The cytoplasm of the egg, and particularly its cortex, starts the chain of reactions which eventually leads to the differentiation of parts of the embryo, it should not be assumed that the structure of the future embryo is rigidly determined by local differences in the cytoplasm of the egg. The local peculiarities of the egg cytoplasm are only some of the factors necessary for the formation of organ rudiments. That this is so can be shown by examining the development of the sea urchin, for instance.
In the 16-cell stage of the sea urchin, the blastomeres are of three different sizes. The animal hemisphere consists of eight blastomeres of medium size, the mesomeres, which are destined to produce most of the ectoderm of the larva. The vegetal hemisphere consists of four very large blastomeres, the macromeres, and of four very small blastomeres, the micromeres, which lie at the vegetal pole of the egg.
The macromeres also contain some material for the ectoderm and all the material for the endoderm. In a subsequent cleavage the future ectoderm and endoderm are segregated from each other into an upper tier of macromeres and a lower tier of macromeres. The upper tier, lying immediately under the equator of the embryo, contains the ectodermal material; the lower tier, lying nearer to the vegetal pole, contains the endodermal material.
The micromeres develop into mesenchyme, which later produces the larval skeleton consisting of calcareous spicules. In the sea urchin Paracentrotus lividus, the cytoplasm of the macromeres possesses a surface layer of red pigment granules, making the macromeres easily distinguishable from the other cleavage cells. The red pigment is already present in the egg at the beginning of cleavage as a broad subequatorial zone.
In the blastula stage, the cytoplasmic substances are found in the same arrangement as at the beginning of cleavage. Subsequently, the descendants of the micromeres migrate into the blastocoele where they develop the skeletal spicules; the descendants of the lower tier of macromeres invaginate, forming a pocket-like cavity—the gut; and the ectoderm produces a ciliary band, serving for locomotion, and a tuft of rigid cilia at the former animal pole.
The ectoderm also sinks inward to produce a stomodeum, coming into communication with the endodermal gut at its anterior end, while the original opening of the pocket-like invagination (the blastopore) becomes the anal opening.
The embryo of the sea urchin develops into a larva called a pluteus. If the blastomeres of the sea urchin are separated in the two-cell stage or in the four-cell stage, each of them develops into a complete pluteus of diminished size. This may be related to the fact that the first two cleavage planes are meridional, passing through the main axis of the egg.
All of the first four blastomeres therefore get equal portions of the three cytoplasmic regions of the egg (ectodermal, endodermal, and mesenchymal). A different result is observed if the egg is separated into the animal and vegetal halves after the third cleavage, the third cleavage plane being equatorial. In this case both halves produce, as a rule, defective embryos. The animal half differentiates as an ectodermal vesicle; it does not produce a gut.
Even purely ectodermal structures are abnormally differentiated- the ciliary band is not developed, whereas the tuft of long cilia on the animal pole develops excessively, growing over a much greater surface than it normally does. The vegetal half is differentiated into an ovoid embryo with a disproportionately large endodermal gut but without a mouth. There may be a few irregular skeletal spicules but no arms and no ciliary band.
In later cleavage stages, it is possible to cut the morula transversely below the equator. In such cases the vegetal part develops still more abnormally; it produces a large endodermal gut and a small ectodermal vesicle. The gut does not lie inside the ectodermal vesicle but is evaginated to the exterior (turned inside out), owing to interference with the normal processes of gastrulation. This phenomenon is known as exogastrulation. The mesenchyme cells migrate into the interior, but they usually produce no skeletal spicules or only very small and abnormal ones.
The result of this experiment is obviously due to each of the two halves lacking some parts contained in the other half. However, each half does not simply produce what would normally have been the fate of the respective part. Instead, the differentiation of each half seems to be “exaggerated” as compared with its normal fate. This is especially clear in the case of the increased animal tuft of cilia.
The same “exaggeration” of the ectodermal or endodermal differentiations can be achieved by exposing the developing eggs to certain chemical substances, even without removing any parts of the egg or embryo. If the fertilized sea urchin egg is exposed to seawater containing some lithium salts in solution, the embryo develops just as if it were only the isolated vegetal half.
The gut is increased and tends to exogastrulate instead of invaginating toward the interior. The skeleton is absent or abnormal, and the ectoderm is represented by an epithelial vesicle and fails to differentiate further. The increase in the size of the gut occurs at the expense of the ectoderm, and in extreme cases, most of the embryo differentiates as an enormous everted gut, with the ectodermal vesicle reduced to a tiny appendage.
The opposite effect is achieved if, before fertilization, the egg is exposed to artificial seawater lacking calcium ions but with sodium thiocyanate (NaSCN) added to it. In this case the gut is diminished or completely absent, the ciliary bands in the ectoderm fail to develop, and the tuft of stiff cilia at the animal pole is increased in size.
