ABSTRACT
The present paper reports basic data on DNA content, protein content, and protein synthesis in Triturus pyrrhogaster embryos during development from cleavage to the hatching stage. Except for measurements of DNA and total protein contents, embryos were labeled with sodium carbonate-14C for 10 h and fractionated into embryonic cell components, i.e. cytoplasmic mass, yolk and pigment granules, and nuclei, in a discontinuous density gradient of sucrose. The protein content and the radioactivity incorporated into protein were measured in each fraction. Those fractions combining protein soluble in buffer at pH 8·3 and in 0·25 N-HCl were further studied with polyacrylamide gel electrophoresis.
In the newt embryo, four stages of active DNA increase were observed when cultured at constant temperature; they were gastrula, neurula, late tail-bud, and before-hatching stages. Total protein per embryo decreased from 3 to 2 mg during the development studied. The content of cytoplasmic soluble protein per embryo was low and constant throughout development. Synthesis of the fraction was observed at the earliest stage of development studied though the rate was not high and specific activity of the soluble protein increased during development. Qualitative changes in the newly synthesized protein were observed. With the yolk fraction, synthesis of protein, other than from probable contamination with the cytoplasmic fraction, was not detected and a detailed description was omitted.
Changes were observed at two stages of development in the synthesis of nuclear protein soluble in buffer at pH 8·3, the first at gastrulation and the second at late tail-bud stage. The change at gastrulation seemed to be the start of syntheses of the nuclear soluble proteins, while quantitative enhancement rather than qualitative change was noticed at late tail-bud stage. Most of the nuclear protein soluble in 0·25 N-HCl was histone. The histone content increased in accordance with increase in the DNA content and the rate of DNA accumulation was accompanied by proportionate incorporation of radioactivity into histone. Among histone fractions, unique behaviour of the very lysine-rich histone was observed.
The availability of [14C]sodium carbonate in rough estimations of protein synthesis in embryos and significance of the data obtained have been discussed.
INTRODUCTION
Syntheses of proteins are some of the most fundamental events in the development of embryos and various proteins with their unique functions are synthesized and degraded. This program, which differs for each organism, is determined genetically and is carried out through intervention of messenger RNAs. Though there can be no doubt that studies on regulation mechanisms of protein synthesis in developing embryos, including regulation mechanisms of gene transcription, are of utmost importance and that the studies have solved not a few problems in developmental biology (see Davidson, 1968), it would be useful to know the gross program of protein synthesis; that is, when and what kind of protein is to be synthesized during development of the embryo, to understand development on a molecular basis.
Concerning protein synthesis in amphibian embryos, soluble and pellet protein in Rana (Brown & Caston, 1962) and cytoplasmic and nuclear protein in Rana (Ecker & Smith, 1971) have been reported among others. The present report is an analysis of protein synthesis in Triturus embryos during development.
MATERIAL AND METHODS
Embryos of Triturus pyrrhogaster (BOIE) were used throughout the experiments. They were cultured at 21 ± 1°C and staged according to the tables of Okada & Ichikawa (1947).
(a) Labeling of embryos and fractionation of embryonic cell components
Preparation of embryos for labeling and the detailed procedures of labeling have been reported elsewhere (Cohen, 1954; Imoh, Sasaki, Kawakami & Hayashi, 1972). Embryos in various developmental stages were labeled together in the same 50 ml vial with 100 μCi of Na214CO3 (2·0 mCi/ml, the Radiochemical Centre, Amersham) for 10 h and, at the end of labeling, they were washed with Holtfreter’s solution and grouped according to their developmental stages. The stages and numbers of embryos are shown in Table 1. The availability of the precursor in the study of protein synthesis in embryos during development is considered in the Discussion.
Labeled embryos were homogenized with 10 ml of 0-88 M sucrose in Tris-HCl (pH 7·2) containing 3 mM-CaCl2 in a glass homogenizer provided with a Teflon pestle (clearance 0·08 mm) at 600 rev/min. The homogenate was floated on two layers of sucrose solution, 14 ml of 1·8 M on 16 ml of 2·1 M, in a centrifugation tube. Each sucrose solution was prepared with 10 mM Tris-HCl (pH 7·2) containing 3 mM-CaCl2. After centrifugation in a Spinco SW 25 rotor at 23000 rev/min for 90 min, embryonic cell components were recovered from each density boundary or the bottom; nuclei were obtained from the bottom, yolk and pigment granules from the boundary between the 1·8 and the 2·1 M sucrose layer, and most of the remaining cytoplasmic mass from the boundary between the 0·88 and the 1·88 M layer. The nuclei were almost free from contamination by cytoplasmic or yolk material, although the latter was a little contaminated with nuclei because of a few intact cells (Imoh & Negami, 1972).
