The dorsal or the ventral cortex of Xenopus laevis eggs, at the one- or two-cell stage, has been slightly injured with a microneedle; 184 of these eggs have been examined cytologically.

The frequency of early (cleavage) and late (gastrulation, neurulation) abnormalities after dorsal injury was about twice that after ventral injury.

Abnormalities of the mitotic apparatus and of the chromosomes are very frequent during cleavage after slight cortical damage.

At later stages, the previously wounded embryos often display aneuploidy, as shown by cytophotometric measurements of the DNA content of the individual nuclei. Aneuploidy is much more frequent after injury of the dorsal cortex than injury of the ventral cortex.

Autoradiography studies have shown that there is a close correlation between morphogenesis and nuclear RNA synthesis.

The results are discussed in relation to current ideas on the morphogenetic and genetic significance of the grey crescent and its cortex.

Classical experiments by A. Brachet (1906) and Pasteels (1932) have demonstrated that localized destructions of the grey crescent are followed, in frogs’ eggs, by abnormalities of the axial organs (nervous system, chorda). The conclusion drawn was that the grey crescent, which is located in the dorsal side of the fertilized egg, behaves like a ‘germinal localization’. Such experiments have been taken as a proof that the organization of the cytoplasm plays a leading role in the control of morphogenesis during the early stages (fertilization, cleavage) of amphibian development. They led Albert Brachet (1931) to the idea that a distinction should be made between two kinds of heredities during embryonic development: a ‘general’ heredity which involves both chromosomal and cytoplasmic factors, and a special heredity which is the classical Mendelian one. The former would be responsible for the development of a species-specific organ (the eye, for instance), while the latter would control the formation of late characters such as the appearance of eye pigmentation.

According to Pasteels (1932), superficial injuries of fertilized eggs have less effect than deep destructions of the dorsal cytoplasm: he concluded that the morphogenetic territories must lie deep inside the recently fertilized egg. But more recent work by Curtis (1960) on Xenopus has led to quite different conclusions: he worked out a technique for the grafting of cortical material from one egg to another and concluded that ‘the cortical material definitely possesses morphogenetic properties which may be transferred with it’. In particular, this cortical material would play a leading role in the initiation of the gastrulation movements and, by way of consequence, in neural induction : this is in agreement with the theory of Dalcq & Pasteels (1937) which attempts to explain amphibian development on the basis of an interaction between a cortical field (with its centre in the grey crescent) and a yolk gradient (with a maximum at the vegetal pole).

Still more recent work by Curtis (1965) has led to the intriguing possibility of the existence of a cortical inheritance in Xenopus: adults originating from eggs which have undergone an operation upon their dorsal cortex were mated with normal animals. An effect, which was arrest of development just before gastrulation, was obtained at low frequency in the same generation and at increasing frequencies in the three following generations. Of special interest is the fact that the source of the sperm (from a normal male or from a male originating from an injured egg) has no effect on the incidence of pregastrular arrest. After discussing various possible explanations for the results, Curtis (1965) concludes that the operation probably sets up a population of replicating molecules responsible for the effect on gastrulation: the cortex of the grey crescent would have properties of self-replication, possibly due to the presence of particles endowed with genetic continuity.

These observations led one of us, in 1967, to the following hypothesis: since there is good reason to think that neural induction is the result of gene derepression (reviews by Brachet, 1967, and by Tiedemann, 1968) and since injuring the cortex of the grey crescent in the fertilized egg prevents gastrulation and medullary plate formation, it is conceivable that this cortex somehow controls the synthesis of specific ribonucleic acids (RNA’s) in the nuclei. This control would be a remote one, exerted by the cytoplasm on the nuclei at a later stage : injury of the dorsal cortex, in the recently fertilized egg, would affect RNA synthesis at the gastrula stage only. The existence of such a remote control has been found by Davidson et al. (1965) in the eggs of Ilyanassa: removal of the polar lobe (which contains no nucleus) delays the onset of RNA synthesis, many hours later, in the ‘lobeless’ embryos.

The aim of the present work was to test this hypothesis: the immediate cytochemical effects of injuries of the egg cortex (which had never been studied) have been examined and the delayed effects of such injuries on nucleic acid and protein synthesis have been studied by autoradiography.

All the experiments have been performed on eggs of Xenopus laevis obtained by injecting females with 750 units of Pregnyl (Organon). The fertilized eggs were operated, at the one- or two-cell stage, following the method of Curtis (1965): a tungsten microneedle was inserted in the cortex of either the dorsal or the ventral side and was used to produce a small superficial lesion. In a few experiments the eggs were pricked more deeply with the needle, in view of the above-mentioned results of Pasteels (1932) on frog eggs, but the results were not markedly different. In one experiment, in which the eggs had been pricked only once and very lightly, the development was normal in all respects. It should be pointed out that the mode of operation was different from that used by Curtis (1960, 1965) in his grafting experiments, where the wounding was carried out in the presence of EDTA: this chelating agent was not used, because previous experiments by one of us (J.B., unpublished) have shown that it can induce cell death in amphibian gastrulae.

Experiments have been performed on 16 different batches of eggs; the total number of operated eggs was 484. But in the great majority of the cases (i.e. in 300 eggs) the operation was followed by fast cytolysis : the eggs of certain batches showed such fragility that very few cleaved normally after injury in the cortex. The eggs cultivated in dilute Holtfreter or in tap water.

