight, ethanol was added to the aqueous phase, and the precipitate was dissolved in 0.3 M NaCl, 0.05 M sodium phosphate, pH 7.0. The solution was applied to a large methylated albumin
URECHIS OOCYTE
DNA
3
kieselguhr (MAK) column (about 5 ml bed volume per milliliter of packed eggs), and the column was washed with 0.3 M NaCl until the absorbancy at 260 rnF was below 0.02. The DNA was then eluted with 1 M NaCl, 0.05 M sodium phosphate, pH 7.0. The solution was dialyzed against 2 mM Tris-HCl, pH 7.4, concentrated in a flash evaporator, and centrifuged to equilibrium in a CsCl gradient for 65 hours at 33,000 rpm in a No. 65 or No. 40 fixed-angle rotor in a Beckman Spinco centrifuge. Fractions were collected and used for hybridization with rRNA (see below) or the DNA was recovered for other experiments. The yield of DNA was approximately 5 Fbg per oocyte, which represents a 50”; yield (see below). A major unidentified contaminant of oocyte DNA precipitates as a fibrous mass with ethanol and makes concentrated solutions very viscous. Most of it is removed in the MAK fractionation step, the remainder bands at the top of CsCl gradients, where it is separated completely from the DNA. DNA from cell fractions of Urechis oocytes. Urechis oocytes were homogenized at 0°C in about 10 volumes of 0.25 M sucrose, 0.03 M Tris-HCl, 1 mM EDTA, pH 7.4. The suspension was centrifuged for 20 minutes at 1500 rpm in the International centrifuge; the supernatant was collected and centrifuged for 20 minutes at 12,000 rpm in a Sorvall SS-34 rotor to pellet mitochondria. DNA was extracted from the low speed pellet as described above for whole-oocyte DNA. The high speed pellet was extracted by methods that have been applied to purified preparations of mitochondria (Dawid, 1966). DNA from mitochondria of chicken liver was prepared by the same method. Other nucleic acid preparations. Urechis sperm DNA was prepared by SLS-phenol extraction after an initial pronase digestion (Dawid, 1965; Brown and Weber, 1968). Ribosomal RNA labeled with uridine5-“H (Amersham-Searle) was prepared from cells of Xenopus grown in culture (Brown and Weber, 1968). The specific activity was 100 cpm/mwg. Nonradioactive rRNA was purified from unfertilized eggs of Xenopus and oocytes of Urechis (Brown and Weber, 1968). Escherichia coli rRNA was prepared by the same method. All RNA preparations were fractionated on MAK columns before use, concentrated, dialyzed, and finally filtered through prewashed HA Millipore filters (Brown and Weber, 1968). Denaturation and reannealing. DNA was denatured either by heating to 100°C for 10 minutes in solutions of an ionic strength below 0.05, or by exposure to 0.1 N KOH at room temperature for 5 minutes. Reannealing of denatured DNA was carried out at concentrations be-
4
DAWID
AND
BROWN
tween 2 and 10 &g/ml in 0.5 M CsCl at 60°C for 12-14 hours (Dawid and Wolstenholme, 1968). Hybridization. Hybridization with RNA-JH was performed with DNA that had been fractionated in CsCl gradients and immobilized on HA Millipore filters (Brown and Weber, 1968). Electron microscopy. DNA was spread by the method of Freifelder and Kleinschmidt (1965) on a hypophase of 0.3 M sodium acetate and 0.5% formaldehyde. The preparations were rotary shadowed with platinum-iridium and examined in an Hitachi 11-E electron microscope. Magnification was calibrated with a carbon replica of a diffraction grating (Ladd Research Inc.). Analytical centrifugation in CsCl. Methods for equilibrium centrifugation of DNA have been described (Mandel et al., 1968). Band sedimentation through CsCl solutions (1.33 gm/cm”) was carried out in a type I centerpiece of a Beckman ultracentrifuge (Vinograd et al., 1965). RESULTS
The Mitochondrial
AND
DISCUSSION
Component of Oocyte DNA
The properties of oocyte DNA and DNA from purified mitochondria were compared. In CsCl gradients native M-DNA (Fig. la, top) banded at the same buoyant density as sperm DNA (1.699 gm/cm”) (Fig. lb, top). Denaturation shifts the density of both DNAs equally (Figs. la and lb, middle). Reannealed M-DNA had the density of native DNA (Fig. la, bottom) indicating rapid and essentially complete reassociation. Such behavior of M-DNA has been attributed to the homogeneity of its sequences (Borst et al., 1967; Dawid and Wolstenholme, 1968). In contrast to M-DNA, the bulk of sperm DNA did not reassociate (Fig. lb, bottom). A fraction of approximately 5% of the reannealed sperm DNA formed a band at 1.707 gm/cm”. This partially reassociated DNA probably consists of one or more families of highly repetitive sequences which reassociate partially but fail to crosslink with the rest of the nuclear DNA (Britten and Kohne, 1968). Native whole-oocyte DNA (Fig. lc, top) has the same density (1.699 gm/cm”) as sperm or mitochondrial DNA, and on denaturation its density increases to 1.716 gm/cm’ (Fig. lc, middle). When the oocyte DNA was denatured and reannealed, bands were observed at the densities of native and of denatured DNA (Fig. lc, bottom). As shown above, M-DNA regains the density characteristic of native DNA under the reannealing conditions used, whereas the bulk of chromoso-
URECHIS
OOCYTE
5
DNA
A 1.699
I.731 cl
1.731
1.699
b
L
I
I .731
1.699 C
FIG. 1. tinaturation studies of Urechis DNA by CsCl equilibrium sedimentation in the analytical ultracentrifuge. The marker DNA from Micrococcus lysodeikticus, density 1.731 gm/cm’, has been deleted from the densitometer tracings. (a) Top, native M-DNA; middle, denatured M-DNA; bottom, reannealed M-DNA; (b) top, native sperm DNA; middle, denatured sperm DNA; bottom, reannealed sperm DNA; (c) top, native oocyte DNA; middle, denatured oocyte DNA; bottom, reamrealed oocyte DNA.
ma1 DNA remains single-stranded. Three separate preparations of DNA were prepared from Urechis oocytes and subjected to reannealing. the proportion of oocyte DNA which reassociated in these experiments was 67%, 45%, and 72% (average of 61%); this fraction of DNA is considered to be M-DNA. The identification of this rapidly reassociating fraction of DNA as M-DNA was supported by its ability to form common networks during reannealing with a known M-DNA of heterologous origin. Different DNAs form common networks during reannealing only if they share some nucleotide sequences. By this criterion the M-DNA of Xenopus has been shown to contain some sequence homology with
DAWID
AND
BROWN
C / cl
b /
1.709
FIG. ‘2. Corenaturation studies of Urechis DNAs with chick mitochondrial DNA. Bands in CsCl gradients of (a) separately reannealed and subsequently mixed M-DNA of Urechis and the chick; (b) jointly reannealed M-DNA of Urechis and the chick; (c) separately reannealed Urechis oocyte DNA and chick M-DNA; (d) Urechis oocyte and chick M-DNA reannealed in mixture.
