ABSTRACT
The rate of incorporation of glycine-l-C14 into the tissues of X enopus tailbud embryos was estimated by counting grains in autoradiographs. The tissues differed significantly, the order of activity being epidermis > somitic mesoderm > neural tube > notochord > endoderm.
Less precise data indicated that the incorporation both of glycine-l-C14 and methionine-S35 is more rapid into the nuclei than the cytoplasm, and that there may be some differential accumulation also in the cell membranes.
The rate of accumulation of glycine-2-C14 in acid- and acetone-insoluble material, estimated by the Geiger counter, is reduced in the presence of the amino-acid analogue p-fluoro-phenyl-alanine. This effect is counteracted by L-phenyl-alanine.
INTRODUCTION
The production of different cell types during embryonic development must depend to a major extent on the synthesis of specific substances, or mixtures of substances, characterizing the various tissues. It seems probable that such differential syntheses begin at an early stage, probably very soon after determination occurs. Up to the present rather little is known about them. The earlier biochemical work on young embryos, summarized in such works as those of Needham (1942) and Brachet (1944), dealt largely with the respiratory or glycolytic metabolism and threw only indirect light on the formation of the substances which cause developing cells to become differentiated. In the last few years a more direct attack on the problem has been made by such workers as Brachet (1952 and previously) in his studies on the appearance and distribution of nucleic acids and sulphydryl-containing proteins, and Ebert (1952), Clayton (1953), and others who have used immunological methods to detect newly synthesized substances.
One method of investigating the processes of protein metabolism in early embryos is to study the rate of incorporation of isotope-labelled amino-acids. A considerable amount of work of this kind has been carried out on echinoderm eggs (cf. Hultin, 1953), the isotopes used being N15 and C14, and the incorporation from inorganic compounds (acetate and ammonium salts) being studied as well as that from amino-acids. In this, and a subsequent paper by Feldman & Waddington (1955), preliminary reports are given of investigations on two types of vertebrates, namely amphibia and birds. The isotopes employed were C14 and S35, made available as glycine-l-C14 and glycine-2-C14 and methionine-S35.
We have been particularly interested to search for evidence of differential incorporation into different tissues, since this would indicate the relevance of protein synthesis to the early phases of differentiation. This problem has been previously studied by Friedberg & Eakin (1949), Eakin, Kutsky, & Berg (1951), and Flickinger (1954). We have, in some experiments, used an autoradiographic method of assessing the concentration of radioactive amino-acid in the various regions of the embryo. Some attention has also been paid to the effect, on the incorporation of these amino-acids, of certain amino-acid analogues, the two most fully studied being p-fluoro-phenyl-alanine (for a sample of which we are very grateful to Professor S. Spiegelman) and ethionine.
A short account of the work was given at a meeting held in London in January 1954 under the auspices of the Journal of Embryology and Experimental Morphology (cf. Nature, 1954,173,517-20). At the same meeting Brachet described work on the incorporation of labelled amino-acids into amphibian embryos carried out in his laboratory by Ficq (Ficq, 1954).
METHODS
Axolotl and Xenopus embryos were placed in small capsules containing the solutions of the labelled amino-acids or other substances and allowed to develop at 18° C. The bottom of the capsule was covered with a layer of 2 per cent, agar to prevent adhesion of the embryos to the glass. The vitelline membranes were removed. Friedberg & Eakin (1949) found that the diffusion of amino-acids was impeded by the surface coat, and therefore in blastulae or gastrulae destined for autoradiography either the blastocoel or the archenteron were slit open.All later stages, and all embryos used for Geiger counter estimations, were cut in halves; this caused some morphological abnormalities in the development of embryos transected at early stages, but it is unlikely that these changes would alter the chemical processes with which the experiments were concerned.
The chemicals were dissolved in one-fifth strength Holtfreter solution. To preserve asepsis, addition was made of 0·02 per cent, sodium sulphadiazine (May & Baker), which is harmless to the embryos and further helps to keep the pH near 6·9 without the presence of buffer. All materials and solutions were sterile, and aseptic precautions were taken throughout. Only embryos of normal appearance after decapsulation were employed in the experiments.
Some preliminary autoradiographs were made with Axolotl embryos. These were fixed in 10 per cent, trichloracetic acid, embedded in paraffin, and made into very thick sections (50μ). They were coated with Kodak stripping film and given an exposure of 15 days. This was found to result in a blackening of the film which was much too intense for quantitative work, although one or two useful qualitative points emerged.
In later work, with Xenopus embryos, the fixation was in Carnoy’s fluid and sections were cut at 8μ. After dewaxing the sections were coated with from one to three layers of gelatin by being plunged into a 0·5 per cent, solution and then allowed to dry. By weighing coverslips of known area treated in a similar way it was found that the thickness of each gelatin layer was about 0·1μ. This is too small to impair the resolution of the autoradiograph (Doniach & Pele, 1950), and it was thought that the treatment would make it easier to discriminate between grains in the photographic emulsion and pigment granules in the cells. All the autoradiographs used for the grain-counting mentioned in this paper have been treated in this way. Experience has shown, however, that in autoradiographs stained through the film after development and mounted in balsam, no difficulty arises in making this distinction even without the aid of a gelatin layer, and the use of such layers has therefore been discontinued.
