Braz Dent J (1990) 1(1): 5-15 ISSN 0103-6440
| Introduction | The enamel organ | Ameloblastoma: benign or malignant tumor? | Why is hard tissue not formed in ameloblastoma? | Epithelium/mesenchyme interaction | Morphodifferentiation | Odontoblastic differentiation | Ameloblastic differentiation | Stratum intermedium | The role of connective stroma | References |
The ameloblastoma, because of its aggressive clinical behavior and its histological feature apparently benign, constitutes a puzzling paradox. Some hypotheses concerning this strange clinical-histologic contradiction are analyzed, as well as the additional paradox represented by the neoplastic parenchyma itself, in which a tissue consisting of cells nominally able to form enamel do not elaborate any of the calcified dental tissues.
Key words: ameloblastoma, odontogenic tumors.
The enamel organ is a structure comparable to the dermal appendages, i.e., the hair follicle, the sweat gland and the sebaceous gland, or to those derived from the oral mucous membrane, i.e., the accessory salivary glands.
Amongst the skin appendages, perhaps the one that most resembles the teeth is the hair, with the main difference being the latter's continual growth, whereas the tooth has limited growth. However, there are some animals - the rodents - in which growth of several teeth (the incisors) is equally continuous, a fact which narrows even more this analogy.
The odontogenic organ considered from the moment of its formation, although appearing to be a simple structure when analyzed superficially, in fact is composed of a variety of different cellular types which undergo a series of morphological, physiological and biochemical modifications throughout the several phases of cellular development and differentiation. These modifications mobilize a complicated biological mechanism (not yet completely explained or understood) which renders the tissue inter-relations in which it is involved very complex, principally at the level of the ectodermal/mesodermal interface of the tissues which constitute the developing dental organ.
Thus, considering the ectodermal component of the odontogenic organ, the dental lamina (which marks the beginning of tooth formation) presents a histological appearance very similar to the deeper cellular layers of the epithelium seen in the primordium of the oral mucosa. The outer epithelium of the enamel organ is composed of short cells, while the inner enamel epithelium is composed of tall cells. The stratum intermedium is formed of flat cells, while the stellate reticulum exhibits a polygonal pattern microscopically.
In regard to the mesodermic component of the odontogenic organ, the dental papilla shows a mesenchymal aspect near the inner enamel epithelium. The dental sac, initially very similar to the dental papilla, becomes more fibrous near the outer enamel epithelium as the odontogenesis evolves. In addition, at the right moment, the cells change their initial aspect at the inner epithelium/dental papilla interface, acquiring columnar characteristics on both sides of the basal membrane which separates the two tissues. The cells of the inner enamel epithelium alter the polarity of their nuclei during the so-called secretory phase.
This group of controlled modifications commands the biological process of tooth formation. The understanding of these phenomena that occur during this complex group of successive transformations (and how they may be altered pathologically) may one day suggest a rational explanation for the existence of this series of clinically and/or histologically different pathological conditions, despite being derived from the same original odontogenic organ. This could explain why the rests of Hertwig's root sheath (epithelial rests of Malassez) generally lead to periodontal cysts, lacking tumoral characteristics, while dental lamina remnants (epithelial rests of Serres) may lead to ameloblastomas, which are extremely aggressive neoplasms. What environmental factors (or of any other nature) determine these differences? This is a question which until now does not have a convincing, satisfactory answer, in spite of the enormous volume of research trying to solve the numerous doubts that persist. Therefore, only hypotheses remain purely speculative despite being founded on observations and logical arguments.
These characteristics of the ameloblastoma remember the basal-cell carcinoma, a known malignant neoplasm, although of low malignancy by its slow, invasive growth and by the fact that it only occasionally produces metastases. Ameloblastoma, on the contrary; known to be benign by its histological aspect, nevertheless, presents a highly aggressive behavior and, despite its slow growth, it is extremely invasive, as are malignant tumors, and produces occasional metastases. The final result, in both cases, is exactly the same.
Basal-cell carcinoma is a neoplasm which develops exclusively in the skin, from the epidermal basal layer. or root sheaths of hair follicles, never occurring in the mucosa (Shafer et al. 1983). As the basal layer of the epidermis is composed of cells potentially capable of differentiating into tiny of the skin appendages, aborted attempts to produce these types of structures arc at times seen in basal-cell carcinomas.
Ameloblastoma is equally a tumor derived indirectly from the epithelial basal layer, however from the covering of the gingival mucosa, perhaps from embryonic remnants of that which could be considered grossly as gengivomaxillary appendages: the teeth.