chick M-DNA, but not with yeast M-DNA (Dawid and Wolstenholme, 1968). Such an experiment was carried out with M-DNAs from Urechis and the chick. Individually reannealed samples of the M-DNAs separate in CsCl (Fig. 2a). When the two DNAs were reannealed together most of the material formed a single band of intermediate density (Fig. 2b) indicating the formation of common networks. A shoulder of density 1.700 gm/cm” suggests that a small fraction of the Urechis M-DNA did not crosslink with chick M-DNA. Since M-DNA from Urechis and the chick share enough nucleotide sequences to form common networks, this test was also applied to oocyte DNA of Urechis. Separately reannealed samples, when mixed, formed two sharp bands in CsCl gradients (Fig. 2c), in addition to an underlying broad band which represents the nonreassociating component of the oocyte DNA. When Urechis oocyte DNA was rean-
URECHIS OOCYTE
DNA
7
nealed together with chick M-DNA, only a single sharp band appeared in addition to the broad band of DNA which did not reassociate (Fig. 2d). Therefore, the rapidly reassociating fraction of Urechis oocyte DNA forms common networks with chick M-DNA. There is no detectable network formation between nuclear DNA and M-DNA of Urechis during reannealing under these conditions. This finding is in accord with the behavior of other pairs of nuclear and mitochondrial DNAs studied previously (Dawid and Wolstenholme, 1968). We can summarize the constituents of oocyte DNA with the following approximate balance sheet. An oocyte contains a total of about 10 ppg DNA, while only 4 PpLgare expected for a 4C complement of chromosomal DNA (Schwartz, 1969). The reassociation experiments show that about 40% of oocyte DNA (4 grg per oocyte) behaves like nuclear DNA, and the remaining 60 % are mitochondrial DNA (6 ppg per oocyte). The Size and Structure
of Urechis M-DNA
The M-DNAs from all animals studied so far are circular molecules which display special physical properties characteristic of closed (twisted) circles (Borst et aZ., 1967; Dawid and Wolstenholme, 1967). M-DNA from Urechis shares these properties. Sedimentation analysis of Urechis M-DNA demonstrated the fast and slow components previously described for frog M-DNAs (Dawid and Wolstenholme, 1967). The presence of a fast component in neutral solutions and a very fast component in alkali indicates that Urechis M-DNA contains supercoiled (twisted) circular molecules that are characteristic of animal M-DNAs. Electron micrographs of Urechis M-DNA showed many open circular molecules as well as some twisted circles. The contour length of Urechis M-DNA circles was found to be 5.85 f 0.021 ,,, SE (n = 20), compared to 5.86 f 0.048 P SE (n = 15) for Xenopus M-DNA prepared under identical conditons. These observations add to the picture of remarkable uniformity of structure and size of animal MDNAs. The size constancy is striking; the M-DNAs of Urechis, Xenopus, and several other vertebrates (Wolstenholme and Dawid, 1967; Borst et al., 1968) fall into a very narrow size range. While size is retained, avian M-DNAs differ from those of other vertebrates in base composition (Borst et al., 1968) and therefore must have large differences in nucleotide sequences. Nevertheless, the experiment shown in
8
DAWID
AND
BROWN
Fig. 2 suggests that some sequence homology exists between Urechis and chick M-DNA. The uniform size of M-DNAs from different animals despite this divergence in sequences cannot be considered as evidence for a minimum “iimctional length” of 5.5 to 6 P for M-DNA since circles below 5 P have been found in the mitochondria of two urodele species (Wolstenholme and Dawid, 1968). The Ribosomal
Genes in Urechis Oocytes
The content of rDNA in Urechis oocyte DNA was compared to that of sperm DNA by hybridization with radioactive rRNA of Xenopus. First, the approximate extent of homology between Xenopus rRNA and the ribosomal genes of Urechis was measured and the validity of estimating the content of Urechis rDNA with Xenopus rRNA was established. Xenopus rRNA has been shown to have sequence homologies with DNAs from a wide variety of eukaryotes (Sinclair and Brown, 1968). An estimate of this homology with Urechis DNA was made in competition experiments in which nonradioactive Urethis rRNA was used to compete in the homologous hybridization reaction between Xenopus DNA and Xenopus radioactive rRNA. Two identical samples of Xenopus erythrocyte DNA were centrifuged to equilibrium in CsCl, the gradients were collected in fractions, and the DNA in each fraction was immobilized on individual Millipore filters. The two filter sets were split into half filters to give four identical sets and each set was hybridized with rRNA-“H containing a different nonradioactive rRNA as competitor (Fig. 3). Xenopus rRNA effectively competes in the hybridization reaction (less than 2% of the radioactive hybrid remained), E. coli rRNA had no effect, and Urechis rRNA reduced the hybridization reaction by about 60%. This experiment demonstrates a high degree of cross homology between the ribosomal sequences of Urechis and Xenopus. Xenopus rRNA hybridizes with a fraction of Urechis DNA of high buoyant density. As in the case of rDNAs from most eukaryotes, the density of rDNA of Urechis is due to its high content of deoxyguanylic and deoxycytidylic acid (GC) as compared to the bulk of the DNA. In the case of Urechis the bulk DNA bands at 1.699 gm/cm”, while rRNA hybridizes with DNA having a buoyant density of approximately I.715 gm/cm”. These values correpsond to GC contents of 40% and 56%, respectively. The latter value is slightly lower than the reported average base composition of rRNA from Urechis (Gould, 1969a). Hybridization of Urechis DNA with excess radioactive Xeno-
URECHIS OOCYTE
9
DNA
1600
06 800
400
IO
Tube
no.
5 a 0
IO
FIG. 3. Competition for the hybridization of Xenopus rRNA-JH with Xenopus DNA by unlabeled rRNA from different sources. Two identical filters sets of CsCl-fractionated DNA were split into half, and each half set was hybridized with 0.7 fig rRNA-“H and 100 pg of unlabeled rRNA as competitor as indicated, in 2 ml of 4 x SSC overnight at 70%. (a) e-0, no unlabeled RNA added; 0 - - 0, Xenopus rRNA. (b) o-0, E. coli rRNA; 0 - - 0, Urechis rRNA. Thin line, A260. Background has not been subtracted. pus rRNA
demonstrated that about 0.06% of the Urechis genome is homologous to it. No measurement of the fraction of ribosomal sequences in Urechis DNA has been performed with its homologous rRNA. By using the competition experiment of Fig. 3 one can estimate that about 0.1% of Urechis DNA is homologous to rRNA; this value is only a rough estimate because of possible problems in using heterologous rRNA for such measurements. The relative content of rDNA in different DNA preparations of Xenopus has been compared previously by hybridization with subsaturating amounts of rRNA (Brown and Weber, 1968). Such measurements are valid when the DNA has been fractionated in CM21 and is immobilized on nitrocellulose filters. CsCl fractionation increases the specificity of the method since only hybridization with DNA of high density is scored. Since the immobilized DNA does not enter into the kinetics of the reaction the degree of saturation depends only on the concentration of RNA in solution. At a given RNA concentration the level of hybridization is proportional to the total DNA on a filter set. Such measurements are also possible in the heterologous reaction of Xenopus rRNA with Urechis sperm DNA (Fig. 4) since the amount of rRNA bound is linearly proportional to the DNA
10
DAWID
AND
BROWN
FIG. 4. Amounts of rRNA hybridized with varying amounts of DNA on filters. Varying amounts of Urechis sperm DNA and of Xenopus erythrocyte DNA were fractionated in CsCl. All the DNA-filters were hybridized together in 8 ml of 4 X SSC with 3.5 rg of rRNA-JH of Xenopw. The radioactivity bound in the high-density region of each gradient is plotted against the total DNA on the filter set.