Cytological observations were made on the remaining 184 eggs: 96 had been operated upon the dorsal cortex (D), 88 upon the ventral cortex (V) and 70 control embryos had not been operated. The eggs were fixed with Zenker acetic at various time intervals, ranging from 45 min to 2 days after the operation. Paraffin sections were alternately stained with Unna (methyl green-pyronine) and Feulgen, under routine conditions.

Some of the Feulgen-stained embryos were submitted to a cytophotometric analysis using a Barr and Stroud microdensitometer: 20 embryos, originating from five different experiments, have been analysed for nuclear DNA content, after reaching the gastrula or the neurula stage. 50–100 measures were made in each of the three following regions: neuroblast, chordomesoblast and entoblast in the case of gastrulae 23 h old; ectoderm, somitic mesoderm and endoderm in neurulae 29 to 48 h old (total number of measures : about 5000). The frequency of the mitoses and pycnoses was also measured in these 20 embryos.

For autoradiography, except for two experiments in which gastrulae, having been cut into two, were immersed in a [3H]uridine-containing solution, the labelled precursors were micro-injected into the blastocele about 8 h after the cortex had been injured. The following precursors were used :
formula

The injected volume was 0 ·018 μI; the solutions were made either in Holt-freter medium or in tap water.

The eggs were fixed 2–3 h after micro-injection of the radioactive precursor, by freeze-substitution and the autoradiographs were developed after 27 days. The autoradiography technique, with Ilford liquid emulsion K5 and Unna staining, was that described by Ficq (1959). Grain counts were made over the nuclei of the animal pole, the dorsal lip of the blastoporus and the vegetal pole in the case of gastrulae (circular blastoporus), in the animal half in blocked blastulae.

Three different experiments were made with [3H]uridine: 20 radioactive embryos were examined under the microscope and grains were counted in 6 of them (3 D, 2 V, 1 control), 8 embryos (3 D, 3 V, 2 controls) were counted after [3H]thymidine injection, and 4 (2 D, 1 V, 1 control) after leucine injection.

1. General observations

Even if one disregards the fragile eggs which cytolyse quickly after operation, it is clear that many of the cortically injured eggs stop development already during cleavage: we shall therefore distinguish between early abnormalities (45 min to 20 h after operation) and late (between 20 and 48 h after operation).

Observation of the living eggs under the dissecting microscope shows that injuries of either the dorsal or the ventral cortex can produce both early and late abnormalities. But cytological examination of our material clearly shows that the frequency of the abnormalities is definitely higher (by a factor of more than two) after dorsal than after ventral injury of the cortex (Table 1).

Table 1.
graphic
graphic

The difference between D and V is very impressive in one particularly ‘good’ experiment in which the eggs withstood the operative trauma exceptionally well; it gave the results shown in Table 2.

Table 2.
graphic
graphic

It is clear that very few eggs, in this experiment, cleaved normally after dorsal injury, while almost all of them did after ventral injury. Furthermore, none of the D eggs which proceeded further than the late blastula or early gastrula stage developed normally; three of them hardly formed a nervous system (Figs. 9, 10). On the other hand, the majority of the V eggs formed perfectly normal neurulae.

2. Early cytological abnormalities

Mitotic abnormalities are extremely frequent in eggs which have been fixed between 45 min and 20 h after slight injury of the dorsal or the ventral cortex. These abnormalities affect both the mitotic apparatus (spindle and asters) and the chromosomes. There is no specific mitotic abnormality which could allow a distinction to be made between eggs injured on the dorsal or the ventral side : the only difference is of a quantitative nature and lies in the frequency of the abnormal mitoses. Cytological examination shows that the latter are often spread all over the eggs after dorsal injury and that they are, in general, localized in the cells which surround the wounded area after ventral injury: this difference explains, to some extent, the results described in Tables 1 and 2.

(a) Abnormalities of the mitotic apparatus

They are exceedingly varied and frequent, and can sometimes be detected as soon as 45 min after the operation : it is clear that slight injury of the cortex can exert quick and deep effects on the organization of the mitotic spindle. The more frequent abnormalities are the formation of cytasters, of pluricentric mitoses and the distortion or the cleavage of the spindle (Figs. 1–3). Anastra mitoses have occasionally been observed.

Fig. 1.

Abnormal formation of spindle and asters 45 min after injury of D cortex (Unna, × 1000).

Fig. 1.

Abnormal formation of spindle and asters 45 min after injury of D cortex (Unna, × 1000).

Fig. 2.

Cleavage of the spindle 5 h after injury of D cortex (Unna, × 1000).

Fig. 2.

Cleavage of the spindle 5 h after injury of D cortex (Unna, × 1000).

Fig. 3.

Pluricentric mitosis and cytasters 4 h after injury of V cortex (Unna, × 400).

Fig. 3.

Pluricentric mitosis and cytasters 4 h after injury of V cortex (Unna, × 400).

The alterations of the spindle formation often lead to an arrest of cytoplasmic division (plasmadieresis) in certain blastomeres in which nuclear division can proceed for several hours; this leads to the formation of large multinucleate cells and to partial and irregular cleavage.

