In this article we will discuss about:- 1. The Problem of Reversibility of Differentiation 2. Chemical Substances as Means of Controlling Differentiation 3. Epithelial-Mesenchymal Interactions.
The Problem of Reversibility of Differentiation:
Differentiation may be reversible to a certain extent. The morphological and physiological peculiarities of tissues require for their maintenance the environment which surrounds them in the normal organism. Except for manifestly nonliving parts such as the chitin cuticle in insects or the hairs in mammals, all other animal structures may become changed or may even dissolve and disintegrate if the normal conditions in the organism are changed.
A disintegration of the animal’s morphological organization is even more to be expected if the integrity of its body is interfered with, such as in the case of wounding or in experiments with explanation of parts of the animal’s organs and tissues.
The cultivation of small portions of tissues in a clot of blood plasma or any other suitable medium (tissue culture) is an especially effective method of investigating the extent to which the histological differentiation of tissues may be reversed. In tissue cultures, intercellular structures (fibers of connective tissue, matrix of the bone and cartilage) become destroyed, the normal arrangement of cells in the tissues becomes dissolved, and the specific organoids of cells may also disappear (e.g., myofibrils or cilia).
ADVERTISEMENTS:
Cells derived from different tissues may acquire a very similar appearance, an appearance not unlike that of cells which have not yet undergone differentiation. These phenomena may be conveniently called de-differentiation.
We have seen that in differentiation there is an increase of proteins in the cells in relation to nucleic acids. The reverse is true in the case of dedifferentiation, which involves a breakdown of functional mechanisms of cells (composed of protein). Thus, it can be expected that the amount of protein in the cells would diminish as compared with the amount of nuclear material.
This is actually the case. Determinations of protein nitrogen and deoxyribonucleic phosphorus were carried out on a culture of fibroblasts taken from a chicken’s heart and grown in vitro. After six days of cultivation, the amount of protein nitrogen per unit of deoxyribonucleic acid phosphorus was diminished by one half.
The dedifferentiation of cells under conditions of tissue culture or the retention of a lower state of differentiation, if embryonic cells are being cultivated, is the result of the change in environment in which the cells are being kept. The separation of a part of the body from other parts and the damage to the tissues caused by cutting them may contribute toward these changed conditions.
ADVERTISEMENTS:
However, the main factor which causes the cells to lose their differentiation and to start growing and proliferating instead is the medium which surrounds them in tissue culture, and which is different from the medium (the fluid bathing the cells) in the intact organism.
The standard medium for tissue cultures consists of blood plasma, embryo extract, and some modification of the Ringer saline solution. The salts are necessary for the upkeep of ionic balance between the cells and the surroundings. The blood plasma contains fibrin, which clots and forms the solid substrate on which the cells can spread out and which also becomes slowly dissolved and supplies some of the necessary nutrients for the cells.
The embryo extract is added as a growth-stimulating agent. Without embryo extract, on blood plasma and saline alone, the cells grow only very slowly, if at all. With embryo extract added, the cells start growing and proliferating rapidly, and dedifferentiation occurs.
The embryo extract naturally contains some of the substances surrounding the cells in the early embryo. The behavior of cells in tissue cultures may thus be attributed to the fact that the medium bathing the cells tends to keep them in a condition characteristic of the embryo in the stage from which the extract had been taken.
ADVERTISEMENTS:
In a seven-day-old embryo, often used for the preparation of embryo extract, growth is rapid, and there is as yet not much histological differentiation. If this interpretation were essentially correct, it should be possible to produce progressive differentiation of cells by exposing them to media containing extracts from consecutively older embryos.
This has been done in the following experiment. Osteoblasts derived from a 16-day-old chick embryo were kept in cultures in two series. The tissues were grown in flasks (Carrel flasks) on the surface of a blood plasma clot and suffused with embryo extract, which was changed every two days. In one series, the embryo extract was always the same, prepared from seven-day-old chick embryos.