The DNA content of embryos during development was measured by the diphenylamine method (Burton, 1956) with the sample prepared by the procedures of Schmidt & Thannhauser (1945) from homogenate of whole unlabeled embryos or with the nuclei isolated and freed from histones (Imoh & Kawakami, 1973). About 90% of DNA in the homogenate was recovered in the nuclei fraction.
(b) Measurement of content and synthesis of protein
The protein content of whole embryos was determined by the method of Lowry, Rosebrough, Farr & Randall (1951) with a sample prepared by the procedures of Schmidt & Thannhauser (1945) from homogenate of whole unlabeled embryos.
The isolated cytoplasmic material from the homogenate of labeled embryos was diluted with Tris-HCl (pH 7·2) and then ethanol was added to the sample to make a final concentration of 67%. This was then centrifuged at 6000 rev/min for 10 min. The resulting precipitate was extracted with Tris-HCl (pH 8·3) for 2 h in the cold and centrifuged at 10000 rev/min for 10 min to remove unextracted components. Perchloric acid (PCA) was added to an approximately equal aliquot of the supernatant, so as to bring the final concentration of PCA to 0·5 N; this was then kept at 90°C for 20 min. The protein content of the sample was determined by the method of Lowry et al. (1951). The same volume of the sample was processed by the same procedures and the protein precipitate produced by the PCA treatment was caught on Millipore filter and dried. Toluene-based scintilation fluid was added for counting in a liquid scintilation counter (Packard TRI CARB). The remainder of the extract of cytoplasmic protein was subjected to electrophoresis on polyacrylamide gel.
The yolk fraction was diluted about fivefold with the buffer (Tris-HCl, pH 7·2) to reduce the sucrose concentration and centrifuged at 10000 rev/min for 10 min. The precipitate was extracted with Tris-HCl (pH 8·3) for 2 h in the cold and centrifuged for 10 min at 10000 rev/min. The extract was analysed by the same procedure as for the cytoplasmic soluble protein and the precipitate was extracted with 0·25 N HCl for 2 h in the cold. The acid-soluble protein was analysed as above, except that the conditions of electrophoresis differed.
The isolated nuclei were washed briefly with Tris-HCl (pH 7·2) and extracted with Tris-HCl (pH 8·3) for 2 h in the cold. The supernatant was analysed as cytoplasmic soluble protein and the precipitate was extracted with 0·25 N-HCl for 2 h in the cold and was analysed by the same procedure as for yolk acid soluble protein.
(c) Gel electrophoresis of soluble and acid soluble protein
The electrophoresis of the protein soluble in buffer at pH 8·3 was performed on 7·5% polyacrylamide gel by the method of Williams & Reisfeld cited by Nagai (1966). Electrophoresis was conducted with current of 4 mA per gel until the tracking dye, bromphenol blue, had run 5·0 cm. At the termination of electrophoresis the gel was stained with amidoblack 10B solution for several hours and destained electrically. The gel was sliced into disks 1 mm thick and these were liquefied by treatment with 35% H2O2 at 60°C for 1 h. To the lysate were added ethanol and the toluene-based scintilation fluid for counting in a liquid scintilation counter. The results were reproducible.
For fractionation of the protein soluble in 0·25 N-HC1, 15% polyacrylamide gel electrophoresis was used according to the method of Shepherd & Gurley (1966) with a glycine buffer (pH 4·0) instead of the valine buffer. After electrophoresis the gel was stained with amidoblack 10 B solution for 12 h and destained electrically. The optical density of the stained protein bands was recorded with a densitometer (Fujiriken, FD-A4). The quantity of protein in a band was calculated from the densitometer trace with reference curves which had been made with known amounts of protein. Radioactivities in the protein bands were determined in the same way as with the protein soluble in buffer at pH 8-3. The characterization of the acid-soluble protein from nuclei has been reported elsewhere (Imoh & Kawakami, 1973).