No abnormality other than localized cytolysis could be observed after one single pricking of the egg with a microneedle.

The basophilic bodies which accumulate at the vegetal pole of the fertilized amphibian eggs and which are thought to be specific to the germ plasm (Czol-owska, 1969; Mahowald & Hennen, 1971) could often be seen after Unna staining: there was no conspicuous difference, regarding their size and number, between control and the injured eggs. In both cases these basophilic bodies remained unstained with the Feulgen reaction.

(b) Chromosomal abnormalities

The spindle and centrosome abnormalities have a serious impact on the behaviour of the nuclei and the chromosomes. Giant and irregular, sometimes polyploid, nuclei are often seen (Fig. 4). Some of them clearly result from nonfusion of the chromosomes at telophase; others, which stain strongly with Feulgen, derive from polyploid or endopolyploid mitoses (Fig. 5). The chromosomes of pluricentric or blocked mitosis progressively undergo pycnotic degeneration in the middle of the spindle (Fig. 6). Images of chromosome breakage or stickiness (formation of anaphase bridges) are also frequent (Figs. 7, 8). Chromatin bridges uniting two adjacent cells and binucleate cells have often been seen in blastulae (Figs. 1–8). Elimination of whole chromosomes or of chromosome fragments in the cytoplasm, nuclear budding and amitotic nuclear division are also common in blastulae arrested in their development 20 h after operation.

Fig. 4.

Giant, irregular nucleus, surrounded by small cytasters, 6 h after injury of D cortex (Unna, × 650).

Fig. 4.

Giant, irregular nucleus, surrounded by small cytasters, 6 h after injury of D cortex (Unna, × 650).

Fig. 5.

Polyploid mitosis, 4 h after injury of D cortex (Feulgen, ×1100).

Fig. 5.

Polyploid mitosis, 4 h after injury of D cortex (Feulgen, ×1100).

Fig. 6.

Pycnotic degeneration of the chromosomes 4 h after injury of D cortex (Feulgen, × 1000).

Fig. 6.

Pycnotic degeneration of the chromosomes 4 h after injury of D cortex (Feulgen, × 1000).

Fig. 7.

Chromatin bridge in a multipolar mitosis, 4 h after injury of V cortex (Feulgen, × 1000).

Fig. 7.

Chromatin bridge in a multipolar mitosis, 4 h after injury of V cortex (Feulgen, × 1000).

Fig. 8.

Chromatin bridge at anaphase, 5 h after injury of D cortex (Feulgen, × 1000).

Fig. 8.

Chromatin bridge at anaphase, 5 h after injury of D cortex (Feulgen, × 1000).

Fig. 9.

Almost complete absence of differentiation, 36 h after injury of D cortex (Unna, × 70).

Fig. 9.

Almost complete absence of differentiation, 36 h after injury of D cortex (Unna, × 70).

Fig. 10.

Degenerating medullary plate, but chorda is present 36 h after injury of D cortex (Unna, × 70).

Fig. 10.

Degenerating medullary plate, but chorda is present 36 h after injury of D cortex (Unna, × 70).

It can be concluded from these cytological observations that injury of the cortex often exerts a very unfavourable influence on the mitotic apparatus, and, by way of consequence, on the chromosomes : DNA itself finally undergoes degradation, as indicated by chromosomal stickiness and breakage, followed by pycnotic degradation.

3. Late abnormalities

(a) Morphogenesis

Exogastrulation is frequent in both the V and D embryos; it is probably the consequence of a lack of co-ordination in the gastrulation movements, directly or indirectly induced by cortical damage.

More interesting are the cases of reduction of the nervous system: in the three already mentioned extreme cases the embryos were almost anidian and lacked proper cephalo-caudal and dorsoventral organization (Figs 9, 10). More frequently, the nervous system is made of a very small number of cells only or is even missing, despite fairly normal organization of the underlying chorda and mesoderm (Fig. 11) : in such cases it is likely that the competence of the neuroblast to the inducing stimuli has been reduced as a consequence of the previous injury of the cortex. Such embryos are very similar to those previously described by one of us (Brachet, 1962) after mercaptoethanol and lipoic acid treatment of amphibian embryos.

Fig. 11.

Absence of medullary plate, 31 h after injury of D cortex (Unna, × 100).

Fig. 11.

Absence of medullary plate, 31 h after injury of D cortex (Unna, × 100).

In more advanced embryos (Fig. 12) basophilic degeneration and anarchic distribution of the cells can be found in the axial organs (nervous system and chorda) in the somites and on the dorsal side of the endoderm.

Fig. 12.

Abnormal axial organs, 47 h after injury of D cortex (Unna, ×200).

Fig. 12.

Abnormal axial organs, 47 h after injury of D cortex (Unna, ×200).

It should be noted that these late developmental abnormalities have also been found, but in very exceptional cases only, after injury of the ventral cortex. Since the position of the grey crescent is not always clearly recognizable in certain batches of eggs, an occasional error, at the time of the operation, cannot be excluded. In any case, the great majority of the eggs, which after injury of the ventral cortex, went through cleavage and gastrulation, underwent normal development afterwards.