It appears that all these phenomena may be accounted for by assuming that in the sea urchin’s egg two factors or principles exist which are mutually antagonistic and yet interact with each other at the same time, and that normal development is dependent on a certain equilibrium between the two principles. Each has its center of activity at one of the poles of the egg. The activity diminishes away from the center, producing a gradient of activity.
The two gradients of activity decline in opposite directions – the one from the animal pole, the other from the vegetal pole. Taking into account this type of distribution of activities, the factors or principles themselves have been called gradients. The two gradients are therefore the animal gradient, with a center of activity at the animal pole, and the vegetal gradient, with a center of activity at the vegetal pole.
According to this theory normal development depends on the presence of both gradients and on an equilibrium between them. If the animal gradient is weakened or suppressed, the vegetal gradient becomes preponderant, and the embryo is vegetalized; i.e., it develops, in excess, parts pertaining to the vegetal gradient, such as the gut.
If the vegetal gradient is weakened or suppressed, the animal gradient becomes preponderant, and the embryo is animalized; i.e., it develops, in excess, parts pertaining to the animal gradient, such as the tuft of stiff cilia at the animal pole. Other structures of the embryo, such as the ciliary bands and the skeleton, can develop only if both gradients are active, and the development is the more nearly normal the more the two gradients approach a correct equilibrium.
The equilibrium between the two gradients may be upset in various ways. The animal gradient may be weakened, with concomitant vegetalization of the embryo, by removing its center of activity (the blastomeres at the animal pole of the egg) or by the action of lithium salts. The vegetal gradient may be weakened by removing its center of activity (the vegetal part of the egg) or by the action of sodium thiocyanate.
The gradient concept aptly covers the results of various experiments on sea urchins’ eggs. In addition, we will consider the following experiments. It is possible to separate from one another the three groups of blastomeres in the 16-cell stage—the mesomeres, the macromeres, and the micromeres—and then to recombine them at will.
In isolation the three groups develop as follows:
1. Isolated mesomeres develop into a vesicle with a tuft of cilia owing to a preponderance of the animal gradient and animalization.
2. Isolated macromeres plus micromeres develop into an extremely vegetalized embryo with evaginated gut, the result of a preponderance of the vegetal gradient.
3. Mesomeres plus macromeres develop into a practically normal embryo; the macromeres bear a sufficiently strong vegetal gradient to counterbalance the animal gradient. Mesenchyme and skeleton develop in such embryos in spite of the absence of micromeres.
4. Micromeres alone are not capable of any development, since they do not keep together but fall apart.
5. Mesomeres (ectoderm) plus micromeres (mesenchyme) develop into a complete and more or less normal pluteus. This combination is especially illuminating since the gut of such embryos develops in spite of the absence of the macromeres, which should normally have supplied the material for the gut. However, the embryo possesses the two gradients, the animal gradient borne by the mesomeres and the vegetal gradient borne by the micromeres, and the possession of the two gradients seems to create the necessary conditions for normal development.
What has been said is sufficient to show that the presence of different cytoplasmic substances in the egg does not necessarily mean that there is a direct relation between these substances and certain specific organs, in the sense that the cells containing the particular kind of cytoplasm develop directly into the corresponding organ. The absence of the cytoplasm with red pigment did not prevent the development of the gut because the necessary conditions for gut development were provided for in another way.
Physicochemical Nature of the Animal-Vegetal Gradient System in Sea Urchin Eggs:
The existence of animal and vegetal gradients is proved not only by the patterns of morphogenetic processes occurring after certain interventions in the normal development of the egg, but also by the direct demonstration of the gradients as peculiar physicochemical states of the cells of the developing embryo.
One way of proving their existence is to study the reduction of dyes by the embryo under conditions of anaerobiosis. Sea urchin embryos (or embryos of other animals for that matter), slightly stained with the vital dye Janus green (diethylsafraninazodimethylaniline), are placed in a small chamber sealed off with petroleum jelly. After some time all the free oxygen contained in the chamber is used up as a result of the respiration of the embryos.
The embryos then, respires using Janus green as an acceptor of hydrogen, which is first reduced to a red dye, diethylsafranin, and further to a colorless substance, leucosafranin. That is, the light grayish-blue color of Janus green first changes to red, and then, as a second step in the reduction of the dye, the color disappears completely.
The reduction of Janus green, shown by the color change, does not occur simultaneously in the whole embryo but follows a characteristic sequence. In the late blastula and early gastrula of the sea urchin, the change in color is first noticeable at the vegetal pole, at the point where the primary mesenchyme is given off. Then the color change spreads, involving the whole vegetal hemisphere, and reaches the equator.