on the filter over a 20-fold range of DNA input. Therefore, the fraction of any DNA sample from Urechis homologous to rRNA can be estimated from such a “calibration” curve. Six different preparations of Urechis oocyte DNA and the sperm DNA from three different males were compared for their relative content of rDNA. Oocyte and sperm DNA were fractionated separately in CsCl gradients and the DNA-containing filters hybridized together with subsaturating amounts of rRNA-3H from Xenopus. One such experiment is shown in Fig. 5, and a compilation of the data is presented in Table 1. Whole-oocyte DNA hybridized with about 3 times more rRNA than did equivalent amounts of sperm DNA. Since the M-DNA present in whole oocyte DNA is not expected to hybridize with rRNA (as shown previously with Xenopus M-DNA, Brown and Dawid, 1968) the degree of amplification of rDNA in nuclear DNA of oocytes is expected to be more than two times higher. This conclusion was verified in hybridization experiments with DNA extracted from
11
URECHIS OOCYTE DNA 04
I
Oocy’e
, 1600
DNA
200 j-
i
1 BOOz v
‘b 02 ?I a
400 !!k
" Tube no.
IO
0
FIG. 5. Hybridization of CsCl-fractionated oocyte and sperm DNAs of Urechis with rRNA-‘H. About 20 pg of whole oocyte DNA and 70 pg of sperm DNA were fractionated in CsCl. The DNA-filters prepared from each fraction were hybridized together with 3.5 rg of rRNA-JH in 8 ml of 4 X SSC. --, A2&0; a---@, cpm. Background has not been subtracted. TABLE AMPLIFICATION
OF rDNA
1 IN Urechis
OOCYTES
Hybridization of rRNA-“H with oocyte DNA relative to sperm DNA
DNA
Measured Whole-oocyte
“Nuclear”
DNA
DNA
2.6 3.0 3.3 5.6 5.9
Corrected’ 6.5 7.5 8.3 -
’ Hybridization with rRNA-“H was performed on DNA fractionated in CsCl as shown in Fig. 5 and described in the text. The amount of DNA on filters was determined as described previously (Brown and Weber, 1968). The numbers are derived by dividing the counts per minute of rRNA bound per microgram of total oocyte DNA by the counts per minute of rRNA bound per microgram of total sperm DNA. This ratio is a valid measure of the relative rDNA content as demonstrated in Fig. 4. b Experiments with whole-oocyte DNA were corrected for the M-DNA present (60%) which does not hybridize with rRNA by multiplying the measured value by 2.5. ’ DNA from low-speed pellet as described in the text.
the low speed pellet of oocytes (see Methods). This pellet contains the nuclear DNA complement of the oocyte since essentially pure MDNA was recovered from the supernatant suspension. DNA extracted from the low speed pellet contained about twice the enrichment of rDNA than did whole oocyte DNA (Table 1). A single Urechis oocyte with its four chromosome sets thus contains
12
DAWID AND BROWN
about 24 haploid equivalents of rDNA. This rDNA is localized presumably in its single large nucleolus. This single nucleolus which is commonly found in the oocytes of many marine invertebrates, some insects, and mammals, is almost certainly the site of rRNA synthesis within these cells. In Urechis, disappearance of the nucleolus at meiosis is associated with cessation of rRNA synthesis (Das et al., 1965; Gould, 1969b; Schwartz, 1969). Urechis oocytes, as well as the oocytes of many organisms containing single nucleoli, accumulate relatively large quantities of rRNA. A mature Urechis oocyte contains about 12 rnccg of RNA, and there is little change in rRNA content during development to the trochophore larva (Schwartz, 1969). The same pattern of synthesis and storage of ribosomes in oocytes and their subsequent use during embryogenesis has been found to a greater or lesser degree in all organisms studied to date (Brown and Dawid, 1969). The extension of the correlation or rDNA amplification with synthesis and storage of rRNA to an organism with a single oocyte nucleus supports the idea (Brown and Dawid, 1968) that extrachromosomal copies of rDNA in oocytes are under different control than the rDNA integrated in the chromosomes of somatic cells. This would permit oocytes to uncouple the synthesis of ribosomes from protein synthesis, in distinction to somatic cells, where these two activities are tightly coupled. SUMMARY
Oocytes of Urechis caupo contain 10 bpg DNA (Schwartz, 1969). The oocytes’ 4 chromosome sets account for 4 ppg DNA, and the remaining 6 PpLgwere shown to be mitochondrial DNA. All the DNA from isolated mitochondria of Urechis oocytes and 60% of the whole oocyte DNA reassociate rapidly after denaturation, share some sequence homologies with chick mitochondrial DNA, and contain circular molecules. The contour length of these circular molecules is 5.85 p. Mitochondrial DNA of Urechis is closely similar in physical structure and size to mitochondrial DNA of Xenopus laevis and other vertebrates. The DNA coding for ribosomal RNA (ribosomal DNA) is amplified in Urechis oocytes. Ribosomal DNA of Urechis was measured by hybridization with Xenopus ribosomal RNA, which is homologous to about 60% of the ribosomal sequences in Urechis. The concentration of ribosomal DNA in the nuclear component of oocyte DNA is about 6 times higher than in sperm DNA. The oocyte therefore contains
URECHIS
OOCYTE
DNA
13
about 24 haploid equivalents of ribosomal DNA, which are localized
presumably in its singlelargenucleolus. ACKNOWLEDGMENTS We thank Dr. M. Schwartz for help with the collection Rehhert and E. Jordan for excellent technical assistance.
of Urechis oocytes, and M.