RESULTS
(a) Changes in contents of DNA and protein during development
The changes in DNA content per embryo during development are shown in Fig. 1. There were four stages where the increase of DNA was greater than at other stages when cultured at constant temperature; this was recognized by calculating the rate of DNA accumulation (Fig. 1, dotted line). They were gastrulation (st. 12), middle neurulation (st. 17), late tail-bud formation (sts. 29-30), and late foreleg (sts. 37-39) stages.
Content of DNA and rate of DNA increase per embryo. From groups of newt embryos cultured at 21 ± 1°C. An appropriate number of embryos at a given stage selected, homogenized, and processed by the method of Schmidt & Thannhauser (1945). The DNA content of the sample was determined by the diphenylamine method (Burton, 1965). From the data, μ g DNA per embryo (○ — ○) and the rate of DNA accumulation per embryo, i.e. μ g DNA increase per 12 h per embryo (.....),were calculated.
Content of DNA and rate of DNA increase per embryo. From groups of newt embryos cultured at 21 ± 1°C. An appropriate number of embryos at a given stage selected, homogenized, and processed by the method of Schmidt & Thannhauser (1945). The DNA content of the sample was determined by the diphenylamine method (Burton, 1965). From the data, μ g DNA per embryo (○ — ○) and the rate of DNA accumulation per embryo, i.e. μ g DNA increase per 12 h per embryo (.....),were calculated.
Fig. 2 shows changes in the content of total or fractions of protein per embryo during development. Total protein was about 2 mg/embryo at the foreleg stage. Because of ethanol soluble lipoprotein, which had been removed in the procedure of determination, the data were lower than the true value by about 1 mg/embryo. The content of cytoplasmic soluble protein was low and constant throughout development; soluble protein obtained from the yolk fraction, which is not shown in the figure, was lower than cytoplasmic soluble protein. Nuclear protein soluble in buffer at pH 8·3 was low before the tail-bud stage, increased rapidly at tail-bud stage, and remained almost constant thereafter. The nuclear protein soluble in 0·25 N-HC1, most of which was histone, increased throughout development in a pattern quite similar with that of the DNA content.
Changes in the contents of protein fractions during development. Protein contents of whole embryo and of cell components isolated by the sucrose density gradient were determined at various stages of development. Notation: total protein in mg embryo (▵ — ▵) and cytoplasmic soluble protein (× — ×), nuclear soluble protein (● — ●),and nuclear acid-soluble protein in μ g per embryo (○ — ○).
Changes in the contents of protein fractions during development. Protein contents of whole embryo and of cell components isolated by the sucrose density gradient were determined at various stages of development. Notation: total protein in mg embryo (▵ — ▵) and cytoplasmic soluble protein (× — ×), nuclear soluble protein (● — ●),and nuclear acid-soluble protein in μ g per embryo (○ — ○).
(b) Quantitative changes in the syntheses of protein fractions
Changes in the rate of radioisotope incorporation into protein fractions per embryo and in the specific activities of protein fractions during development are represented in Figs. 3 and 4, respectively. The rate of cytoplasmic soluble protein synthesis per embryo gradually increased during development and the rate of soluble protein synthesis from the yolk fraction showed the same pattern as cytoplasmic soluble protein synthesis, though much lower and not shown in the figure, suggesting that soluble protein from the yolk fraction was contamination with cytoplasmic soluble protein. Synthesis of HC1 soluble protein in the yolk fraction was not detected. Nuclear soluble protein synthesis was not detected at the blastula stage, became positive at gastrula stage, and was enhanced at the late tail-bud stage. It may be noted that the rate of nuclear acidsoluble (basic) protein synthesis was greater, as was the rate of DNA accumulation, at gastrula, neurula and tail-bud stages than at other stages. The specific activity of acid-soluble protein from nuclei was very high at the gastrula stage and decreased thereafter, while that of cytoplasmic soluble protein increased throughout development with a temporal decrease at the late neurula stage, and that of nuclear soluble protein was roughly constant after the gastrula stage (Fig. 4).
Radioactivities incorporated into protein fractions. Embryos were labelled with 14CO2 for 10 h and homogenized. The cell components were isolated and the protein fractions were extracted from them. Radioactivities in the protein fractions were measured and calculated on a per embryo basis. Notation: cytoplasmic soluble protein (× — ×), nuclear soluble protein (● — ●), and nuclear acid-soluble protein in cpm per embryo (○ — ○).