At this stage a spot of changed color appears at the animal pole and also gradually increases, so that the bluish color remains only in the form of a ring lying well above the equator in the animal hemisphere. Even this ring eventually disappears. The red color then begins to fade in the same sequence, starting first from the vegetal pole and then from the animal pole.
The order in which the dye is reduced in the embryo shows such a nice correspondence to the postulated animal and vegetal gradients that this alone would justify our mentioning these experiments. However, the connection goes much further. When either the animal or vegetal gradient is suppressed, the corresponding gradient of reduction also disappears.
In isolated animal halves of sea urchins’ eggs, the animal tendencies of development are predominant, and it is found that such animalized embryos start reducing Janus green at the animal pole only; there is no center of reduction at the vegetal pole.
In isolated vegetal halves the animal tendencies of development are suppressed and the vegetal tendencies are supreme. Correspondingly, the only center of reduction is at the vegetal pole; no center of reduction appears at the animal pole. Embryos animalized or vegetalized chemically show reduction gradients similar to isolated animal and vegetal halves of the egg.
The micromeres are carriers of the highest point of the vegetal gradient, and they preserve this property after transplantation. Accordingly, the micromeres can also serve as the center of a reduction gradient.
If a group of four micromeres is implanted laterally and the embryo is tested for reduction of Janus green, it can be seen that in addition to the two normal gradients, one from the vegetal pole and one from the animal pole, a third gradient appears, having the implanted micromeres as its center but spreading out from these to the adjacent cells.
In short, every modification of the gradient system of the embryo that is postulated from the morphogenetic behavior of the embryo is reflected in the gradients of reduction of Janus green.
If the micromeres are able to establish a vegetal gradient in the adjoining parts of the blastoderm, we would expect that they do so by producing some substance which diffuses into the nearby cells. To prove the existence of such a transmission, advantage was taken of an unusual peculiarity of the micromeres. After the fourth division of the egg, the micromeres lag behind in cleavage.
While the mesomeres and macromeres continue rapid division and accordingly synthesize DNA, the micromeres pause with the next divisions and start synthesizing RNA instead. This synthesis can be shown very clearly if the embryo is supplied with radioactively tagged uridine; only the micromeres take up the label.
If the micromeres are then transplanted into an isolated animal half of a 16- or 32-cell embryo, radioactivity can be detected as being widely spread in the cytoplasm of the host half-embryo. The RNA synthesized by the micromeres thus diffused into the cytoplasm of the adjoining cells. It is known that micromeres implanted in an animal half-embryo cause the development of an endodermal gut from cells which normally would have produced only ectoderm.
The foregoing experiment does not prove, in itself, that the RNA diffusing from the micromeres is responsible for the newly established vegetal gradient in the animal hemisphere cells. This explanation, however, is shown to be more likely by another modification of the labeling experiment. If a 16-cell embryo, instead of being provided with uridine, is supplied with 8-azaguanine, an analogue of uridine, this substance is incorporated into the RNA by the micromeres.
An abnormal RNA is produced. After the treatment the progeny of the micromeres—the primary mesenchyme—develop quite normally, but the gut completely fails to develop. This experiment has been interpreted as proving that the RNA diffusing from the micromeres is essential for gut development and that if this RNA is of an abnormal composition, it cannot perform its function.
From the setup of the preceding experiments it appears highly probable that the spreading of substances from the micromeres into the other parts of the blastoderm occurs directly from cell to cell and not through-the fluids surrounding the cells. The fluid filling the blastocoele in particular does not transmit any morphogenetic influences.
In this connection it is pertinent to refer to experiments that prove the possibility of substances moving directly from cell to cell without passing into the intercellular medium. It was first shown that electric currents may flow from cell to cell, which involves essentially the transmission of sodium and potassium ions.
Subsequently it was found that larger molecules, of molecular weight over 1000, but not greater than 10,000, can also pass from cell to cell by simple diffusion. The passage of these molecules (as well as of simple ions) is affected through a special type of junction, which can be established between most kinds of cells where they come in contact with one another.
The junctions are the so-called “gap junctions”. These are in fact tiny channels (20 Å in diameter) which connect the interior cytoplasm of adjacent cells. It has been found that gap junctions, permeable to intracellular substances, are established in the morula-early blastula stage in amphibians, fishes, birds, and squids.
Gap junctions are established between cells of the embryo very rapidly – within seconds of contact between cells of an early newt embryo. There is every reason to believe that wherever the interaction between cells and groups of cells is in the form of gradients; this interaction is performed through the medium of gap junctions.
The animal and vegetal gradients in the sea urchin embryo can be considered to be definite metabolic processes or systems of metabolic reactions which involve oxidation and which have their points of highest intensity at the animal and vegetal poles respectively.
The nature of these reactions is probably very complex and is not, as yet, fully understood, but some indications concerning these reactions may be deduced from a comparison of the chemical agents which may cause vegetalization or animalization.