REFERENCES BORST, P., KROON, A. M., and RUTTENBERG, G. J. C. M. (1967). Mitochondrial DNA and other forms of cytoplasmic DNA. In “Genetic Elements: Properties and Functions” (D. Shugar, ed.), pp. 81-116. Academic Press, New York. BORST, P., VAN BRUGGEN, E. F. J., and RLJTTENBERG,G. J. C. M. (1968). Size and structure of mitochondrial DNA. In “Biochemical Aspects of the Biogenesis of Mitochondria,” (E. C. Slater, J. M. Tager, S. Papa, and E. Quagliariello, eds.), pp. 51-69. Adriatica Editrice, Bari. BR~I-~EN, R. J., and KOHNE, D. E. (1968). Repeated sequences in DNA. Science 161, 529-540. BROWN, D. D., and DAWID, I. B. (1968). Specific gene amplification in oocytes. Science 160, 272-280. BROWN, D. D., and DAWID, I. B. (1969). Developmental genetics. Ann. Reu. Genet. 3, 127-154. BROWN, D. D., and WEBER, C. S. (1968). Gene linkage by RNA-DNA hybridization. I. J. Mol. Biol. 34, 662-680. DAS, N., LUYKX, P., and ALFERT, M. (1965). The nucleolus and RNA metabolism in Urechis eggs. Develop. Biol. 12, 72-78. DAWID, I. B. (1965). Deoxyribonucleic acid in amphibian eggs. J. Mol. Biol. 12, 581599. DAWID, I. B. (1966). Evidence for the mitochondrial origin of frog egg cytoplasmic DNA. Proc. Natl. Acad. Sci. U.S. 56, 269-276. DAWID, I. B., and WOLSTENHOLME,D. R. (1967). Ultracentrifuge and electron microscope studies on the structure of mitochondrial DNA. J. Mol. Biol. 28, 233-245. DAWID, I. B., and WOLSTENHOLME,D. R. (1968). Renaturation and hybridization studies of mitochondrial DNA. Biophys. J. 8, 65-81. FREIFELDER, D., and KLEINSCHMIDT, A. K. (1965). Single-strand breaks in duplex DNA of coliphage T7 as demonstrated by electron microscopy. J. Mol. Biol. 14, 271-278. GALL, J. G. (1968). Differential synthesis of the genes for ribosomal RNA during amphibian oogenesis. Proc. Natl. Acad. Sci. U.S. 60, 553-560. GALL, J. G., MACGREGOR, H. C., and KIDSTON, M. E. (1969). Gene amplification in the oocytes of dytiscid water beetles. Chromosoma 26, 1699187. GOULD, M. C. (1967). Echiuroid worms: Urechis, In “Methods in Developmental Biology” (F. H. Wilt and N. K. Wessells, eds.), pp. 163-171. Crowell, New York. GOULD, M. C. (1969a). RNA and protein synthesis in the unfertilized eggs of Urechis caupo. Deuelop. Biol. 19, 460-481. GOULD, M. C. (1969b). A comparison of RNA and protein synthesis in fertilized and unfertilized eggs of Urechis caupo. Develop. Biol. 19, 482-497. LIMA-DE-FARIA, A., BIRNSTIEL, M., and JAWORSKA,H. (1969). Amplification of ribosomal cistrons in the heterochromatin of Acheta. Genetics 61, Suppl., 145-159.
DAWID ANDBROWN MANDEL, M., SCHILDKRAUT, C. L., and MARMUR, J. (1968). Use of CsCl gradient analysis for determination of guanine plus cytosine content of DNA. In “Methods in Enzymology” (S. P. Colowick and N. 0. Kaplan, eds.), Vol. XII B, pp. 184-195. Academic Press, New York. NEWBY, W. W. (1940). “The Embryology of the Echiuroid Worm Urechis caupo.” American Philosophical Society, Philadelphia. PIKO, L., TYLER, A., and VINOGRAD, J. (1967). Amount, location, priming capacity, circularity and other properties of cytoplasmic DNA in sea urchin eggs. Biol. Bull. 132, 68-90. SCHWARTZ, M. (1969). Nucleic acid metabolism in oocytes and embryos of the echiuroid Urechis cuupo. Ph.D. Thesis, The Johns Hopkins University, Baltimore, Maryland. SINCLAIR, J. H., and BROWN, D. D. (1966). Comparative studies on genes for ribosomal RNA in eukaryotes. Carnegie Inst. Wash. Year Book 67, 404-409. VINCENT, W. S., HALVORSON, H. O., CHEN, H. R., and SHIN, D. (1968). Ribosomal RNA cistrons in single and multinucleolate oocytes. Biol. Bull. 135, 441. VINOGRAD, J., RADLOFF, R., and BRUNER, R. (1965). Band-forming centerpieces for the analytical ultracentrifuge. Biopolymers 3, 481-489. WOLSTENHOLME, D. R., and DAWID, I. B. (1967). Circular mitochondrial DNA from Xenopus laeois and Rana pipiens. Chromosoma 20, 445-449. WOLSTENHOLME, D. R., and DAWID, I. B. (1968). A size difference between mitochondrial DNA molecules of urodele and anuran amphibia. J. Cell Biol. 39, 222-228.
The Mitochondrial
1-14 (1970)
and Ribosomal
of Oocytes IGOR Deportment
B. DAWID
DNA Components
of Urechis AND
DONALD
caupo D. BROWN
of Embryology,
Carnegie Institution Baltimore, Maryland 21210 Accepted
of Washington,
October 6, 1969
INTRODUCTION
Among the substances stored by oocytes for use during embryogenesis are large numbers of mitochondria and ribosomes. The abundance of these structures is correlated with an increased content of two special types of DNA. These DNAs and their occurrence in oocytes have been studied most extensively in amphibians. The presence of DNA in mitochondria is a general phenomenon, and the preeminence of mitochondrial DNA (M-DNA)’ in frog (Dawid, 1965, 1966) and in sea urchin eggs (Pike et al., 1967) is a consequence of the large number of mitochondria in each egg. The nuclei of oocytes contain extra copies of genes for 28 S and 18 S ribosomal RNA which function in the synthesis of rRNA during oogenesis. Amplification of ribosomal DNA (rDNA) has been demonstrated in oocytes of amphibians (Brown and Dawid, 1968; Gall, 1968) and fish (Vincent et al., 1968) which have multiple nucleoli, and in some insect oocytes which contain a large DNA body surrounding the nucleolus (Lima-de-Faria et al., 1969; Gall et al., 1969). No cytological indication for excess rDNA has been found in oocytes of marine invertebrates which contain only a single large nucleolus in each nucleus. Evidence for rDNA amplification in oocytes of two such organisms, Spisula solidissima and Urechis caupo has been mentioned earlier (Brown and Dawid, 1968), but a recent study did not detect any rDNA amplification in starfish oocytes (Vincent et al., 1969). This paper presents evidence on the abundance and some properties of M-DNA, and on the amplification of rDNA in mature oocytes of Urechis caupo, an echiuroid worm. ’ Abbreviations: SLS, sodium lauryl sulfate; rRNA, ribosomal RNA, defined as 28 S and 18 S RNA; rDNA, ribosomal DNA, defined as the DNA component containing the sequences homologous to rRNA; M-DNA, mitochondrial DNA: GC content, content in deoxyguanylic acid and deoxycytidylic acid; SSC, 0.15 M sodium chloride, 0.015 M sodium citrate; MAK, methylated albumin adsorbed on kieselguhr.