In the other series, older embryos were used to prepare the extract at each subsequent change, namely- 10-day, 12-day, 15-day, and 18-day embryos, then the extract from the heart of a newly hatched chick, and lastly blood serum of an adult hen.
The changing extracts were to imitate the changes in the tissue fluids with progressing development. The experiment yielded results according to expectation; the first culture grew and proliferated without any differentiation, while bone developed in the second culture.
In spite of the apparent simplification of the cells which have undergone dedifferentiation, they do not revert to the state of embryonic cells. Evidence acquired from numerous experiments, which cannot be considered here in detail, proves that dedifferentiated cells retain their histological specificity and do not acquire new competences.
If cultivated alone, renal tissue may become disintegrated, and its component cells grow out in the form of a disorganized sheet or layer, but the cells remain kidney cells, and given suitable conditions, they again arrange themselves into the shape of tubules.
Similarly, cartilage may be dedifferentiated, and the cartilage cells then grow as a disorderly mass, scarcely distinguishable from a mass of connective tissue cells. But if the culture is kept under conditions which do not favor rapid growth, such as when the culture medium is poor in growth-promoting substances (small amounts of embryo extract) or if it is not changed often enough to a fresh medium, a new differentiation becomes possible. When this happens, the former cartilage cells again secrete cartilage matrix, thus showing that they retained their functional specificity in spite of morphological simplification.
In tissue cultures, cells are dissociated from one another as a result of the breakdown of the bonds which bind them together in a normal tissue. The dissociation is not complete, especially in the case of epithelial tissues, which remain joined in a sheet even when cultivated in vitro, nor is it quite under the control of the experimenter. Therefore, the attempt was made to separate tissue cells by more direct methods in order to test to what extent individual cells can retain the peculiarities that they had acquired previously in conjunction with other cells.
By grinding tissues, especially embryonic tissues, in a specially prepared small glass mortar, they may be disaggregated, and a sufficient number of individual cells remain alive and may be used for further study. A more delicate method is to treat tissues with a weak solution of trypsin in a calcium- and magnesium-free saline. This treatment causes the cells to separate from one another. A suspension can be obtained in this way consisting almost exclusively of completely separated individual cells which appear to be quite healthy.
ADVERTISEMENTS:
The suspension may then be put into a medium in which the cells can re-aggregate and, under favorable conditions, resume differentiation. In some experiments, the medium was that used for tissue culture with reduced embryo extract to facilitate differentiation.
In other experiments, the cell suspension was injected into the veins of chicken embryos, and the cells became disseminated through the vascular route. Individual cells or small clusters of cells then settled at various sites in the body of the embryo or on the chorioallantois and either were incorporated into the tissues of the host or gave rise to small local growths, teratomas.
The main result of these experiments was that the cells, which had passed through a condition of complete disaggregation, were able to resume specific tissue differentiation, whether on a plasma clot or on the chorioallantois of a living embryo.
Masses of brain tissue, muscle, cartilage, bone, nephric tubules, glandular tissue, or epidermis, usually in the form of cysts with clearly differentiated feather germs, were observed in various experimental series. The assortment of tissues appearing in any experiment depended on the origin of the cell suspension, the stage of the embryo from which the cells were derived, and the part of the embryo taken.
The results available so far are compatible with the assumption that every type of cell differentiates after disaggregation in conformity with its previous differentiation. Thus, if cell suspensions were prepared from whole embryos, the teratomas contained a wide variety of tissues including nervous tissue, musculature, and glands. If only limb-buds were used for preparing the suspension, the teratomas contained epidermis, cartilage, bone, and mesenchyme, but no nervous tissue, muscle, or glands.