Apart from the use of lithium ions, vegetalization may be caused by sodium azide and dinitrophenol. Both of these substances belong to a group of enzyme poisons; the azide is known to inactivate the cytochrome oxidase system, and the dinitrophenol disturbs respiration by preventing oxidative phosphorylation, that is, the formation of energy-rich bonds between phosphoric acid and adenosine diphosphate (resulting in the formation of adenosine triphosphate). This might mean that vegetalization is based essentially on a disturbance of oxidative processes in the embryo and, perhaps even more specifically, a disturbance of the processes of phosphorylation.
That the action of lithium ions is along the same lines may be deduced from a number of observations of which we shall single out the following. Lithium salt treatment suppresses the rise of oxygen consumption which occurs normally at the beginning of gastrulation, that is, at the same time as the morphological effects of vegetalization begin to be apparent.
Furthermore, inorganic phosphate accumulates in the seawater during the development of lithium ion-treated embryos—again a hint that these embryos are incapable of utilizing the energy of oxidation for phosphorylation and synthesis of adenosine triphosphate.
The whole sequence of reactions involved in vegetalization must, of course, be far more complicated. Thus, it has been found that the relative amounts of various amino acids change as a result of vegetalization in lithium ion-treated embryos.
Furthermore, if a modification of the respiratory system of the embryo is the essential feature of vegetalization, it remains to be discovered why the processes of morphogenesis at the animal pole of the embryo (essentially the ectodermal organs) are more severely affected than those of endodermal parts developing at the vegetal pole.
With regard to animalizing agents, sodium thiocyanate, has been found to be rather unreliable. Some batches of eggs treated by this chemical do not react at all. In other batches individual embryos become animalized to greatly varying degrees. Subsequently, many other substances have been found whose animalizing action is much more predictable.
These belong to several distinct groups:
i. Some metals – zinc, mercury.
ii. Some proteolytic enzymes – trypsin, chymotrypsin
iii. Many acidic dyes – in particular some possessing sulfonic (HSO3) groups in their molecules, such as Evans blue, chlorazol sky blue, trypan blue, and Congo red; and others possessing carboxyl groups (COOH), such as uranin and rose bengal.
iv. Some other sulfonated organic compounds – germanin and others.
v. Some anionic (acid) detergents.
Animalization may be achieved by extreme dilutions of a chemical; for instance, 1:50,000 in the case of Evans blue.
The common property of most, if not all, of the preceding agents seems to be their ability to attack proteins or to form compounds with proteins, especially basic proteins. In the case of the many sulfonated and carboxylated dyestuffs, it is fairly certain that their acidic groups become attached to the functional side groups of protein molecules.
The two metals, zinc and mercury, also become easily bonded to side groups of protein molecules. The case of mercury is very remarkable – mercuric chloride, a powerful poison used therefore as a fixative for proteins, causes animalization when applied for a very short time in great dilutions (1:90,000).
By blocking the active side groups of the protein molecules, the agents referred to probably “immobilize” these molecules, preventing them from interacting normally with other cell constituents. Even the steric configurations of protein molecules may thus be altered, since they are based on lateral bonding between parts of the polypeptide chains. A far-reaching change in the properties of proteins may thus be brought about.
Fortunately, in the case of animalizing agents, their point of attack can be clearly demonstrated. The dyes which have been used for this purpose stain the cells of the embryo, and it can be seen that the first cells to be stained are the primary mesenchyme cells at the vegetal pole.
With higher concentrations of the dye or with a longer duration of the treatment, the adjoining presumptive endodermal cells show the staining too. The zinc taken into the cells of embryos may be made visible because it yields a pink coloration with dithizone.
In embryos first treated with zinc and later immersed in a dithizone solution, the pink color is detectable in the same position as in the case of animalizing dyes. It is thus quite clear that animalization is due to damage to the center of the vegetal gradient.
While the treatment of embryos with various chemicals of simple and known composition (and, in part, of known mode of action) is useful in establishing the existence of gradients, it does not reveal the chemical mechanism of the gradients under normal conditions. A more direct approach is here indicated. Such an approach was made in attempts to isolate animalizing and vegetalizing substances from the embryo itself.
By homogenizing sea urchin unfertilized eggs or early cleavage stages and by separation of the materials by centrifugation and column chromatography, some fractions were obtained which had a distinct animalizing action on developing embryos. Several fractions with a slightly different effect could be detected.
Less success was obtained with vegetalizing substances, though some vegetalizing activity was also recorded. The chemical nature of the fractions has not, as yet, been determined, but the absorption curve of one fraction in ultraviolet light resembles that of the amino acid tryptophan, while the absorption curve of another suggests nucleotide structure. Further characterization of the substances is to be expected.