The number of grains appearing over various regions of the sections was counted under an oil objective, the area covered for each count being three adjacent squares of a graticule, each covering 54μ2. Areas to be counted were chosen some distance apart, so that there was no question of mutual overlap by the spreading of the radiations; the resolution in the autoradiograph is about 3μ.
For activity measurements in the Geiger counter the embryos after removal from the glycine-C14 were rinsed twice in saline with a similar concentration of non-labelled glycine. They were then fixed in 10 per cent, trichloracetic acid, and a precipitate, mainly composed of protein, was obtained by the method of Friedberg & Eakin (1949). The precipitate was spread on several small aluminium planchettes, air-dried overnight under a 60-watt lamp, and the activity measured with a thin windowed Geiger counter. Counting was timed so as to ensure a standard deviation less than 5 per cent., and the rates were corrected for self-absorption of β-particles. Finally the weight of the precipitate was obtained.
RESULTS
1. Autoradiographs
For the first set of autoradiographs, cut at 50μ, Axolotl embryos of various stages between the two-cell and mid tail-bud stages were kept for periods of from 1 to 3 days either with glycine-1-C14 or with methionine-S35. As mentioned above, the autoradiographs were too dark for detailed quantitative examination. Visual estimation of the degree of blackening, however, gave a clear indication that, at least during stages up to the early tail-bud, there is a preferential incorporation of the amino-acids into the nuclei as contrasted with the cytoplasm. There was also probably some preferential incorporation into the cell membranes, but the evidence for this was not quite so unequivocal. Both amino-acids behaved rather similarly in these respects, although the tendency of methionine to be accumulated in the nucleus was perhaps rather less marked than that of glycine. Further experiments on the distribution of labelled amino-acids within the cell are in progress.
The incorporation into different regions of the embryo was investigated in three anterior halves of Xenopus embryos of rotation to gill-bud stages (Weisz’s stages 14–8), which were kept for 21 hours with glycine-1-C14 (activity 16·6 μc./c.c.). The counts in the autoradiographs are given in Table 1. It will be seen that the overall grain densities correspond with the exposure times. In any one embryo the differences of the mean counts in the various tissues are statistically significant, except for the neural tube-notochord comparison in embryo 4L and the epidermis-mesoderm in 4P. In the latter case, however, the epidermis may be underestimated, since the grain density is near the upper limit that can be accurately assessed in this way.
It will be seen that, in all embryos, the serial order of the tissues in tracer uptake is the same, except for the non-significant difference of neural tube and notochord in 4L. Epidermis always shows the greatest uptake. It cannot be excluded that this is partly due to the greater availability of the amino-acid to this tissue, but differences in diffusion path cannot be held responsible for all the variations found; it could not account, for instance, for the lower uptake in neural tube as compared with somitic mesoderm, or for the very low uptake in the endoderm.
It will be noticed that, although the order of the tissues remains the same in the three embryos, the ratios between them alter rather considerably. Further study will be necessary before the significance of this can be properly assessed, but it seems likely that it is related to differences in the age of the specimens. As judged by the development of the eye-cups in the sections, the developmental stage increased from 4L through 4V to 4P. On this basis one might advance the tentative hypothesis that the rate of incorporation in the axial mesoderm and neural tube increases in later stages in comparison with that into the epidermis, while that into the notochord remains low, perhaps as a consequence of the swelling of the cells of this organ with a relatively watery sap.
2. Effect of an amino-acid analogue on incorporation
Geiger counter measurements of incorporation into parts of the embryo have not been made, but the instrument has been used to investigate the incorporation into whole embryos and the effect on this of ‘the antagonistic amino-acid analogue p-fluoro-phenyl-alanine (FPA). The upper part of Table 2 gives the results of two replicated experiments. In experiment Pl7, 20 transected late blastulae of Xenopus were kept at 18° C. for 18 hours in a solution of glycine-2-C14 (activity 2·7 μc./c.c.) and the same number in a similar solution with FPA added, the molecular ratio of glycine to FPA being 1:30. In experiment Pl8 conditions were similar, except that 30 embryos were used in each solution. At the end of the experiment the embryos appeared perfectly healthy, there being no obvious effect of the amino-acids on their morphological development. The two experiments should not be compared with one another; but comparison of the two halves of each experiment provides definite evidence that the analogue has caused a reduction in the rate of uptake of the glycine.
In another experiment, No. Pl9, the specificity of the inhibition by FPA was tested by seeing whether it could be alleviated by L-phenyl alanine (PA), which is the structurally related normal amino-acid. In this experiment 20 transected late blastulae of Xenopus were allowed to develop for 24 hours at 18° C. in a solution containing glycine-2-C14 and FPA in the molecular ratio 1:73, the activity of the solution being 116μc./c.c.; and another 20 embryos were kept in a similar solution with added PA, the ratio of PA to FPA being 1:2. The activity of the proteins is given in the lower part of Table 2. It will be seen that the uptake in the solution with added PA is considerably higher (113·5 per cent.) than that with FPA alone.