It is remarkable that structures developing from cell suspensions do not present a chaotic assemblage of different types of differentiation. Rather, they produce parts resembling organ rudiments of a normal embryo, with the various tissues segregated from one another and each tissue arranged in a recognizable morphological unit- nerve cells form brain vesicles with a central lumen; cartilage cells are sometimes arranged in elongated rods with perichondrial ossification; epidermis cells are arranged in layers with a clear distinction between proximal and distal surfaces; and feather germs may show a very high degree of internal organization.
As it is unlikely that each unit of tissue is always derived from one single cell, it follows that the cells sort themselves out in some way—that cells of any one kind join together and group themselves anew in an order similar to what they had before they had been separated. This is the same process that we have found in relation to cells at the time of gastrulation and neurulation and is further evidence of the “affinities” between particular kinds of cells.
Chemical Substances as Means of Controlling Differentiation:
Since the actual differentiation of tissue cells is dependent on conditions in their environment, it is possible, to direct their development in definite pathways by exposing them to appropriate treatments. Further experiments along these lines will now be described, first of all in relation to the control of differentiation by such chemical substances as vitamins and hormones.
Stratified epithelium in vertebrates appears to be a suitable tissue for experiments of this kind. Under normal conditions, stratified epithelium takes on various forms, both in different groups of vertebrates and in different parts of the same animal. In terrestrial vertebrates, stratified epithelium is squamous and cornified on its surface, but in fishes the epidermis, though stratified, is not squamous and contains mucus-secreting cells. The degree of cornification in terrestrial vertebrates varies, being very strong on the surface of the body but weak in the lining of the oral cavity, the pharynx and the esophagus.
In the vagina, the stratified epithelium undergoes cycles of cornification accompanying the menstrual cycle. The epidermis gives rise to a number of glands, among them the mammary glands, in which the epithelium becomes simple columnar, but under pathological conditions it may revert to the cornified type, such as in some kinds of cancer. Lastly, in the endodermal part of the alimentary canal of mammals the esophagus is lined with stratified epithelium, while the posterior parts, starting with the stomach, are lined with columnar mucus-secreting epithelium.
The transformation of stratified epithelium from the cornified to the non-cornified type may be achieved by purely chemical methods under conditions of explanation in vitro. If small pieces of vaginal wall of juvenile female mice are cultivated in the standard media for tissue cultures, the epithelium remains, without any traces of cornification. If, however, the female sex hormone, 3, 17β-estradiol, is added to the culture medium, the epithelium becomes squamous and cornified on its surface. Other preparations of estrogens have a similar effect.
The opposite transformation may be achieved by treatment of the cells with vitamin A. If the skin from young seven-to eight-day incubated chick embryos is grown in the ordinary tissue culture medium (blood plasma + embryo extract), it develops a squamous cornified layer on the surface of the epithelium. An addition of vitamin A to the medium causes a complete transformation of the epithelium, which now becomes a cuboidal or columnar mucus-secreting one.
It is remarkable that the cells do not need to be continuously in a vitamin A-enriched medium to be converted into the non-keratinized type; a short treatment with the vitamin suffices to switch the development from the one channel into the other. To enable the vitamin to reach each epithelial cell in a short time, the skin of a chick embryo was dissociated into single cells by trypsin treatment, and these were then immersed in a 0.06 per cent solution of vitamin A for 15, 30, or 60 minutes.
After this, the cells were put onto a plasma clot and cultivated for several days. The cells re-aggregated and in the best cases formed cysts or vesicles, with the distal surface of the epithelium turned inward and the outer surface surrounded by connective tissue. Typical stratified squamous epithelium developed in controls not exposed to vitamin A.
As to the vitamin A-treated preparations, it was found that even a 15-minute sojourn in the vitamin solution is sufficient to transform the cells into the non-keratinizing type, while additional periodic washings (30 minutes every two days) furthered the development of the epithelium into a typical columnar epithelium with goblet cells.