(b) Cytology

A frequent histological observation is that the axial organs, at the neurula stage, are made of a mosaic of basophilic and yolk-laden cells : breakdown of the yolk platelets is slowed down in some individual cells, for no obvious reasons. The result is that the dorsoventral basophilic gradients are less regular and conspicuous in the operated eggs than in the controls. It is clear that individual cells can escape the influence of these gradients and that they apparently remain unaffected by the presence of neighbouring RNA-rich cells.

Mitotic abnormalities are less frequent than during earlier development, as one could have expected; but, as shown in Figs. 13–15, abnormal mitoses can still be found, even 36 h after operation. Giant nuclei, pluricentric mitoses (Fig. 13) sometimes with dispersed chromosomes (resembling c-mitoses), anastral mitoses, polyploid mitoses (Fig. 14) and multinucleate cells (Fig. 15) have been observed many times. Obviously the formation of a normal mitotic apparatus, after early cortical damage, remains a difficult problem for many cells at a much later stage (neurula).

Fig. 13.

Pluricentric mitosis, 30 h after injury of D cortex (Unna, × 1000).

Fig. 13.

Pluricentric mitosis, 30 h after injury of D cortex (Unna, × 1000).

Fig. 14.

Polyploid mitosis, 36 h after injury of D cortex (Unna, × 1000).

Fig. 14.

Polyploid mitosis, 36 h after injury of D cortex (Unna, × 1000).

Fig. 15.

Multinucleate cell, 36 h after injury of D cortex (Unna, × 1150).

Fig. 15.

Multinucleate cell, 36 h after injury of D cortex (Unna, × 1150).

Most often the mitoses are either very rare or even absent in the operated embryos; the arrest of DNA synthesis is probably one of the reasons why the operated embryos often lag behind the controls. Two days after operation abnormally large cells are found frequently, particularly in the nervous system.

The nucleoli are variable in size and basophily; in large nuclei which have stopped dividing, giant or multiple very basophilic nucleoli can always be found. In many embryos a mosaic of cells containing very large and inconspicuous nuclei has been observed.

The most obvious difference between operated and control embryos is the much greater variability in size and in intensity of Feulgen staining among the various nuclei of the previously injured eggs. This variability is particularly conspicuous in the three already mentioned almost anidian embryos of Figs. 9 and 10, which also show a very abnormal distribution of the cytoplasmic RNA: there are no basophily gradients but the embryos are made of a mosaic of strongly basophilic and lightly staining cells after Unna staining. The development, basophily and number of nucleoli in the individual cells are also completely anarchic in these embryos, where the mitoses were still frequent, but often abnormal and pluricentric.

(4) Cytophotometry

The above-reported cytochemical observations very strongly suggest that the embryos originating from injured eggs (especially after wounding of the dorsal cortex) display a large degree of aneuploidy. This possibility was checked by a cytophotometric analysis of 20 embryos, which had been fixed at the following times after the beginning of the experiment: 23–24 h (4 embryos); 26–29 h (4 embryos) ; 30–36 h (8 embryos) ; 43–48 h (4 embryos).

The overall results are given in Fig. 16 (a–c): each graph summarizes the results of more than 1000 measurements, made on six different embryos (a, control; b, ventral injury; c, dorsal injury). It is clear that the histograms for the controls and the V embryos are only slightly different; on the other hand, the histogram for the D embryos shows the presence of nuclei which contain either a lower or a very much higher amount of DNA than the controls. Obviously, aneuploidy is very marked in the D embryos and moderate in the V ones.

Fig. 16.

Results of cytophotometric analysis of Feulgen-stained embryos. Each histogram summarizes the results of more than 1000 measurements on six different embryos.

Fig. 16.

Results of cytophotometric analysis of Feulgen-stained embryos. Each histogram summarizes the results of more than 1000 measurements on six different embryos.

Fig. 17, which shows histograms for embryos from the same experiment, demonstrates that aneuploidy was indeed very marked in the two almost anidian embryos which originated from eggs wounded in their dorsal cortex.

Fig. 17.

Results of cytophotometric analysis of Feulgen-stained embryos.

Fig. 17.

Results of cytophotometric analysis of Feulgen-stained embryos.

As mentioned in the section on Methods, cytophotometric measurements of the DNA content of the individual nuclei were made separately on the ectoderm, the mesoderm and the endoderm: the results of this analysis lead to the conclusion that aneuploidy and polyploidy are more frequent in the mesoderm and in the endoderm than in the neurectoderm. This conclusion remains valid for the 2-day-old embryos, which had a closed neural tube. This difference between neurectoderm and mesentoderm is not surprising in view of the localization and prospective significance of the grey crescent in the fertilized egg.

That aneuploidy is more frequent after dorsal than after ventral injury is further shown by the following observation: the histograms were indistinguishable from those of the controls in 3 embryos out of 6 after ventral wounding and only in 1 out of 10 after dorsal injury.

(5) Autoradiography

(a) Uridine incorporation

The two experiments in which bisected gastrulae originating from normal or wounded eggs had been immersed in radioactive precursors failed to give clear answers : radioactivity was weak, irregular and patchy, as already reported by Bachvarova, Davidson, Allfrey & Mirsky (1966) and by Tencer (1971). It is likely that, in such experiments, the uptake of [3H]uridine varied greatly from cell to cell, in connexion with the degree of injury inflicted on the cells by the section of the embryos into two in addition to injury due to previous wounding of the cortex.