Radioactivities incorporated into protein fractions. Embryos were labelled with 14CO2 for 10 h and homogenized. The cell components were isolated and the protein fractions were extracted from them. Radioactivities in the protein fractions were measured and calculated on a per embryo basis. Notation: cytoplasmic soluble protein (× — ×), nuclear soluble protein (● — ●), and nuclear acid-soluble protein in cpm per embryo (○ — ○).
Changes in the specific activities of protein fractions during development. The specific activities of protein fractions were calculated from the data shown in Fig. 2 and Fig. 3. Notation is the same as in Fig. 3.
(c) Qualitative changes in the newly synthesized soluble protein fractions
In Fig. 5 are shown electrophoretic patterns of the radioactivity incorporated into the cytoplasmic soluble protein at several stages of development. As about 10 μg of soluble protein was applied to each gel, the patterns were comparable to each other. At the blastula stage protein synthesis occurred, but only small amounts of newly synthesized protein were found in the protein applied to the gel. The radioactivity incorporated in the protein increased with development and at the tail-bud stage one fraction showed extremely high radioactivity (Fig. 5d), though it gradually decreased through later development (Fig. 5e–g). The nature of the fraction has not been examined. Fig. 6 shows electrophoretic patterns of newly synthesized nuclear soluble protein. Radioactivity incorporated into the protein at the blastula stage was not significantly above background level. Incorporation into several protein fractions was evident after the gastrula stage and the pattern was essentially identical with that of tail bud embryos in the embryo beyond the tail-bud stage.
Polyacrylamide gel separation of radioactivities incorporated into cytoplasmic soluble protein. About 10 μ g of cytoplasmic soluble protein from labeled embryos was subjected to electrophoresis in 7·5% polyacrylamide gel. After electrophoresis, gel was stained with amidoblack, destained electrically, and sliced into 1 mm disks. The slices were liquefied and radioactivities in them were measured. The sample gel was at the side of slice number zero, (a) Blastula, sts. 9–10; (b) gastrula, st. 12; (c) neurula, sts. 16–17; (d) tail-bud stage, sts. 26–27; (e) balancer developing stage-c, st. 32; (f) the second foreleg stage-a, 34; (g) the third foreleg stage-a, st. 36.
Polyacrylamide gel separation of radioactivities incorporated into cytoplasmic soluble protein. About 10 μ g of cytoplasmic soluble protein from labeled embryos was subjected to electrophoresis in 7·5% polyacrylamide gel. After electrophoresis, gel was stained with amidoblack, destained electrically, and sliced into 1 mm disks. The slices were liquefied and radioactivities in them were measured. The sample gel was at the side of slice number zero, (a) Blastula, sts. 9–10; (b) gastrula, st. 12; (c) neurula, sts. 16–17; (d) tail-bud stage, sts. 26–27; (e) balancer developing stage-c, st. 32; (f) the second foreleg stage-a, 34; (g) the third foreleg stage-a, st. 36.
Polyacrylamide gel separation of radioactivities incorporated into nuclear soluble protein. About 10 μg of soluble protein from nuclei was fractionated. For details, see text or the legend of Fig. 5. (a) Blastula sts. 9–10; (b) gastrula st. 12; (c) neurula sts. 16–17; (d) tail-bud sts. 26–27.
(d) Changes in contents and syntheses of fractions of nuclear acid-soluble protein
Figs. 7 and 8 show electrophoretic patterns of nuclear protein soluble in 0·25 N-HCl, from embryos at four stages. In Fig. 7 traces of density patterns of the stained gels are shown and the peaks observable between electrophoretic mobilities of 1·5–3·3 cm are histone fractions: very lysine-rich (fl), arginineand alanine-rich (f3), two slightly lysine-rich (lysine- and serine-rich f2b and alanine- and leucine-rich f2a2) as one continuous peak, and arginine- and glycine-rich (f2al) in the order of increasing mobility. It should be noted that fl at the blastula stage consisted of two peaks and one of them with lower mobility increased at the gastrula stage. The peak between fl and f3 has not been identified. It should also be noticed that the mutual ratio between the three peaks (f3, f2b + f2a2, and f2al) did not change much, depending on stage, while the ratio of fl to the three peaks changed during development, being low at blastula and tail bud stages and high at the neurula stage. This is shown more accurately in Fig. 9(a), which shows changes in the ratio of content of each histone fraction to the total histone content during development. Though the experimental values were considerably scattered, the ratio of f2b + f2a2, f2al, or f3 could be regarded as constant during development. On the other hand, the ratio of fl positively elevated at the neurula stage. Almost the same statements could be made about radioactivities incorporated into newly synthesized histone fractions (Fig. 8, Fig. 9b).