The histology of a typical basal-cell carcinoma (Figure 18) is very similar to that of primordial or basaloid type of ameloblastoma (Figure 1C). The follicular aspect observed in many of the cellular islets of the tumor represents only an aborted attempt of the neoplastic tissue to form teeth (Figure 1D), in the same manner in which the basal-cell carcinoma also attempts to form hair follicles.
This analogy of clinical behavior as well as the histological picture between the two tumors was what led Willis (1948) to categorically affirm that "attempts to distinguish benign and malignant adamantinomas are futile; they are all malignant in that they are locally invasive and prone to recur".
Henceforth, the following question remains: May there not be several cases of ameloblastomas - mainly extra-osseous tumors which develop in the gingiva - the type of basal-cell carcinoma that is said not to exist, I.e., the basal-cell carcinoma with its seat in the mucosa?
This detail of embryonic cellular remnants from the tooth germ evoked by Willis appears interesting, and perhaps explains much in respect to the question. In fact, embryonic development involves spacial as well as temporal details which cannot be upset - i.e., things ought to always occur at an established place and at an established moment, all in a pre-established order, as if it was previously programmed by a computer. If there is any change, be it ambiental conditions, be it the time in which the phenomenon ought to occur, there is always a detour in programming, with results and consequences quite different. from those foreseen by the biological determinism.
In relation to ameloblastoma, it is necessary to analyze step by step the circumstances and the moments in the odontogenic process in which it is possible for local or temporal errors to occur which could determine the biological impossibility of the ameloblastomatous tissues to produce calcified dental tissues in the tumoral mass. For this, it is necessary to invoke each evolutive step of odontogenesis, trying to associate the observed histological details common in ameloblastomas, in order to detect eventual space-time errors capable of impairing the elaboration of dental tissues in a neoplasm which theoretically is composed of forming cells of these tissues.
I. Epithelium/mesenchyme interaction. Odontogenesis begins about the 6th week of intrauterine life, with a thickening of the epithelium of the embryonic oral mucosa resultant from the multiplication of the cells of the basal layer of this epithelium, forming the dental lamina and the tooth buds. The local changes occurring in the epithelium induce a condensation of mesenchymal cells around the tooth buds, principally joined to its distal extremity, giving origin to the dental papilla and an outline of the dental sac. What determines these alterations in cellular behavior, epithelial as well as mesenchymal, at exactly that instant in embryonic development? To explain the phenomena which occur at predetermined moments of tissue development, and others which appear to occur in the interaction of different tissues at their interfaces, Slavkin (1988) suggests the existence of chemical mediators produced by the cells, autocrine and paracrine factors. The autocrine factors are those which, elaborated by the cell at a determined time of its development, act on the actual cell, modifying its behavior from then on. The paracrine factors, formed by some cells, are liberated into the tissue space, acting on other nearby cells altering the behavior of those and not the behavior of the secreting cells, as is the case with autocrine factors.
These autocrine factors, in the specific case of the odontogenesis, act at the time of proliferation of the cells of the basal layer orienting them to form the dental lamina and the dental buds. Paracrine factors are responsible for mesenchyme condensation near this epithelial tissue in proliferation.
The important factor to consider at this point of tooth germ development is the moment when the phenomenon occurs, i.e., the embryonic period, in which the epithelial cells are prone to proliferate, and the connective cells still have a mesenchymal character, appearing only slightly differentiated.
In ameloblastoma: Exactly in what are the environmental conditions different in ameloblastomas? In everything. In general, due to the fact that the moment of its installation is another, the connective stroma present in the tumor is dense fibrous and already perfectly differentiated. The histological aspect of the neoplasm makes one believe that probably the inductive power of the proliferating epithelium, which acts effectively on mesenchymal cells, perhaps does not act as appropriately on differentiated fibrous connective tissue. In consequence, instead of activating cell proliferation in the sense of causing cellular condensation dud dental papilla formation, the result of the epithelial influence on the connective tissue is rather a degenerative process of present collagenous fibers. This fact explains the clear area of less dense tissue often seen around the epithelial islands and cords of the neoplastic parenchyma (Figure 2A,B). It gives the impression that the paracrine factors, in the presence of a differentiated tissue, try to revert this differentiation process and return the surrounding connective tissue again to the mesenchymal state.