In the micro-injection experiments plain microscopical observation shows that labelling of the nuclei was insignificant when development was blocked at the blastula stage, whether it was the result of dorsal or ventral cortex injury. At the gastrula stage a clear-cut gradient of uridine incorporation in the nuclei becomes visible after micro-injection : the animal pole nuclei are much more radioactive than those of the vegetal one, while the nuclei present in the blastoporal lips seem to occupy an intermediate position. Labelling strongly increases in the nuclei while gastrulation and neural induction proceed.

Only 6 embryos out of 17 have been thoroughly analysed by counting the tracks present over the nuclei (3 D, 2 V, 1 control). Part of the slides have been treated with DNase, RNase, buffer alone and acid hydrolysis (6 min in 1 N-HCI at 60 °C) before counting. The results of the enzymic digestion were not very conclusive because the buffer alone took away too much of the radioactivity after the mild fixation methods used in this autoradiographic study.

Table 3 gives the results of the countings corrected for the background values. Counts were made over 100 nuclei at the animal pole, 50 nuclei in the dorsal lip of the blastoporus and 20 nuclei in the vegetal hemisphere. Values obtained before and after acid hydrolysis (which removes the RNAs) have been included in the table.

Table 3.

[3H]uridine incorporation: number of grains per nucleus after deduction of the background

[3H]uridine incorporation: number of grains per nucleus after deduction of the background
[3H]uridine incorporation: number of grains per nucleus after deduction of the background

The results are not numerous enough to draw definite conclusions but a few trends are indicated. For instance, in controls and in the V eggs, a clear animal-vegetal gradient in the radioactivity of the nuclei is found in the gastrulae; but there is no difference between the two poles after injury of the dorsal cortex (D). The lowering of the incorporation in the dorsal lip, after wounding the dorsal cortex, might be correlated with the frequent decrease in the development of the neural system when such embryos are allowed to develop.

Another feature of the results is the strong radioactivity of the neuroblast, which was undergoing neural induction in the only late gastrula which was analysed; this is in agreement with earlier autoradiographical and biochemical findings (Brahma, 1966; Tiedemann, 1968).

Finally, it is clear that the nuclei of the blocked blastulae were practically inactive in RNA synthesis, especially after the dorsal injury.

When one compares the values obtained before and after acid hydrolysis, the results are too variable to draw any other conclusion than that there was a tendency for preferential incorporation of uridine into RNA when there was a strong incorporation of the precursor in the nuclei.

(b) Thymidine incorporation

Nine embryos (4 D, 3 V, and 2 controls) have been analysed; digestion with DNase has clearly shown that the whole of the label has been incorporated into DNA. The data are summarized in Table 4.

Table 4.

[3H]thymidine incorporation: number of grains per nucleus after deduction of the background

[3H]thymidine incorporation: number of grains per nucleus after deduction of the background
[3H]thymidine incorporation: number of grains per nucleus after deduction of the background

Three embryos (1 D, 1 V and 1 control) were at almost exactly the same stage (young blastulae) ; those (D and V) whose cortex had been injured shortly after fertilization were already arrested in their development: the incorporation of thymidine was very much lower in these two blocked embryos than in the control. Furthermore, examination of the sections of these two operated embryos showed that they contained a mosaic of labelled and unlabelled nuclei. There was much less coordination of DNA synthesis in the various nuclei in these operated blastulae than in the control one. No definite trend can be detected when one compares the operated and control gastrulae, where DNA synthesis was still proceeding normally when the embryos were fixed.

(c) Leucine incorporation

Only four gastrulae (2 D, 1 V, 1 control) have been studied. The results are given in Table 5.

Table 5.
graphic
graphic

One will notice that the nuclei of the gastrula which had developed after injury of the ventral cortex (V) are less radioactive than those of the blocked blastula, which had been wounded on the dorsal side. The latter (in contrast to what we have seen for uridine incorporation in Table 4) has rather strongly radioactive nuclei; this is not surprising since it is known that protein synthesis is possible in amphibian eggs in which RNA synthesis has been blocked by micro-injection of actinomycin D (Brachet, Denis & de Vitry, 1964).

There is another difference between RNA and protein synthesis in these eggs: the nuclei of the vegetal hemisphere are very radioactive after leucine injection, while they contain little radioactive material after uridine incorporation. It would be unwise to draw conclusions from this observation about the existence or absence of an animal-vegetal gradient for nuclear protein synthesis, since recent work of Gurdon & Woodland (1970) has clearly shown that a large proportion, if not all, of the nuclear proteins are of cytoplasmic origin. It might well be that the large nuclei of the vegetal pole incorporate a larger quantity of cytoplasmic proteins than the smaller nuclei of the upper hemisphere of the gastrula.

The fact that slight injury of the cortex is often followed by mitotic abnormalities, resulting in more or less pronounced aneuploidy, is a matter of concern for the interpretation of some classical experiments : for instance, it is no longer possible to conclude that experiments such as those of Curtis (1960), in which only the cortex apparently had been injured, demonstrate that, in early stages of development, morphogenesis is controlled by purely cytoplasmic factors. The same word of caution is also valid for experiments in which the grey crescent region was destroyed by deep pricking (A. Brachet, 1906; Pasteels, 1932): in the few experiments in which we destroyed the grey crescent of the Xenopus eggs by repeated deep prickings with the microneedle, the cleaving eggs showed the same kind of cytological abnormalities as in those where only the cortex had been superficially injured according to the technique of Curtis (1965).