No effects of the FPA on morphogenesis were noted, nor did it cause any obvious slowing-down of development, but no exact studies on development rate have yet been made.
DISCUSSION
The results reported in this paper are of a preliminary nature and do more to reveal the magnitude of the field of investigation opened up by these methods than to provide secure material for an extended discussion.
It is to be expected that glycine will be incorporated into the nucleic acids of the embryo as well as the proteins. Its partition between these two groups of substances is at present under investigation. The results reported in this paper do not make possible any discrimination between them, since the histological procedures used in preparing the autoradiographs would have removed the fats and free non-protein substances, but would leave the greater part of both proteins and nucleic acids.
There are, however, some points which emerge clearly even at this early stage. One of these is the preferential incorporation of glycine and methionine into the nuclei in early stages of development. The same effect has been noticed by Ficq (1954) using a different technique of autoradiography. Ficq showed that there is a considerable decrease in activity after digestion with ribonuclease or after hydrolysis with HC1, so that presumably much, though by no means all, of the glycine in the nuclei is incorporated into nucleic acid. Even so, the high activity of the nuclei is somewhat surprising, since it is usually considered that protein synthesis occurs mainly in the cytoplasm rather than the nuclei.
It is, of course, not obvious at first sight that the incorporation of labelled amino-acids into the material precipitable with trichloracetic acid is a direct indication of the synthesis of proteins; it might merely be a consequence of exchanges between the free amino-acids and those present in already formed protein. The latter possibility is, however, rendered less likely by the observation that the rate of incorporation of glycine is reduced in the presence of a structural analogue of an unrelated amino-acid, namely FPA. The simplest explanation of this is the hypothesis that proteins are synthesized in the embryo from a pool of free amino-acids, and that an abnormal analogue inhibits the whole process of protein formation, and thus lowers the rate of incorporation even of structurally unrelated amino-acids. It is more difficult to see how the exchange-rates of unrelated amino-acids could be affected.
In considering the effects of added FPA and PA it is important to remember that considerable quantities of the normal amino-acids are likely to be made available to the embryo by the breakdown of the yolk granules which are already inside the cells. This process is already becoming visible in microscopical preparations by the gastrulation stage. The presence of free phenyl alanine in the amphibian egg has been made probable by Holtfreter, Koszalka, & Miller (1950), who, however, were unable, with their technique, to distinguish it from leucine; and it was confirmed by Eakin, Berg, & Kutsky (1950), whose methods made this distinction possible. Kutsky, Eakin, Berg, & Kavanau (1953) have produced evidence of the existence of glycine, probably largely in the form of peptide, in amphibian embryos.
The results of the autoradiographs make it clear that, whether or not the incorporation of glycine is an indication of protein synthesis, the rate at which it occurs certainly differs considerably from tissue to tissue. Friedberg & Eakin (1949) and Eakin, Kutsky, & Berg (1951) had already found that the incorporation of labelled glycine proceeds more rapidly in the dorsal than in the ventral halves of frog gastrulae, but their method did not allow the discrimination of individual tissues. Our results show that the greater part of this difference should be attributed to the high uptake of the somite mesoderm and, to a lesser extent, to that of the neural tube. Flickinger (1954) tested the incorporation of C14O2 into a number of different parts of the gastrula, and into the axial system (neural plate, notochord, and somitic mesoderm) as compared with the remainder of the neurula (epidermis and endoderm). The latest stages he studied were at a considerably younger developmental age than the embryos which we have so far investigated by autoradiographs. Our results agree with his, however, in showing that there are marked differences between different regions.
These differences are not always of the kind which previous results would lead one to expect. The low activity of the notochord is perhaps explained by the fact that, shortly after the stages investigated, it becomes distended with a fluid cell sap. But the high activity of the epidermis and low activity of the neural tube are rather unexpected. There are indications, however, that the rates of activity are changing with developmental age even within the narrow range of the specimens described here, and it may be that considerable differences in tissue activity will be found at other stages. The autoradiographic technique has the advantage that it facilitates the assessment of the activity of rather precisely defined parts of the embryo without calling for the very laborious process of dissecting sufficient material to obtain a reading on a Geiger counter. It must, however, be admitted that the counting of grains on autoradiographs is itself a rather tiring and time-consuming occupation. On the other hand, one is limited to some extent in the fractionation procedures which can be employed, although much more might be done in this way than we have yet attempted. Further, the autoradiographs only enable one to assess activity per unit volume of tissue. It is notoriously difficult in the biochemistry of amphibian embryos to know what is the best measure of the metabolizing system, and how to allow for the presence of variable quantities of inactive yolk (cf. Needham, 1942; Boell, 1948). It may be one of the defects of the autoradiographic technique that it is rather inflexible in this respect.
ACKNOWLEDGEMENTS
While carrying out this work the junior author held a scholarship from the British Council, which he acknowledges with gratitude. The work received financial support from the Agricultural Research Council. We should also like to express our gratitude to Dr. B. Woolf for statistical assistance.