This degeneration of the fibrous stroma is responsible for the formation of cystic cavities in the interior of connective tissue, found principally in the type of ameloblastoma known as plexiform (Figure 2C). It has already been suggested that this degeneration may be caused by a nutritional deficiency, as a result of the complicated path of the blood vessels through the existing connective tissue labyrinth within the mesh of the neoplastic epithelial network (Spouge, 1973). However, the actual aspect of the tissue seems to contradict this hypothesis, since the blood vessels, which are directly responsible for the tissue nourishment, are exactly the last structures to disappear in the degenerating tissue (Figure 2D).
2. Morphodifferentiation: The condensation of mesodermic cells around the free end of the epithelial buds induces this epithelium to differentiate morphologically, adopting the specific form of the determined dental type that this bud would form in the future. This mutual induction in the epithelium/connective tissue interface was studied extensively by Kollar (1972), Koch (1972) and Slavkin (1972). According to Kollar (1972), for example, the basal membrane which separates these two tissues of different embryological origin ought to fulfil a very important role in this interface, as does the quality of the adjacent connective tissue. In case the environmental conditions are altered in relation to the normal embryological situation, the epithelium of the enamel organ also changes its behavior. Experimental research in animals, involving the transplantation of maxillary dental germs to diverse areas shows that if, for example, these germs are transferred to the plantar tissues of the animal paw, where the stroma is the dense fibrous connective type, the epithelium of the enamel organ begins to irregularly proliferate, adopting the pattern of growth histologically similar to that of the ameloblastoma. This detail possibly explains the irregular proliferation of ameloblastomatous neoplastic tissue since the stroma in this tumor usually is also the differentiated fibrous type (Figure 2A,B).
3. Odontoblastic differentiation: In odontogenesis, once the dental papilla is formed, the induction by this part, besides the general dental form, also transforms the inner epithelial cells of the enamel organ into pre-ameloblasts. These induce the cells of the adjacent dental papilla to differentiate into odontoblasts, forming two layers of different embryonic origin, which face vis-a-vis, separated by-the basal membrane.
In ameloblastoma: This induction does not occur in ameloblastoma because a mesenchyme capable of responding to this inductive action of the pre-ameloblasts does not exist; rather, there is a dense fibrous connective tissue (Figure 2A,B). This detail explains the lack of odontoblasts in the neoplastic tissue and consequently the lack of formation of dentinary matrix and calcified dentin.
4. Ameloblastic differentiation: Once the odontoblastic layer is complete, the formation of the dentin matrix begins with a thickening of the basal membrane which separates the two tissues. This initial thickening progressively increases and, as the odontoblasts are pushed apart, initiates the depositing of calcium salts on the matrix so formed. With the formation of the dentin matrix there is an interruption of the source of nourishment for the cells of the inner epithelium of the enamel organ which obliges these cells to acquire their nutrients at another source, I.e., from the stellate reticulum. This leads to an internal recomposition of the intracellular organelles of the pre-ameloblasts: the nucleus polarizes at the opposite end to the basal membrane, adjacent to the stellate reticulum, while the other structures, such as the Gold complex and the endoplasmic reticulum, position between the nucleus and the basal membrane, preparing for the depositing of organic enamel matrix and posterior calcification. In short, the pre-ameloblasts differentiate into actual ameloblasts (Figure 3A).
In ameloblastoma: In spite of the fact that in ameloblastoma the differentiation of connective cells in odontoblasts does not occur, and consequently dentin matrix is not formed or calcified, in the peripheral layer of the epithelial islands and cords which compose the neoplastic tissue, cells very similar to ameloblasts are present. There is evidence that in these cells there exists a clear polarization of the nuclei in the cellular extremity opposite the connective cells and thus adjacent to that which would be the stellate reticulum. This leads to conjecture if the simple presence of a thickening of the basal membrane would not be sufficient to determine the migration and polarization of the nuclei in the opposite cellular extemity, giving the tumoral cells a histological aspect similar to those of ameloblasts, since this type of thickening is commonly encountered around the epithelial islands and cords of the tumoral tissue (Figure 3B).
However, if these neoplastic cells are similar to differentiated ameloblasts, why are they unable to form a recognizable enamel matrix? Perhaps because they lack some detail which transforms them in biologically active ameloblasts. This functional detail is probably the absence of a stratum intermedium adjacent to the ameloblastic layer. In fact, the stratum intermedium cells are looked upon as the suppliers of elements adequate for posterior metabolism of the ameloblasts, these elements that the ameloblasts are not able to mobilize from the stellate reticulum. For some reason, there is no stratum intermedium in the ameloblastoma (Figure 3C).