2
DAWID
AND
BROWN
METHODS
Collection of Urechis eggs and sperm. Urechis caupo were collected by Mr. H. Little of Stanford University from the mud flats near San Francisco and shipped by air to Baltimore, where they were maintained at 11-14’C without feeding in an Instant Ocean Culture System (Aquarium Systems, Inc., Wickliffe, Ohio). Urechis oocytes develop free in the coelomic cavity without any follicle cell envelope and are collected into storage sacs when they are mature (Newby, 1940). Mature oocytes were obtained by probing through the external gynophores or by dissecting them from the storage sacs (Gould, 1967). After settling 3 times in sterile seawater, oocytes were suspended in about 4 volumes of 5% bovine serum albumin in seawater and overlaid on 10 ml of 12% albumin layered over 10 ml of 18% albumin in seawater. After centrifugation at 2000 rpm for 15 minutes in a horizontal rotor of the International HN centrifuge, the oocytes were dispersed through the 12% albumin layer and red blood cells and debris had sedimented. The oocytes were removed with a syringe and washed twice with sterile seawater by sedimentation. Such preparations contained less than one blood cell for each 100 oocytes. Purification of DNA from oocytes. Washed oocytes were homogenized in about 20 volumes of SSC (0.15 M NaCl, 0.015 M sodium citrate), sodium lauryl sulfate (SLS) was added (2%), and the homogenate was gently shaken with one volume of phenol overnight at 4“C. The aqueous phase was recovered after centrifugation, two volumes of ethanol were added, and the precipitate was collected and dissolved in about one-tenth of the original volume of 0.02 M Tris-HCl, pH 7.4. The solution was made 0.5% with respect to SLS and 1 M with NaCl and incubated at 4’C for several hours with gentle shaking. More than 80% of the RNA precipitates under those conditons with little if any loss of DNA. The nucleic acids in solution were precipitated with two volumes of ethanol and the precipitate was dissolved in one-tenth of the original volume of 0.02 M Tris, pH 7.4. Pancreatic RNase (0.1 mg/ml), and RNase Tl (50 units/ml), both heated to 80°C for 10 minutes to destroy DNase, were added, and the solution was incubated for 30 minutes at 37°C. Pronase was preincubated for 30 minutes at 37°C and then added at 1 mg/ml; 0.2% SLS was also added and incubation continued for 1 hour. The solution was extracted with phenol overnight, ethanol was added to the aqueous phase, and the precipitate was dissolved in 0.3 M NaCl, 0.05 M sodium phosphate, pH 7.0. The solution was applied to a large methylated albumin
URECHIS OOCYTE
DNA
3
kieselguhr (MAK) column (about 5 ml bed volume per milliliter of packed eggs), and the column was washed with 0.3 M NaCl until the absorbancy at 260 rnF was below 0.02. The DNA was then eluted with 1 M NaCl, 0.05 M sodium phosphate, pH 7.0. The solution was dialyzed against 2 mM Tris-HCl, pH 7.4, concentrated in a flash evaporator, and centrifuged to equilibrium in a CsCl gradient for 65 hours at 33,000 rpm in a No. 65 or No. 40 fixed-angle rotor in a Beckman Spinco centrifuge. Fractions were collected and used for hybridization with rRNA (see below) or the DNA was recovered for other experiments. The yield of DNA was approximately 5 Fbg per oocyte, which represents a 50”; yield (see below). A major unidentified contaminant of oocyte DNA precipitates as a fibrous mass with ethanol and makes concentrated solutions very viscous. Most of it is removed in the MAK fractionation step, the remainder bands at the top of CsCl gradients, where it is separated completely from the DNA. DNA from cell fractions of Urechis oocytes. Urechis oocytes were homogenized at 0°C in about 10 volumes of 0.25 M sucrose, 0.03 M Tris-HCl, 1 mM EDTA, pH 7.4. The suspension was centrifuged for 20 minutes at 1500 rpm in the International centrifuge; the supernatant was collected and centrifuged for 20 minutes at 12,000 rpm in a Sorvall SS-34 rotor to pellet mitochondria. DNA was extracted from the low speed pellet as described above for whole-oocyte DNA. The high speed pellet was extracted by methods that have been applied to purified preparations of mitochondria (Dawid, 1966). DNA from mitochondria of chicken liver was prepared by the same method. Other nucleic acid preparations. Urechis sperm DNA was prepared by SLS-phenol extraction after an initial pronase digestion (Dawid, 1965; Brown and Weber, 1968). Ribosomal RNA labeled with uridine5-“H (Amersham-Searle) was prepared from cells of Xenopus grown in culture (Brown and Weber, 1968). The specific activity was 100 cpm/mwg. Nonradioactive rRNA was purified from unfertilized eggs of Xenopus and oocytes of Urechis (Brown and Weber, 1968). Escherichia coli rRNA was prepared by the same method. All RNA preparations were fractionated on MAK columns before use, concentrated, dialyzed, and finally filtered through prewashed HA Millipore filters (Brown and Weber, 1968). Denaturation and reannealing. DNA was denatured either by heating to 100°C for 10 minutes in solutions of an ionic strength below 0.05, or by exposure to 0.1 N KOH at room temperature for 5 minutes. Reannealing of denatured DNA was carried out at concentrations be-
4
DAWID
AND
BROWN
tween 2 and 10 &g/ml in 0.5 M CsCl at 60°C for 12-14 hours (Dawid and Wolstenholme, 1968). Hybridization. Hybridization with RNA-JH was performed with DNA that had been fractionated in CsCl gradients and immobilized on HA Millipore filters (Brown and Weber, 1968). Electron microscopy. DNA was spread by the method of Freifelder and Kleinschmidt (1965) on a hypophase of 0.3 M sodium acetate and 0.5% formaldehyde. The preparations were rotary shadowed with platinum-iridium and examined in an Hitachi 11-E electron microscope. Magnification was calibrated with a carbon replica of a diffraction grating (Ladd Research Inc.). Analytical centrifugation in CsCl. Methods for equilibrium centrifugation of DNA have been described (Mandel et al., 1968). Band sedimentation through CsCl solutions (1.33 gm/cm”) was carried out in a type I centerpiece of a Beckman ultracentrifuge (Vinograd et al., 1965). RESULTS
The Mitochondrial
AND
DISCUSSION
Component of Oocyte DNA
The properties of oocyte DNA and DNA from purified mitochondria were compared. In CsCl gradients native M-DNA (Fig. la, top) banded at the same buoyant density as sperm DNA (1.699 gm/cm”) (Fig. lb, top). Denaturation shifts the density of both DNAs equally (Figs. la and lb, middle). Reannealed M-DNA had the density of native DNA (Fig. la, bottom) indicating rapid and essentially complete reassociation. Such behavior of M-DNA has been attributed to the homogeneity of its sequences (Borst et al., 1967; Dawid and Wolstenholme, 1968). In contrast to M-DNA, the bulk of sperm DNA did not reassociate (Fig. lb, bottom). A fraction of approximately 5% of the reannealed sperm DNA formed a band at 1.707 gm/cm”. This partially reassociated DNA probably consists of one or more families of highly repetitive sequences which reassociate partially but fail to crosslink with the rest of the nuclear DNA (Britten and Kohne, 1968). Native whole-oocyte DNA (Fig. lc, top) has the same density (1.699 gm/cm”) as sperm or mitochondrial DNA, and on denaturation its density increases to 1.716 gm/cm’ (Fig. lc, middle). When the oocyte DNA was denatured and reannealed, bands were observed at the densities of native and of denatured DNA (Fig. lc, bottom). As shown above, M-DNA regains the density characteristic of native DNA under the reannealing conditions used, whereas the bulk of chromoso-
URECHIS
OOCYTE
5
DNA
A 1.699
I.731 cl
1.731
1.699
b
L
I
I .731
1.699 C
FIG. 1. tinaturation studies of Urechis DNA by CsCl equilibrium sedimentation in the analytical ultracentrifuge. The marker DNA from Micrococcus lysodeikticus, density 1.731 gm/cm’, has been deleted from the densitometer tracings. (a) Top, native M-DNA; middle, denatured M-DNA; bottom, reannealed M-DNA; (b) top, native sperm DNA; middle, denatured sperm DNA; bottom, reannealed sperm DNA; (c) top, native oocyte DNA; middle, denatured oocyte DNA; bottom, reamrealed oocyte DNA.