Since even superficial injuries to the egg cortex can exert deep effects on the nuclei, especially during cleavage stages, one should also be careful in the interpretation of experiments in which radioactive precursors of nucleic acid and protein synthesis, or metabolic inhibitors such as actinomycin and puromycin (Brachet et al. 1964; Legros & Brachet, 1965; Legros, 1970) have been micro-injected into amphibian eggs. However, it would be going much too far to conclude that such experiments, which remain the basis of much of our present knowledge on the molecular bases of amphibian development, are meaningless. But it would be unwise to forget that micro-injection of substances into the eggs can produce, in an unspecific way, serious mitotic and nuclear abnormalities. In fact, such abnormalities have repeatedly been found after micro-injection of tissue extracts (Markert & Ursprung, 1963; Melton, 1965) and isolated nuclei in amphibian eggs (Hennen, 1970; Briggs & King, 1957; Signoret & Picheral, 1962; Di Berardino & Hoffner, 1970, etc.). Several of the early chromosomal abnormalities we encountered in the present study are very similar to those recently described by Di Berardino & Hoffner (1970) a few hours (2–8 h) after the transplantation of nuclei in anucleate unfertilized eggs of Rana pipiens.

Our cytological observations do not, however, invalidate the main conclusions of A. Brachet (1906) and Pasteels (1932), since they studied embryos fixed many hours after the destruction of the grey crescent : by that time the embryos which had shown irregular cleavage (as a result of the formation of an abnormal mitotic apparatus) or arrest at the blastula stage (which is classical in aneuploid amphibian embryos, as shown by Fankhauser, 1934, and Dalcq, 1968) had been discarded. Our own results confirm that developmental abnormalities (both early and late) are at least twice as frequent after dorsal as after ventral injury of fertilized eggs. However, one of the almost anidian embryos described in this study was very strongly aneuploid and nearly all the embryos which originated from dorsally wounded eggs showed some aneuploidy by cytophotometric analysis; it is difficult to rule out a nuclear, genetic effect as the cause of poor development.

In the more advanced embryos (neurulae), mitotic abnormalities became very few in number when development was apparently normal: there was always a good parallelism between aneuploidy and failure for the embryo to develop normally. It is quite conceivable that aneuploid cells are eliminated by selection when development proceeds further in a normal way : this is indicated by the fact that the frequency of the pycnoses was much higher in the pricked embryos which have been analysed by cytophotometry than in the controls (69 after dorsal pricking, 67 after ventral pricking, 4 in the controls for 1000 nuclei). The frequency of the mitoses, ranging from 50 to 54 for 1000 nuclei, was the same in the operated and control embryos of this series.

The importance of the egg cortex, emphasized by Curtis (1960), has been recently challenged by Nieuwkoop (1969). Our own experiments confirm that there is a difference between the dorsal and the ventral cortex; but we think that, for the already mentioned reasons, one should be very cautious in their interpretation. For instance, when Curtis (1960) obtained a block in development at the blastula stage after removal of a piece of the dorsal cortex, this result might have been the consequence of aneuploidy rather than of direct control of morphogenesis by the dorsal cortex : only a cytological analysis or, better, a quantitative analysis of the nuclei, by cytophotometry, could allow a choice to be made between the two alternatives. The few cases where Curtis (1960) obtained double embryos after grafting dorsal cortical material on the ventral side of fertilized eggs speak in favour of a direct effect of the cortex on morphogenesis; but it is unfortunate that these unusually interesting eggs have not been submitted to a thorough cytological analysis. Recent experiments by one of us (Brachet, 1969) on sea-urchin eggs indicate that cortical material exerts, in these eggs, an inhibitory effect on development. Such a finding would agree with the view expressed by Dollander (1962) and by Lovtrup (1965) that the dorsal cortex, in amphibian eggs, might be less developed than the ventral cortex: this would lead to increased permeability and metabolism on the dorsal side (see reviews by Dollander (1962), Lovtrup (1965) and Pasteels (1964)) on the formation and the properties of the cortex in amphibian eggs.

Our main conclusion, i.e. that slight injury of the cortex often produces aneuploidy which remains detectable a couple of days after the operation, throws some doubt on the conclusions of the genetic experiments of Curtis (1965) : it will now be very difficult to prove that the lethality he has observed in the offspring of frogs originating from eggs which had suffered injury in their cortex soon after fertilization is due to a ‘cortical heredity’ rather than to chromosomal defects. Proof of the existence of cortical heredity will be particularly difficult to give since, as we have seen, aneuploidy is especially frequent in the mesoderm and in the endoderm: slight chromosomal damage of the future gonocytes is a possibility which cannot be easily dismissed. In experiments such as those of Curtis (1965) and those of Buehr & Blackler (1970), on the formation of the gonocytes in amphibian eggs, controls in which the eggs are injured in other areas than those which are thought to be of morphogenetic importance (grey crescent, vegetal pole) are required before safe conclusions can be drawn.