5. Stratum intermedium: In normal odontogenesis, when polarization of the nuclei occurs and the epithelium passes from the inductive to the secretory phase, differentiating into active ameloblasts, profound modifications in the stellate reticulum and in the outer epithelium of the enamel organ occur. The stellate reticulum atrophies driving the outer epithelium to approximate the stratum intermedium, forming the reduced epithelium of the enamel organ. At the same time, the outer epithelium acquires a meshed aspect, becoming permeable to nutritive elements from the blood capillaries of the dental sac, which are then more conspicuous close to the reduced epithelium, protruding to the stellate reticulum (Figure 3A). This all facilitates the arrival of nutritional elements to the stratum intermedium where they will be pre-metabolized, passing later to the ameloblasts.
In ameloblastoma: Although forming an ameloblastic layer in ameloblastoma, it is not able to elaborate enamel matrix, probably because it lacks the stratum intermedium. Perhaps, the explanation for the lack of stratum intermedium is in the absence of outer epithelium in the neoplastic islands, which could hinder the formation of reduced epithelium. In fact, it is only necessary to histologically examine an ameloblastoma to verify that the neoplastic epithelial islands are surrounded by a peripheral cellular layer resembling the inner epithelium of the enamel organ, either in the pre-ameloblastic or the ameloblastic phase (Figure 3C,D). There is no simultaneous occurrence of the cellular layers which could recall the inner and outer epithelia occurring at the same neoplastic island, or the formation of the reduced epithelium of the enamel organ.
Lucas and Thackray (1952) attribute the formation of intrafollicular cystic cavities to a deficiency in absorption and diffusion of nutritive elements (coming from the perifollicular blood capillaries) to the center of the cellular islands, causing their degeneration by nutritive insufficiency, since the neoplastic growth causes extremely large follicles.
However, this same central degeneration could have been caused by the polarization of the nuclei at the cellular end facing the stellate reticulum. This probably causes the cells of the peripheral layer of the follicles to remove nutritive elements from the interior of these cellular islands and not from the connective tissue facing the other cellular extremity. This nutritive competition can cause metabolic deficiencies for the cells of the stellate reticulum, which can explain the degeneration of the central cells of the islands and the consequent formation of cystic cavities in its interior (Figure 3D).
6, The role of connective stroma: The influence of dense fibrous connective tissue around the epithelial islands, as well as the possible morphological and functional changes occurring in the cells of the peripheral layer of the follicles, could be responsible for the squamous metaplasia often observed in the ameloblastoma (Figure 3D). These alterations are able to impart characteristics of the basal cell layer to the peripheral cells of the follicle (Figure 3E). In fact, according to Kollar (1972), the epithelium of the enamel organ, when transplanted to regions which are normally covered by stratified squamous epithelium and have their own densely fibrous lamina propria or dermic layer - for example, the oral diastema and the plantar region of the rat paw - tends to grow as squamous stratified epithelium, loosing its ability to form dental tissues.
In conclusion, it can be said that ameloblastoma is probably no more than epithelial cells remaining from the embryonic phase, reactivated from their latent state and engaged in vain attempts to resume their original functions - an obligation that they, having been awakened in a different world and in another time, are not able to conclude.
Kollar EJ: Histogenetic aspects of dermal-epidermal interactions. In: Developmental aspects of oral biology. Slavkin HC & Bavetta LA ed. Chapter 7: 125-149. Academic Press, New York 1972
Lucas RB, Thackray AC: Cyst formation in adamantinomata. British Dent J 93: 62-65, 1952
Shafer WG, Hine MK Levy BM: A textbook of oral pathology. 4th ed. Saunders, Philadelphia 1983
Slavkin HC: Intercellular communication during odontogenesis. In: Developmental aspects of oral biology. Slavkin HC & Bavetta LA ed. Chapter 9: 165-199. Academic Press, New York 1972
Slavkin HC: Molecular biology of dental development: a review. In: Biological mechanisms of tooth eruption (An international conference). Davidovitch Z ed. EBSCO Media, Birmingham 1988
Spouge JD: Oral pathology. Mosby, Saint Louis 1973
Willis RA: Pathology of tumours. Butterworth, London 1948
Correspondence: Dr. Geraldo Maia Campos, Departamento de Estomatologia, Faculdade de Odontologia de Ribeirão Preto, Universidade de São Paulo, 14050 Ribeirão Preto, SP, Brasil.
Accepted September 27, 1990