ma1 DNA remains single-stranded. Three separate preparations of DNA were prepared from Urechis oocytes and subjected to reannealing. the proportion of oocyte DNA which reassociated in these experiments was 67%, 45%, and 72% (average of 61%); this fraction of DNA is considered to be M-DNA. The identification of this rapidly reassociating fraction of DNA as M-DNA was supported by its ability to form common networks during reannealing with a known M-DNA of heterologous origin. Different DNAs form common networks during reannealing only if they share some nucleotide sequences. By this criterion the M-DNA of Xenopus has been shown to contain some sequence homology with
DAWID
AND
BROWN
C / cl
b /
1.709
FIG. ‘2. Corenaturation studies of Urechis DNAs with chick mitochondrial DNA. Bands in CsCl gradients of (a) separately reannealed and subsequently mixed M-DNA of Urechis and the chick; (b) jointly reannealed M-DNA of Urechis and the chick; (c) separately reannealed Urechis oocyte DNA and chick M-DNA; (d) Urechis oocyte and chick M-DNA reannealed in mixture.
chick M-DNA, but not with yeast M-DNA (Dawid and Wolstenholme, 1968). Such an experiment was carried out with M-DNAs from Urechis and the chick. Individually reannealed samples of the M-DNAs separate in CsCl (Fig. 2a). When the two DNAs were reannealed together most of the material formed a single band of intermediate density (Fig. 2b) indicating the formation of common networks. A shoulder of density 1.700 gm/cm” suggests that a small fraction of the Urechis M-DNA did not crosslink with chick M-DNA. Since M-DNA from Urechis and the chick share enough nucleotide sequences to form common networks, this test was also applied to oocyte DNA of Urechis. Separately reannealed samples, when mixed, formed two sharp bands in CsCl gradients (Fig. 2c), in addition to an underlying broad band which represents the nonreassociating component of the oocyte DNA. When Urechis oocyte DNA was rean-
URECHIS OOCYTE
DNA
7
nealed together with chick M-DNA, only a single sharp band appeared in addition to the broad band of DNA which did not reassociate (Fig. 2d). Therefore, the rapidly reassociating fraction of Urechis oocyte DNA forms common networks with chick M-DNA. There is no detectable network formation between nuclear DNA and M-DNA of Urechis during reannealing under these conditions. This finding is in accord with the behavior of other pairs of nuclear and mitochondrial DNAs studied previously (Dawid and Wolstenholme, 1968). We can summarize the constituents of oocyte DNA with the following approximate balance sheet. An oocyte contains a total of about 10 ppg DNA, while only 4 PpLgare expected for a 4C complement of chromosomal DNA (Schwartz, 1969). The reassociation experiments show that about 40% of oocyte DNA (4 grg per oocyte) behaves like nuclear DNA, and the remaining 60 % are mitochondrial DNA (6 ppg per oocyte). The Size and Structure
of Urechis M-DNA
The M-DNAs from all animals studied so far are circular molecules which display special physical properties characteristic of closed (twisted) circles (Borst et aZ., 1967; Dawid and Wolstenholme, 1967). M-DNA from Urechis shares these properties. Sedimentation analysis of Urechis M-DNA demonstrated the fast and slow components previously described for frog M-DNAs (Dawid and Wolstenholme, 1967). The presence of a fast component in neutral solutions and a very fast component in alkali indicates that Urechis M-DNA contains supercoiled (twisted) circular molecules that are characteristic of animal M-DNAs. Electron micrographs of Urechis M-DNA showed many open circular molecules as well as some twisted circles. The contour length of Urechis M-DNA circles was found to be 5.85 f 0.021 ,,, SE (n = 20), compared to 5.86 f 0.048 P SE (n = 15) for Xenopus M-DNA prepared under identical conditons. These observations add to the picture of remarkable uniformity of structure and size of animal MDNAs. The size constancy is striking; the M-DNAs of Urechis, Xenopus, and several other vertebrates (Wolstenholme and Dawid, 1967; Borst et al., 1968) fall into a very narrow size range. While size is retained, avian M-DNAs differ from those of other vertebrates in base composition (Borst et al., 1968) and therefore must have large differences in nucleotide sequences. Nevertheless, the experiment shown in
8
DAWID
AND
BROWN
Fig. 2 suggests that some sequence homology exists between Urechis and chick M-DNA. The uniform size of M-DNAs from different animals despite this divergence in sequences cannot be considered as evidence for a minimum “iimctional length” of 5.5 to 6 P for M-DNA since circles below 5 P have been found in the mitochondria of two urodele species (Wolstenholme and Dawid, 1968). The Ribosomal
Genes in Urechis Oocytes
The content of rDNA in Urechis oocyte DNA was compared to that of sperm DNA by hybridization with radioactive rRNA of Xenopus. First, the approximate extent of homology between Xenopus rRNA and the ribosomal genes of Urechis was measured and the validity of estimating the content of Urechis rDNA with Xenopus rRNA was established. Xenopus rRNA has been shown to have sequence homologies with DNAs from a wide variety of eukaryotes (Sinclair and Brown, 1968). An estimate of this homology with Urechis DNA was made in competition experiments in which nonradioactive Urethis rRNA was used to compete in the homologous hybridization reaction between Xenopus DNA and Xenopus radioactive rRNA. Two identical samples of Xenopus erythrocyte DNA were centrifuged to equilibrium in CsCl, the gradients were collected in fractions, and the DNA in each fraction was immobilized on individual Millipore filters. The two filter sets were split into half filters to give four identical sets and each set was hybridized with rRNA-“H containing a different nonradioactive rRNA as competitor (Fig. 3). Xenopus rRNA effectively competes in the hybridization reaction (less than 2% of the radioactive hybrid remained), E. coli rRNA had no effect, and Urechis rRNA reduced the hybridization reaction by about 60%. This experiment demonstrates a high degree of cross homology between the ribosomal sequences of Urechis and Xenopus. Xenopus rRNA hybridizes with a fraction of Urechis DNA of high buoyant density. As in the case of rDNAs from most eukaryotes, the density of rDNA of Urechis is due to its high content of deoxyguanylic and deoxycytidylic acid (GC) as compared to the bulk of the DNA. In the case of Urechis the bulk DNA bands at 1.699 gm/cm”, while rRNA hybridizes with DNA having a buoyant density of approximately I.715 gm/cm”. These values correpsond to GC contents of 40% and 56%, respectively. The latter value is slightly lower than the reported average base composition of rRNA from Urechis (Gould, 1969a). Hybridization of Urechis DNA with excess radioactive Xeno-
URECHIS OOCYTE
9
DNA
1600
06 800
400
IO
Tube
no.