It should be pointed out, however, that the eggs used in our experiments, probably because of a greater fragility than for technical reasons, gave a much lower percentage of normally gastrulating and neurulating embryos (42/88 normal embryos) than in those of Curtis (1965), who obtained 194 normal embryos out of 200. For this reason we do not think that our experiments disprove the theory of cortical heredity proposed by Curtis (1965): they merely show that it will be very difficult to prove it. Even if the embryos which finally become adults are free of gross aneuploidy, one cannot exclude the possibility that minor chromosomal damage (deletions, inversions, etc.) has persisted. If this happened in the gonads, important lethality in the offspring could of course ensue. The methods used in the present paper can only detect gross cytological abnormalities; in the absence of a cytogenetic analysis of the larvae and adults, all one can say is that the results of Curtis (1965) might be due to chromosomal damage after pricking the eggs.

Our own autoradiography experiments are not numerous enough to allow us to draw any serious conclusions: in fact, our work in this direction was interrupted when it became clear that, due to the frequent aneuploidy, a more extensive and time-consuming analysis would be worthless. It is clearly not possible to test the hypothesis put forward by one of us (Brachet, 1967), i.e. that the dorsal cortex controls nuclear RNA synthesis in a system where injury of the cortex induces changes in the nuclei : as already pointed out, the anidian embryos described in this paper are probably the result of both aneuploidy and cortical damage to the grey crescent. All we can say is that destruction of the dorsal cortex probably reduces RNA synthesis in the nuclei of the dorsal lip of the blastoporus and that there is, as one would expect, a close correlation between morphogenesis and nuclear RNA synthesis.

Recherches sur les interactions nucléocytoplasmiques au cours des stades précoces du développement embryonnaire des Amphibiens

I. Destructions localisées du cortex de l’œuf

Le cortex d’œufs de Xénope, au stade indivis ou deux blastomeres, a été légèrement lésé au moyen d’une fine aiguille. Les lésions ont porté sur le cortex dorsal et le cortex ventral. 184 œufs opérés ont été examinés cytologiquement.

La fréquence des anomalies précoces (segmentation) et tardives (gastrulation, neurulation) est à peu près double après lésion du cortex dorsal qu’après blessure du cortex ventral.

Des anomalies de l’appareil mitotique et des chromosomes sont très fréquentes pendant la segmentation, après de légères lésions du cortex.

Aux stades plus avancés, les embryons dont le cortex avait été lésé précédemment présentent souvent de l’aneuploïdie, comme l’ont montré des mesures cytophotométriques de la teneur en ADN des noyaux. Cette aneuploïdie tardive est beacoup plus fréquente après lésion du cortex dorsal qu’après blessure du cortex ventral.

Des expériences autoradiographiques ont montré qu’il existe une corrélation étroite entre la morphogénèse et la synthèse de RNA par les noyaux.

Les résultats sont discutés dans le cadre de nos connaissances actuelles sur le rôle morphogénétique et génétique du cortex et du croissant gris.

We wish to thank Professor A. Curtis and Dr R. Tencer for initiating us to the micro-surgical technique used in this work. We are grateful to Dr P. Malpoix for revising the English, to Mrs N. Hulin for the cytophotometric measurements and to Mrs E. De Saedeleer for the preparation of the microphotographs.

