EVOLUTION & DEVELOPMENT

3:2, 104–108 (2001)

Teeth outside the mouth in teleost fishes: how to benefit from a developmental accident Jean-Yves Sire CNRS UMR 8570, Université Paris 7, Case 7077, 75251 Paris Cedex 05, France Correspondence (email: [email protected])

SUMMARY Evolution proceeds by the selection of characters that enhance survival rates so that the long-term outcome for a species is better adaptation to its environment. These new characters are “accidentally” acquired, mostly through mutations leading to modifications of developmental events. However, changes that lead to the ectopic expression of an organ are rare and, whereas their subsequent selection for a new role is even more rare, such a scenario has apparently occurred for denticles in some teleost fish. Small, conical denticles are present, mainly on the dermal bones of the head, in a few, unrelated lineages of living teleosts. Here, I show that the morphology and structure of the denticles in

Atherion elymus, an atheriniform, is similar to those of teeth inside the oral cavity. These denticles are not derived evolutionarily from odontodes of early vertebrates, nor do they represent a re-expression as such (i.e., a long-lasting ability to make odontodes outside the oral cavity). Teeth and odontodes are homologous organs but the origin of the denticles is to be found in teeth, not in odontodes. The denticles are simply teeth that form outside the mouth, probably derived from a sub-population of odontogenically pre-specified neural crest cells. These “accidental” extra-oral teeth have arisen independently in these lineages and were selectively advantageous in a hydrodynamic context.

INTRODUCTION

RESULTS AND DISCUSSION

Throughout the second half of the twentieth century, experiments involving grafting (including tissue recombinations), ectopic expression of genes, and mutagenesis have produced numerous chimeras, monsters, and mutants that show the development of organs in ectopic locations, their duplication, or their absence. Initially, such experiments have provided information on tissue interactions controlling organogenesis (e.g., epithelial-mesenchymal interactions); now, most efforts concentrate on elucidating the function of the numerous genes involved in the control of development. In nature, ectopic expression of organs (i.e., organs that are expressed out of place) is rare. The occurrence of dermal denticles (teeth) outside the oral cavity in some lineages of living teleost fish demonstrates that such a natural developmental experiment is possible and may even be rather easily achieved because it has been replicated in several, unrelated lineages. Not only have teeth been expressed in extra-oral locations, but also these new phenotypes have been maintained during subsequent evolution, suggesting that these denticles most likely represent a selective advantage for their owners. The finding of a new teleost lineage (the fourth) with extraoral denticles provides an opportunity to consider some evolutionary, developmental, and functional implications of this phenomenon.

Atherion elymus, a small (not exceeding 40 mm in standard length), pelagic, atheriniform teleost, possesses numerous denticles attached to the surface of dermal bones of the head (Fig. 1). A detailed morphological and structural description of the Atherion denticles is out of the scope of this article, but the main features are briefly reported herein. In Atherion elymus, the denticles are particularly numerous on the whole front of the snout, the chin, and the undersides of the lower region of the head, where they are aligned forming a denticulate keel. Like teeth located in the oral cavity and odontodes (Fig. 2), these denticles consist of a pulp cavity surrounded by dentine, itself covered by enameloid (Fig. 3). The denticles are held onto the bony surface by means of ligaments and bone of attachment. These features make them identical to teeth and odontodes. Moreover, the denticles project beyond the skin, as do teeth, and are certainly replaced as indicated by the observation of several buds in serial sections. These findings add to previous descriptions of similar dermal denticles in three other teleost lineages, namely swordfish and billfish (xiphiids and istiophorids, perciforms), Denticeps clupeoides (a clupeomorph), and armored catfish (callichthyids and loricariids, siluriforms) (reviewed in Huysseune and Sire 1998). In the armored catfish, dermal denticles are attached to all bony elements, including fin rays

© BLACKWELL SCIENCE, INC.

104

Sire

Teeth outside the mouth in teleost

105

Fig. 2. Longitudinal section of an odontode ornamenting the scale surface of the living coelacanth, Latimeria chalumnae (after Roux 1942, modified).