5 a 0
IO
FIG. 3. Competition for the hybridization of Xenopus rRNA-JH with Xenopus DNA by unlabeled rRNA from different sources. Two identical filters sets of CsCl-fractionated DNA were split into half, and each half set was hybridized with 0.7 fig rRNA-“H and 100 pg of unlabeled rRNA as competitor as indicated, in 2 ml of 4 x SSC overnight at 70%. (a) e-0, no unlabeled RNA added; 0 - - 0, Xenopus rRNA. (b) o-0, E. coli rRNA; 0 - - 0, Urechis rRNA. Thin line, A260. Background has not been subtracted. pus rRNA
demonstrated that about 0.06% of the Urechis genome is homologous to it. No measurement of the fraction of ribosomal sequences in Urechis DNA has been performed with its homologous rRNA. By using the competition experiment of Fig. 3 one can estimate that about 0.1% of Urechis DNA is homologous to rRNA; this value is only a rough estimate because of possible problems in using heterologous rRNA for such measurements. The relative content of rDNA in different DNA preparations of Xenopus has been compared previously by hybridization with subsaturating amounts of rRNA (Brown and Weber, 1968). Such measurements are valid when the DNA has been fractionated in CM21 and is immobilized on nitrocellulose filters. CsCl fractionation increases the specificity of the method since only hybridization with DNA of high density is scored. Since the immobilized DNA does not enter into the kinetics of the reaction the degree of saturation depends only on the concentration of RNA in solution. At a given RNA concentration the level of hybridization is proportional to the total DNA on a filter set. Such measurements are also possible in the heterologous reaction of Xenopus rRNA with Urechis sperm DNA (Fig. 4) since the amount of rRNA bound is linearly proportional to the DNA
10
DAWID
AND
BROWN
FIG. 4. Amounts of rRNA hybridized with varying amounts of DNA on filters. Varying amounts of Urechis sperm DNA and of Xenopus erythrocyte DNA were fractionated in CsCl. All the DNA-filters were hybridized together in 8 ml of 4 X SSC with 3.5 rg of rRNA-JH of Xenopw. The radioactivity bound in the high-density region of each gradient is plotted against the total DNA on the filter set.
on the filter over a 20-fold range of DNA input. Therefore, the fraction of any DNA sample from Urechis homologous to rRNA can be estimated from such a “calibration” curve. Six different preparations of Urechis oocyte DNA and the sperm DNA from three different males were compared for their relative content of rDNA. Oocyte and sperm DNA were fractionated separately in CsCl gradients and the DNA-containing filters hybridized together with subsaturating amounts of rRNA-3H from Xenopus. One such experiment is shown in Fig. 5, and a compilation of the data is presented in Table 1. Whole-oocyte DNA hybridized with about 3 times more rRNA than did equivalent amounts of sperm DNA. Since the M-DNA present in whole oocyte DNA is not expected to hybridize with rRNA (as shown previously with Xenopus M-DNA, Brown and Dawid, 1968) the degree of amplification of rDNA in nuclear DNA of oocytes is expected to be more than two times higher. This conclusion was verified in hybridization experiments with DNA extracted from
11
URECHIS OOCYTE DNA 04
I
Oocy’e
, 1600
DNA
200 j-
i
1 BOOz v
‘b 02 ?I a
400 !!k
" Tube no.
IO
0
FIG. 5. Hybridization of CsCl-fractionated oocyte and sperm DNAs of Urechis with rRNA-‘H. About 20 pg of whole oocyte DNA and 70 pg of sperm DNA were fractionated in CsCl. The DNA-filters prepared from each fraction were hybridized together with 3.5 rg of rRNA-JH in 8 ml of 4 X SSC. --, A2&0; a---@, cpm. Background has not been subtracted. TABLE AMPLIFICATION
OF rDNA
1 IN Urechis
OOCYTES
Hybridization of rRNA-“H with oocyte DNA relative to sperm DNA
DNA
Measured Whole-oocyte
“Nuclear”
DNA
DNA
2.6 3.0 3.3 5.6 5.9
Corrected’ 6.5 7.5 8.3 -
’ Hybridization with rRNA-“H was performed on DNA fractionated in CsCl as shown in Fig. 5 and described in the text. The amount of DNA on filters was determined as described previously (Brown and Weber, 1968). The numbers are derived by dividing the counts per minute of rRNA bound per microgram of total oocyte DNA by the counts per minute of rRNA bound per microgram of total sperm DNA. This ratio is a valid measure of the relative rDNA content as demonstrated in Fig. 4. b Experiments with whole-oocyte DNA were corrected for the M-DNA present (60%) which does not hybridize with rRNA by multiplying the measured value by 2.5. ’ DNA from low-speed pellet as described in the text.