Epithelial-Mesenchymal Interactions:
A special type of condition for the differentiation of some tissues is the interaction between epithelium and mesenchyme. The organized growth of many epithelial structures and the very preservation of the epithelial arrangement of cells are dependent on the presence of connective tissues on one (the proximal) surface of the epithelium. An example of this is presented by the embryonic epidermis of the early amphibian embryo.
When isolated without mesoderm or mesenchyme of any kind, embryonic epidermis soon loses its epithelial arrangement; the cells acquire a reticulate arrangement and eventually degenerate and die off. In the presence of mesodermal mesenchyme, the epithelial arrangement is preserved; the ectoderm remains healthy and differentiates as normal skin epidermis.
When epithelial tissues are grown in tissue culture they tend to grow as sheets of cells which, though preserving contact with one another, do not follow the arrangement that was present in the original tissue or that should have arisen in the course of development, if the part taken for cultivation was the early rudiment of some epithelial structure.
For example, if epithelial cells of renal tubules are cultivated alone, they form a disorganized sheet or layer spreading out on the surface of the plasma clot. If, however, some connective tissue cells are added to the culture, the spreading out of the sheet of renal cells is arrested, and they become reconstituted into tubules, which bear a similarity to the normal tubules of the kidney.
The dependence of epithelia on the connective tissue can be demonstrated very clearly when rudiments of glands are cultivated in vitro with or without connective tissue. The submandibular glands of a mouse embryo appear on the thirteenth day of gestation in the form of a pair of solid, club-shaped buds, growing from the buccal epithelium down into the connective tissue layer, one on each side of the tongue. Soon the epithelial bud becomes surrounded with dense mesenchyme forming a “capsule.”
On the fourteenth day, the tip of the epithelial bud becomes indented. This is the beginning of the branching of the gland rudiment, and the two primary branches grow out, each bearing a knoblike thickening at its end. The end knobs divide repeatedly and eventually give rise to the secreting acini of the gland, while the more proximal parts form the ramified system of ducts. While this outgrowth and branching of the epithelial parts goes on, the capsule mesenchyme surrounds the branches and penetrates between them, forming the connective tissue of the gland. The ducts and acini become hollowed out at a later stage.
This development can be observed on whole gland rudiments cultivated in vitro in a plasma clot. By treatment of the early thirteenth day rudiment of the submandibular gland with a 3 per cent solution of trypsin for three to five minutes in the absence of calcium and magnesium ions, the cohesion of the mesenchyme with the epithelium can be destroyed, and the two components of the rudiment may be separated from each other.
On placing them in a culture medium, both components are found to be fully viable, but their differentiation is no longer normal- the capsule connective tissue produces a typical culture of mesenchyme, with individual cells spreading out radially from the initial piece of tissue; the epithelial bud loses shape and becomes transformed into a sheet of growing cells. This behavior, however, does not show that the cells of either the capsule or the epithelial part have changed in their essential nature.
If the two components are set in the culture medium near to each other, the mesenchyme comes to surround the epithelial rudiment, whereupon the latter starts sprouting and branching in a very nearly normal fashion. The interaction between the epithelium and mesenchyme restores the system to its normal state.
The mutual influence of the epithelium and mesenchyme of the submandibular gland rudiment may be considered to be a special case of induction, and as with inductions occurring in earlier stages, it may be questioned whether the influence of one component on the other is specific or not.
To answer this question, isolated epithelium of the submandibular gland was cultivated with mesenchyme of different origins – mesenchyme from the rudiment of the maxilla, from the somites, from the lateral plate, or from the lung rudiment. With all foreign mesenchyme, the epithelial rudiment of the submandibular gland failed to sprout and ramify. The sheet-like spreading out of the epithelial cells was arrested, however, and the rudiment developed eventually into an epithelial cyst.