Fig. 1. Scanning electron micrographs of the anterior region of the head of Atherion elymus (a) and detail of the teeth, white arrow (b) and of the denticles on the chin, black arrowhead (c). Scale bars ⫽ 100 ␮m. (Donation of the Smithsonian Institution, Natural Museum of Natural History, USNM 139865-140, Marianas Islands)

and scutes of the post-cranial dermal skeleton. The dermal denticles develop in the same way as do teeth (Sire and Huysseune 1996; Sire et al. 1998), and we may infer that identical morphogenetic mechanisms underlie their formation. The identity of these extra-oral denticles being unquestionable, their evolutionary and developmental origin and biological role remain to be explained.

Are the dermal denticles derived from odontodes of early vertebrates? Tooth-like elements developing outside the mouth are not a novelty for vertebrates. On the contrary, dermal denticles (called odontodes) are known to have been present either free or fixed to a bony support at the body surface of a number of jawless vertebrates, 450 to 500 Mya, long before jaws appeared. In jawed vertebrates, the embryonic tissues responsible for odontode morphogenesis and differentiation have certainly colonized the oral cavity, and gave rise to oral teeth that have been conserved until now in most vertebrate lineages. In chondrichthyans (sharks and rays), isolated extra-oral odontodes have been retained on the body surface whereas in the osteichthyan lineage, the odontodes have either been lost or progressively modified into odontocom-

plexes (Ørvig 1977; Reif 1982), then into elasmoid scales (Sire 1989; see review in Huysseune and Sire 1998). The available fossil record does not indicate that early teleosts possessed isolated extra-oral odontodes (the teleost level of evolution was reached by the end of the Triassic, but the group has diversified at the base of the Upper Cretaceous, c. 100 Mya; Carroll 1988). However, extra-oral denticles have been reported in related extinct species of three out of the four lineages: a denticipid from the Late Oligocene of East Africa (Greenwood 1960); denticulate rostra of xiphioids from the Eocene of the Mississippi (Fierstine and Applegate 1974); and fossil armored catfish (callichthyids or loricariids) from the Eocene/Oligocene of South America (Bardack 1961; Malabarba 1988). These records suggest that the origin of the extra-oral denticles in the living teleost taxa is not older than the Upper Cretaceous, when the modern teleost families diversified. Somewhere between basal actinopterygian lineages possessing odontodes (polypterids, lepisosteids) and early teleosts, the skin lost the ability to develop isolated odontodes. Thus, the dermal denticles in living teleosts cannot have evolved directly from ancestral odontodes. Moreover, an optimization of this character indicates that extra-oral dermal denticles have been acquired independently in the four teleost lineages (Fig. 4). Are the dermal denticles a re-expression of ancestral odontodes or simply teeth developing outside the mouth? The assertion that organs (odontodes), which had been lost from the skin of the ancestors, could be re-expressed in the skin of various descendant lineages is not in conflict with Dollo’s law (1893) on the irreversibility of evolution. This law concerns the inability of an organism to return to a state identical to an ancestral condition. Thus, given the evolutionary relationships between teeth sensu stricto and odontodes, the latter are still present in teleosts as teeth in the oral cavity. Due to the strong selective pressure on oral teeth, the genetic potential to produce them has been conserved until now in the genome of most vertebrates (except in turtles and birds). Re-expression of odontodes outside the mouth would

106

EVOLUTION & DEVELOPMENT Vol. 3, No. 2, March–April 2001

processes are thought to have been the privileged targets for spatial and temporal modifications that have occurred in the dermal skeleton during a very long period of geological time (Reif 1982; Smith and Hall 1993). Once the odontodes were lost, or were severely modified to form odontocomplexes and other dermal skeletal derivatives, it is difficult to understand how the underlying developmental mechanisms could have led to a re-expression of isolated odontodes similar to those present in the ancestors, and independently so in four teleost lineages, probably at different times during evolution. The denticles in these teleosts, therefore, cannot be considered to be direct re-expressions of ancestral odontodes; they are simply teeth being expressed in extra-oral locations. This assumption is supported by the strong morphological and structural similarities between oral teeth and dermal denticles on the head as illustrated here in Atherion (Figs. 1b and c). How can teeth form outside the mouth in teleosts? From experiments in amphibians and in the mouse, we know that the information for identity and shape of teeth resides in dental mesenchyme that originates from neural crest cells