the low speed pellet of oocytes (see Methods). This pellet contains the nuclear DNA complement of the oocyte since essentially pure MDNA was recovered from the supernatant suspension. DNA extracted from the low speed pellet contained about twice the enrichment of rDNA than did whole oocyte DNA (Table 1). A single Urechis oocyte with its four chromosome sets thus contains
12
DAWID AND BROWN
about 24 haploid equivalents of rDNA. This rDNA is localized presumably in its single large nucleolus. This single nucleolus which is commonly found in the oocytes of many marine invertebrates, some insects, and mammals, is almost certainly the site of rRNA synthesis within these cells. In Urechis, disappearance of the nucleolus at meiosis is associated with cessation of rRNA synthesis (Das et al., 1965; Gould, 1969b; Schwartz, 1969). Urechis oocytes, as well as the oocytes of many organisms containing single nucleoli, accumulate relatively large quantities of rRNA. A mature Urechis oocyte contains about 12 rnccg of RNA, and there is little change in rRNA content during development to the trochophore larva (Schwartz, 1969). The same pattern of synthesis and storage of ribosomes in oocytes and their subsequent use during embryogenesis has been found to a greater or lesser degree in all organisms studied to date (Brown and Dawid, 1969). The extension of the correlation or rDNA amplification with synthesis and storage of rRNA to an organism with a single oocyte nucleus supports the idea (Brown and Dawid, 1968) that extrachromosomal copies of rDNA in oocytes are under different control than the rDNA integrated in the chromosomes of somatic cells. This would permit oocytes to uncouple the synthesis of ribosomes from protein synthesis, in distinction to somatic cells, where these two activities are tightly coupled. SUMMARY
Oocytes of Urechis caupo contain 10 bpg DNA (Schwartz, 1969). The oocytes’ 4 chromosome sets account for 4 ppg DNA, and the remaining 6 PpLgwere shown to be mitochondrial DNA. All the DNA from isolated mitochondria of Urechis oocytes and 60% of the whole oocyte DNA reassociate rapidly after denaturation, share some sequence homologies with chick mitochondrial DNA, and contain circular molecules. The contour length of these circular molecules is 5.85 p. Mitochondrial DNA of Urechis is closely similar in physical structure and size to mitochondrial DNA of Xenopus laevis and other vertebrates. The DNA coding for ribosomal RNA (ribosomal DNA) is amplified in Urechis oocytes. Ribosomal DNA of Urechis was measured by hybridization with Xenopus ribosomal RNA, which is homologous to about 60% of the ribosomal sequences in Urechis. The concentration of ribosomal DNA in the nuclear component of oocyte DNA is about 6 times higher than in sperm DNA. The oocyte therefore contains
URECHIS
OOCYTE
DNA
13
about 24 haploid equivalents of ribosomal DNA, which are localized
presumably in its singlelargenucleolus. ACKNOWLEDGMENTS We thank Dr. M. Schwartz for help with the collection Rehhert and E. Jordan for excellent technical assistance.
of Urechis oocytes, and M.
REFERENCES BORST, P., KROON, A. M., and RUTTENBERG, G. J. C. M. (1967). Mitochondrial DNA and other forms of cytoplasmic DNA. In “Genetic Elements: Properties and Functions” (D. Shugar, ed.), pp. 81-116. Academic Press, New York. BORST, P., VAN BRUGGEN, E. F. J., and RLJTTENBERG,G. J. C. M. (1968). Size and structure of mitochondrial DNA. In “Biochemical Aspects of the Biogenesis of Mitochondria,” (E. C. Slater, J. M. Tager, S. Papa, and E. Quagliariello, eds.), pp. 51-69. Adriatica Editrice, Bari. BR~I-~EN, R. J., and KOHNE, D. E. (1968). Repeated sequences in DNA. Science 161, 529-540. BROWN, D. D., and DAWID, I. B. (1968). Specific gene amplification in oocytes. Science 160, 272-280. BROWN, D. D., and DAWID, I. B. (1969). Developmental genetics. Ann. Reu. Genet. 3, 127-154. BROWN, D. D., and WEBER, C. S. (1968). Gene linkage by RNA-DNA hybridization. I. J. Mol. Biol. 34, 662-680. DAS, N., LUYKX, P., and ALFERT, M. (1965). The nucleolus and RNA metabolism in Urechis eggs. Develop. Biol. 12, 72-78. DAWID, I. B. (1965). Deoxyribonucleic acid in amphibian eggs. J. Mol. Biol. 12, 581599. DAWID, I. B. (1966). Evidence for the mitochondrial origin of frog egg cytoplasmic DNA. Proc. Natl. Acad. Sci. U.S. 56, 269-276. DAWID, I. B., and WOLSTENHOLME,D. R. (1967). Ultracentrifuge and electron microscope studies on the structure of mitochondrial DNA. J. Mol. Biol. 28, 233-245. DAWID, I. B., and WOLSTENHOLME,D. R. (1968). Renaturation and hybridization studies of mitochondrial DNA. Biophys. J. 8, 65-81. FREIFELDER, D., and KLEINSCHMIDT, A. K. (1965). Single-strand breaks in duplex DNA of coliphage T7 as demonstrated by electron microscopy. J. Mol. Biol. 14, 271-278. GALL, J. G. (1968). Differential synthesis of the genes for ribosomal RNA during amphibian oogenesis. Proc. Natl. Acad. Sci. U.S. 60, 553-560. GALL, J. G., MACGREGOR, H. C., and KIDSTON, M. E. (1969). Gene amplification in the oocytes of dytiscid water beetles. Chromosoma 26, 1699187. GOULD, M. C. (1967). Echiuroid worms: Urechis, In “Methods in Developmental Biology” (F. H. Wilt and N. K. Wessells, eds.), pp. 163-171. Crowell, New York. GOULD, M. C. (1969a). RNA and protein synthesis in the unfertilized eggs of Urechis caupo. Deuelop. Biol. 19, 460-481. GOULD, M. C. (1969b). A comparison of RNA and protein synthesis in fertilized and unfertilized eggs of Urechis caupo. Develop. Biol. 19, 482-497. LIMA-DE-FARIA, A., BIRNSTIEL, M., and JAWORSKA,H. (1969). Amplification of ribosomal cistrons in the heterochromatin of Acheta. Genetics 61, Suppl., 145-159.
DAWID ANDBROWN MANDEL, M., SCHILDKRAUT, C. L., and MARMUR, J. (1968). Use of CsCl gradient analysis for determination of guanine plus cytosine content of DNA. In “Methods in Enzymology” (S. P. Colowick and N. 0. Kaplan, eds.), Vol. XII B, pp. 184-195. Academic Press, New York. NEWBY, W. W. (1940). “The Embryology of the Echiuroid Worm Urechis caupo.” American Philosophical Society, Philadelphia. PIKO, L., TYLER, A., and VINOGRAD, J. (1967). Amount, location, priming capacity, circularity and other properties of cytoplasmic DNA in sea urchin eggs. Biol. Bull. 132, 68-90. SCHWARTZ, M. (1969). Nucleic acid metabolism in oocytes and embryos of the echiuroid Urechis cuupo. Ph.D. Thesis, The Johns Hopkins University, Baltimore, Maryland. SINCLAIR, J. H., and BROWN, D. D. (1966). Comparative studies on genes for ribosomal RNA in eukaryotes. Carnegie Inst. Wash. Year Book 67, 404-409. VINCENT, W. S., HALVORSON, H. O., CHEN, H. R., and SHIN, D. (1968). Ribosomal RNA cistrons in single and multinucleolate oocytes. Biol. Bull. 135, 441. VINOGRAD, J., RADLOFF, R., and BRUNER, R. (1965). Band-forming centerpieces for the analytical ultracentrifuge. Biopolymers 3, 481-489. WOLSTENHOLME, D. R., and DAWID, I. B. (1967). Circular mitochondrial DNA from Xenopus laeois and Rana pipiens. Chromosoma 20, 445-449. WOLSTENHOLME, D. R., and DAWID, I. B. (1968). A size difference between mitochondrial DNA molecules of urodele and anuran amphibia. J. Cell Biol. 39, 222-228.