When the epithelial rudiment was implanted into a culture of capsular mesenchyme, which had been previously killed by heat, the rudiment behaved much the same as when it was surrounded by foreign mesenchyme; the flattening out and spreading of the epithelium was suppressed, but no growth or ramification took place. Apparently the stimulus necessary for the normal development of the duct and acini system of the submandibular gland can only be given off by the capsular mesenchyme of this organ and only in the living state.
The mode of development described for the submandibular gland is a common feature in other structures of the vertebrate body – other salivary glands, the mammary gland, the pancreas, and even in non-glandular structures such as the lungs.
The development starts with a club-shaped bud, which elongates and then starts branching, producing eventually the acini of the gland, or the alveoli in the case of the lungs. In all these cases it has been shown that the development of the epithelial parts is dependent on the surrounding mesenchyme. A somewhat different type of tissue interaction is encountered in the development of the renal tubules of the metanephros.
The development of renal tubules of the metanephros is dependent on the presence of the growing and ramifying ureter bud. Under the influence of the latter, the loose mesenchyme of the metanephric rudiment (the metanephrogenic mesenchyme) becomes partially converted into epithelium and forms convoluted tubes, the renal tubules, which in normal development link up with the terminal ramifications of the ureteric ducts.
The interaction of the ureteric bud and the metanephrogenic mesenchyme can be observed in vitro in the same way as the interaction of epithelium and mesenchyme of the submandibular gland, that is, after separating the epithelial and the mesenchymal components by trypsin treatment and reuniting them in the culture medium, though the results of the reaction here are, of course, essentially different.
By combining metanephrogenic mesenchyme taken from an 11-day-old mouse with other tissues, it was found that the ureteric duct is not the only part that causes the mesenchyme to be converted into renal tubules. The epithelial part of the submandibular gland cultivated together with metanephrogenic mesenchyme produces the same effect, although there was no reciprocal action; the gland epithelium remained un-branched.
The epithelium of the submandibular gland can thus serve as an “abnormal inductor” of renal tubules, just as adult liver can serve as an abnormal inductor of a neural plate. Furthermore, the spinal cord, especially its dorsal half, proved to be a very efficient inductor of renal tubules, when placed in a culture of metanephrogenic mesenchime.
The renal tubules are always developed in the immediate vicinity of the inducing tissue. An attempt was made therefore to test whether immediate contact is necessary for the induction to take place. For this purpose, the inducing and reacting tissues were separated by thin membranes of various degrees of porosity.
Cellulose ester membrane filters were used, varying in thickness from 20 to 150 µ, and with pores approximately 0.8 µ, 0.4 µ, and 0.1 µ in diameter. The experiment consisted essentially in arranging for two tissue cultures to grow, one on each side of the membrane filter.
The inducing culture (spinal cord) was grown on one side of the membrane and the reacting tissue (the metanephrogenic mesenchyme) on the other side. It was soon discovered that the inductive influence could easily pass through the coarser filters (with pores 0.8 µ and 0.4 µ in diameter) of up to 60 µ in thickness but could not pass if the filter was 80 µ thick (or consisted of four or more layers each 20 µ thick).
With finer filters (pores approximately 0.1 µ in diameter) the induction became weaker, and the influence could cross only a thin membrane not exceeding 30 µ in thickness. A still finer filter, such as a cellophane membrane 20 µ thick, effectively stopped the inducing influence.
Positive results have also been obtained when the submandibular gland components were cultivated on the opposite sides of a membrane filter, the epithelial part on one side and the capsule mesenchyme on the other. The epithelial bud produced numerous ramified outgrowths, thus showing that the inducing principle could penetrate through the filter.
In experiments with nephrogenic mesenchyme the electron microscope was applied to see what was going on in the pores of the filter membrane separating the inductor and the reacting tissue. In ultrathin sections made perpendicular to the separating membrane, it could be seen that in the case of coarser membranes, the pores contained cytoplasmic outgrowths of cells, coming both from the neural tissue and from the mesenchyme.