Polyacrylamide gel separation of acid soluble protein from nuclei. The acidsoluble protein was extracted with 0-·25 N-HCI from nuclei isolated from labeled embryos and fractionated by 15% polyacrylamide gel electrophoresis. The gel was stained with amidoblack for 12 h, destained electrically, and traced with a densitometer. The amount of sample applied to the gel was about 0·05 ml and protein content varied among samples, (a) Blastula sts. 9–10; (6) gastrula st. 12; (c) neurula sts. 16–17; (d) tail-bud sts. 26–27.
Polyacrylamide gel separation of acid soluble protein from nuclei. The acidsoluble protein was extracted with 0-·25 N-HCI from nuclei isolated from labeled embryos and fractionated by 15% polyacrylamide gel electrophoresis. The gel was stained with amidoblack for 12 h, destained electrically, and traced with a densitometer. The amount of sample applied to the gel was about 0·05 ml and protein content varied among samples, (a) Blastula sts. 9–10; (6) gastrula st. 12; (c) neurula sts. 16–17; (d) tail-bud sts. 26–27.
Ratio of each histone fraction to total histone. The content and radioactivity in each histone fraction was measured from Fig. 7 and Fig. 8 respectively, and the total histone content or the total radioactivity was determined as the sum of them. The ratio of each histone fraction to the total was calculated in percent, (a) The ratio in content; (b) the ratio in radioactivity. ○ — ○, The very lysine-rich fl; (× — ×), arginine and alanine-rich f3: ▵ — ▵, two slightly lysine-rich f2b + f2a2: ● — ●, arginine and glycine rich f2al.
Ratio of each histone fraction to total histone. The content and radioactivity in each histone fraction was measured from Fig. 7 and Fig. 8 respectively, and the total histone content or the total radioactivity was determined as the sum of them. The ratio of each histone fraction to the total was calculated in percent, (a) The ratio in content; (b) the ratio in radioactivity. ○ — ○, The very lysine-rich fl; (× — ×), arginine and alanine-rich f3: ▵ — ▵, two slightly lysine-rich f2b + f2a2: ● — ●, arginine and glycine rich f2al.
DISCUSSION
It is probable that permeability of embryonic cells to the radioactive precursor, 14CO2, may change during development and, furthermore, the ratelimiting steps in fixation of CO2 to amino acids are unknown. However, as noted by Brown & Caston (1962), it may be assumed that once an amino acid has been formed, it will be incorporated into the class of proteins being most actively synthesized at that particular time, in the statistical sense. Thus, by using 14CO2 as precursor, a correct answer may be obtained concerning what class of proteins is more actively synthesized than others in the embryo, at a given stage of development. This statement does not exclude the possible increase of radioisotope incorporation into all classes of proteins during development. Examination of the data on histone synthesis, however, suggests that the incorporation of 14CO2 into the various protein classes can be used for rough comparison of protein syntheses at different stages of development. The histone is synthesized without any major qualitative changes in composition after gastrulation (Kischer & Hnilica, 1967; Hnilica & Johnson, 1970) with a definite ratio to newly synthesized DNA (Imoh & Kawakami, 1973), though histone synthesis before gastrulation and synthesis of fl fraction during development are controversial (Vorobyev, Gineitis & Vinogradova, 1969; Asao, 1969). Therefore, a major part of histone synthesis after gastrulation seems to take place in relation to DNA synthesis and, in the present experiment, the rate of 14CO2 incorporation into nuclear basic protein was roughly parallel to the rate of DNA accumulation (compare Figs. 1 and 3). Accordingly, the rate of 14CO2 incorporation into histone may roughly represent the true rate of histone synthesis. As the mechanisms of histone synthesis are the same as those of the synthesis of any other protein (Borun, Scharff & Robbins, 1967; Kedes, Gross, Cognetti & Hunter, 1969; Nemar & Lindsay, 1969), it may be assumed that an approximately correct estimate of the true rate of synthesis of each protein during development would be obtained from determination of the rate of 14CO2 incorporation.