Bachvarova
,
R.
,
Davidson
,
E. H.
,
Allfrey
,
V. G.
&
Mirsky
,
A. E.
(
1966
).
Activation of RNA synthesis associated with gastrulation
.
Proc. natn. Acad. Sci. U.S.A.
55
,
358
365
.
Brachet
,
A.
(
1906
).
Recherches expérimentales sur l’œuf non segmenté de Rana fusca.
Wilhelm Roux Arch. EntwMech. Org.
22
,
325
341
.
Brachet
,
A.
(
1931
).
L’Oeuf et les facteurs de l’ontogénèse,
Paris
:
Doin et fils
.
Brachet
,
J.
(
1962
).
Effects of β-mercaptoethanol and lipoïc acid on morphogenesis
.
Nature, Lond.
193
,
87
88
.
Bracket
,
J.
(
1967
).
De l’Embryologie Expérimentale à la Biologie Moléculaire
(ed.
E.
Wolff
).
Paris
:
Dunod
.
Brachet
,
J.
(
1969
).
Présence dans l’œuf d’oursin, d’inhibiteurs diffusibles du développement
.
C. r. hebd. Séanc. Acad. Sci., Paris
268
,
1768
1770
.
Brachet
J.
,
Denis
,
H.
&
De Vitry
,
F.
(
1964
).
The effects of actinomycin D and puromycin on morphogenesis in amphibian eggs and Acetabularia mediterránea
.
Devl Biol.
9
,
398
434
.
Brahma
,
S. K.
(
1966
).
An autoradiographic study of RNA synthesis during primary embryonic induction
.
J. Embryol. exp. Morph.
16
,
203
210
.
Briggs
,
R. W.
&
King
,
T. S.
(
1957
).
Changes in the nuclei of differentiating ectoderm cells as revealed by nuclear transplantation
.
J. Morph.
10
,
269
312
.
Buehr
,
M. L.
&
Blackler
,
A. W.
(
1970
).
Sterility and partial sterility in the South African clawed toad following the pricking of the egg
.
J. Embryol. exp. Morph.
13
,
375
384
.
Curtis
,
A. S. G.
(
1960
).
Cortical grafting in Xenopus laevis.
J. Embryol. exp. Morph.
8
,
163
173
.
Curtis
,
A. S. G.
(
1965
).
Cortical inheritance in the amphibian Xenopus laevis: preliminary results
.
Archs Biol., Liège
76
,
523
546
.
Czolowska
,
R.
(
1969
).
Observations on the origin of the germinal cytoplasm in Xenopus laevis.
J. Embryol. exp. Morph.
22
,
229
251
.
Dalcq
,
A.
(
1968
).
Sur la production d’une lèvre blastoporale chez les Amphibiens
.
Acta biotheor.
18
,
330
337
.
Dalcq
,
A.
&
Pasteels
,
J.
(
1937
).
Une conception nouvelle des bases physiologiques de la morphogénèse
.
Archs Biol., Liège
48
,
669
710
.
Davidson
,
E. H.
,
Haslett
,
G. M.
,
Finney
,
R. J.
,
Allfrey
,
V. G.
&
Mirsky
,
A. E.
(
1965
).
Evidence for prelocalization of cytoplasmic factors affecting gene activation in early embryogenesis
.
Proc. natn. Acad. Sci. U.S.A.
54
,
696
704
.
Di Berardino
,
M. A.
&
Hoffner
,
N.
(
1970
).
Origin of chromosomal abnormalities in nuclear transplants. A revaluation of nuclear differentiation on nuclear equivalence in amphibians
.
Devl Biol.
23
,
185
209
.
Dollander
,
A.
(
1962
).
Conceptions actuelles et terminologie relatives à certains aspects de l’organisation corticale de l’œuf d’Amphibien
.
Archs Anat. Histol. Embryol.
44
,
93
103
.
Fankhauser
,
G.
(
1934
).
Cytological studies on egg fragments of the salamander Tritton
.
J. exp. Zool.
67
,
349
.
Ficq
,
A.
(
1959
).
In
Autoradiography
(ed.
J.
Brachet
and A. Mirsky), chap. 3 (
The
Cell
), pp.
67
90
. New York and
London
:
Academic Press
.
Gurdon
,
J. B.
&
Woodland
,
H. R.
(
1970
).
On the long-term control of nuclear activity during cell differentiation
.
Current Topics Devl Biol.
5
,
39
70
.
Hennen
,
S.
(
1970
).
Influence of spermine and reduced temperature on the ability of transplanted nuclei to promote normal development in eggs of Rana pipiens.
Proc. natn. Acad. Sci. U.S.A.
66
,
630
637
.
Legros
,
F.
(
1970
).
Effets de divers inhibiteurs de la synthèse des protéines sur le développement des œufs d’Amphibiens
.
Archs Biol., Liège
81
,
109
137
.
Legros
,
F.
&
Brachet
,
J.
(
1965
).
Effets de la puromycine sur la mitose, la synthèse des protéines et celle du DNA, au cours de la segmentation de l’œuf de Batracien
.
J. Embryol. exp. Morph.
13
,
195
206
.
Løvtrup
,
S.
(
1965
).
Morphogenesis in the Amphibian embryo. Determination of the site of invagination
.
Acta zool., Stockh.
46
,
119
165
.
Mahowald
,
A. P.
&
Hennen
,
S.
(
1971
).
Ultrastructure of the ‘germ plasm’ in egg, and embryos of Rana pipiens.
Devl Biol.
24
,
37
53
.
Markert
,
L. L.
&
Ursprung
,
H.
(
1963
).
Production of replicate persistent changes in zygote chromosomes of Rana pipiens by injected proteins from adult liver nuclei
.
Devl Biol.
7
,
560
577
.
Melton
,
C. G.
(
1965
).
Blastular arrest and partial cleavage of frog embryos injected with actinomycin D, mitomycin C and homologous liver macromolecules
.
Devl Biol.
12
,
287
308
.
Nieuwkoop
,
P. D.
(
1969
).
The formation of the mesoderm in urodelean amphibiens
.
Wilhelm Roux Arch. EntwMech. Org.
163
,
298
315
.
Nieuwkoop
,
P. D.
&
Faber
,
J.
(
1967
).
Normal Table of Xenopus laevis (Daudin).
Amsterdam
:
North Holland Publishing Co
.
Pasteels
,
J.
(
1932
).
Expériences de piqûres localisées sur l’œuf non segmenté des Anoures
.
C. r. Assoc. Anat. de Nancy 1932
.
Pasteels
,
J.
(
1964
).
The morphogenetic role of the cortex of the amphibian egg
.
Adv. Morphog.
3
,
363
388
.
Signoret
,
J.
&
Picheral
,
B.
(
1962
).
Transplantation de noyaux chez Pleurodeles waltlii.
C. r. Acad. Sci. Belg.
254
,
1150
1151
.
Tencer
,
R.
(
1971
).
Quelques observations sur les hybrides liteaux
.
Archs. Biol., Liège. (In the Press
.)
Tiedemann
,
H.
(
1968
).
Factors determining embryonic differentiation
.
J. cell. Physiol.
72
(
suppl. 1
),
129
144
.