The possibility was therefore not excluded that the processes of the two kinds of cells met somewhere inside the membrane and thus established a direct contact between the inducing and reacting cells. These processes were, however, more scarce when filters with 0.4 µ pores were used, and in the filters with 0.1 µ pores there was practically no penetration of the filter by cytoplasmic processes of the cells, except for a few small in-pocketing’s which did not go more than 1 to 2 µ into the substance of the filter.
More recently these results have been disputed, as it was found in new experiments that processes of the inducing cells (spinal cord cells) do penetrate through the pores of a porous membrane. However, the membrane used in the new experiments was much thinner than in the experiments mentioned before – only 10 µ and 20 µ, instead of up to 60 µ in the older experiments.
Similar methods were used in another very interesting case, the differentiation of the pancreatic rudiment in the mouse embryo.The epithelial part of the pancreatic rudiment does not differentiate if it is separated from the adjoining mesoderm.
This case is of particular interest, as the signs of differentiation were not confined to morphological changes (formation of tubules and the like) as in the kidney or salivary gland tissue, but the actual process of specific pancreatic secretion, as seen in the formation of secretory granules, could also be detected. Thus, the mechanism of specific protein synthesis was obviously involved.
The tissues were put to the same test as the metanephrogenic tissues; that is, the epithelium of the pancreatic rudiment and the mesodermal tissue were allowed to grow in culture on opposite sides of a porous membrane. The pancreatic epithelium of 11-day embryos, which when cultivated alone did not produce pancreatic enzymes, differentiated fully with the formation of secretory granules when exposed to the influence of mesodermal tissue acting through the porous membrane.
The action of the mesoderm turned out to be even less specific than in the case of kidney tissue, since not only was pancreatic mesoderm able to stimulate pancreatic cells to secretion, but so were other kinds of mesoderm (from the kidney, from the salivary gland, and from the lung). The diffusible substance emanating from the mesenchyme and stimulating the development of glandular and other similar structures has been called the “mesenchymal factor”. It can be isolated from chick embryos.
The action of this substance has been tested on the growth and development of the pancreas rudiment by binding the mesenchymal factor to Sepharose beads and bringing the beads into contact with pancreas rudiments in vitro. The pancreatic tissue wrapped itself around the beads and started to proliferate. In this way it was established that the mesenchymal factor is a chemical substance which may cause pancreatic rudiments (and other glandular and non-glandular structures) to grow and proliferate by mitosis.
That the mesenchymal factor may act as an “inducing substance” is evident from a following experiment carried out on the lung rudiment in the mouse. In the lung rudiment the branching and proliferating distal parts of the rudiment give rise to the bronchioles and the alveoli of the lung. The proximal “stem” of the rudiment becomes the trachea. In the trachea the proliferation is retarded in rather early stages, and it remains a simple tube, while the distal parts continue to grow and branch out for a long time.
The active proliferation of the distal parts is dependent on the influence of the surrounding “bronchial” mesenchyme. If the mesenchyme coat of the trachea is removed (in rudiments grown in vitro), and a mass of bronchial mesenchyme is brought into contact with the tracheal epithelium, an additional bud is formed on the trachea, and in due course it develops into an arborescent growth—a supernumerary lung lobule. This is as clear a case of induction as occurs in any developing system. The induction here has, as its main component, the local stimulation of growth and proliferation.
Induction of supernumerary tracheal buds also has been achieved with another “growth factor” – the so-called epithelial growth factor, which may be extracted from sub-maxillary glands and also from some other sources. The experiment was performed on lung rudiments of the chick embryo grown in vitro. Agarose pellets impregnated with the epithelial growth factor were placed next to the tracheal epithelium, and this led to the appearance of new growing and proliferating buds.
As the epithelial growth factor is a powerful stimulator of cell proliferation in epithelial tissues, it could be concluded that stimulation of growth and cell division is an essential component in the formation of buds on the trachea, and probably also in other similar morphogenetic processes in later development.