Cytoplasmic soluble protein was extracted from the cytoplasmic mass precipitated by ethanol. It is possible that not all the soluble protein was recovered after ethanol precipitation because of denaturation, though our preliminary study with polyacrylamide gel electrophoresis suggested that the qualitative change in the protein fractions before or after ethanol precipitation was small. The content of this fraction per embryo was almost constant throughout development despite a tremendous increase in cell number. The result resembled that of Brown & Caston (1962) on Rana soluble protein. The specific activity of the fraction increased during development without accumulation of protein content, suggesting a high rate of turnover. Localization to cytoplasm, solubility in pH 8-3 buffer, and high rate of turnover suggested that this fraction was composed of cytoplasmic enzymes. The fraction consisted of many protein species. Their synthesis was found at the earliest stage of development studied and one of them was synthesized at quite a high rate at the tail-bud stage though its nature was unknown.
The nature of the yolk fraction protein soluble in buffer at pH 8·3 was identical with that of cytoplasmic soluble protein; and acid-soluble protein of yolk showed no incorporation of radioactivity.
In relation to nuclear protein soluble in buffer at pH 8·3, two stages of development require comment. The first is gastrulation. At the blastula stage, incorporation of radioactivity into the fraction was not found and no peak relating to newly synthesized protein was observed in the electrophoretic pattern. As incorporation of radioactivity into histone was fairly high and peaks of the histone fractions were apparent in the electrophoretic pattern at the same stage, the failure to detect radioactivity in the nuclear soluble protein could not be attributed to deficiency of labeled amino acid and it must be concluded that synthesis of nuclear soluble protein at the blastula stage was nonexistent or very low. On the other hand, at the gastrula stage, incorporation of radioactivity in the fraction and the specific activity of the fraction were higher and several fairly high peaks were observed in the electrophoretic pattern. The change at gastrulation thus seemed to be the start of synthesis of the fraction, which might have some relation to the beginning of gene transcription occurring at gastrulation in amphibian embryos (see Davidson, 1968). The second stage to be noted is the tail-bud stage when the content of the fraction per embryo or the rate of synthesis per embryo increased greatly. Electrophoretic analysis of the newly synthesized protein suggests that the change was a quantitative enhancement of synthesis rather than a qualitative switching-over. The enhancement could be related to a probable increase in differentiating cells; but this remains to be studied.
Most of the nuclear protein soluble in 0·25 N-HCl was histone and it could be fractionated into five major subfractions and identified (Johns & Butler, 1962; Johns, 1967; Imoh & Kawakami, 1973). As discussed above, in rough approximation, the histone seemed to increase in relation to DNA increase. There are a few points to be noticed, however. The very-lysine-rich, fl, fraction at early stages of development consisted of a few subfractions. The heterogeneity of verylysine-rich histone reported in adult tissues (Nelson & Yunis, 1969; Panyim, Bilek & Chalkley, 1971) seemed to have its origin in early development, though examination of embryonic tissues by electrophoretic analysis with higher resolution will be needed. The ratio of very-lysine-rich histone to total histone, with both content and incorporated radioactivity, changed during development; low before gastrulation, high at neurulation, low again at late tail-bud formation, and high again before hatching, while the ratios of other histone fractions to the total histone were almost constant. The percentage of fl cannot be independent from those of other histone fractions and the change of fl during development would be emphasized by taking a fraction, e.g. f2b + f2a2, as unity, though experimental fluctuation in the fraction taken as standard, e.g. f2b + f2a2, cannot be excluded. The significance of the fl change during development, especially at early stages, is now under study.
Acknowledgement
Looking over the results, it may be noted that the rate of nuclear protein synthesis exceeded largely that of cytoplasmic soluble protein, especially during early development, and it was only after the tail-bud stage that the specific activity of cytoplasmic soluble protein became higher than those of nuclear proteins. Though insoluble protein has not been studied in the present experiments, the larger synthesis of nuclear protein than cytoplasmic protein observed in the early development of fishes and echinoderms (Krigsgaber, Kostomarova, Terekhova & Burakova, 1971) or in early cleavage of Rana (Ecker & Smith, 1971) may have some relation to the present observations.