Fig. 3. One ␮m-thick section of teeth pointing attached to the premaxillary bone and pointing toward the oral cavity (a) and of denticles fixed to the mandibular bone and pointing outside (b). The enameloid is not visible on this EDTA-decalcified material. Abbreviations: ba, attachment bone; d, dentine; l, ligament; m, mandibular bone; p, pulp cavity; pm, premaxillary bone. Scale bar ⫽ 100 ␮m.

mean that all the odontogenic factors in the skin of early osteichthyans would have been re-activated after a quiescence of more than a 100 My. However, whereas odontodes in the osteichthyan lineage have been modified into odontocomplexes and then into scales, extra-oral denticles in living teleosts morphologically more resemble oral teeth rather than ancestral odontodes. Because the factors underlying odontode and denticle development are the same as those in tooth development, reciprocal interactions between epithelial cells and a particular mesenchymal cell population from neural crest origin (ectomesenchymal cells) must be involved to generate dermal denticles. The epithelial-mesenchymal interactions are under the control of a large number of genes that are temporally and spatially regulated (Thesleff and Nieminen 1996; Peters and Balling 1999). These regulative

Fig. 4. Optimization of the denticle character for living actinopterygians. The position of the four teleost lineages possessing denticles indicates that these have been acquired independently.

Sire

(NCCs) (Lumsden 1988). Recently, using a two-component genetic system in the mouse, Chai et al. (2000) have demonstrated that cranial NCCs contribute to the formation of dental mesenchyme, dental papilla, odontoblasts, cementum, and periodontal ligament. Are these cranial NCCs committed to an odontogenic fate before or during migration, or do they only become specified after migration is complete? In amphibians, it appears that determination of cranial NCCs is acquired before migration, and Graveson et al. (1997) have shown that the odontogenic potential extends to the most rostral trunk NCCs. In mammals, however, the issue remains controversial, the cranial NCCs being thought either to be already pre-specified (Ruch 1985; Sharpe 1995) or still unspecified when they enter the mandibular arch (Lumsden 1988). In this latter study, premigratory trunk NCCs have been shown to be able to form tooth-like material. Lumsden suggests that the odontogenic potential is not restricted to the NCCs of presumptive tooth-forming levels, but that this potential does require an interaction with region-specific epithelium (i.e., the oral epithelium). In contrast, Graveson et al. (1997) propose that the trunk NCCs in amphibians and mice have conserved the odontogenic potential from their common ancestor (a basal sarcopterygian); in other words, trunk NCCs would be able to produce odontodes or odontocomplexes in amphibians and mice, but the basal-layer cells of the epidermis have lost the ability to interact with them. However, in the absence of specific markers, it is not possible to check if the tooth-like material obtained in both experimental studies are either odontodes or teeth. The second hypothesis is supported by the presence of teeth inside the mouth cavity and the presence of odontodes anywhere on the body in early osteichthyans and in living chondrichthyans. Indeed, this indicates that in these basal vertebrate taxa, (1) cranial and trunk NCC subpopulations with an odontogenic specification are able to colonize most regions of the skin, and (2) that these populations have acquired this specification prior to (or during early) migration. A similar postulate has been proposed by Smith and Hall (1993): “. . . neural crest is necessary for induction, morphogenesis, and differentiation of odontodes wherever they occur on the body.” For teleosts, nothing is known concerning the time of commitment of NCCs to an odontogenic fate. The presence of teeth outside the mouth in four teleost lineages suggests that the odontogenic NCC population could be specified prior to, or during, early migration. This NCC population could subsequently divide into two or several subpopulations that colonize the jaws and other regions of the skin to give rise to teeth inside the mouth and denticles outside the mouth. This also suggests that the basal-layer cells of the epidermis have conserved the ability to interact with these NCCs as does the oral epithelium. This leads to the following evolutionary hypotheses. In

Teeth outside the mouth in teleost

107

early, jawless vertebrates, cells from a NCC population that has acquired an odontogenic potential were able to colonize various regions of the body skin and interact with the basallayer cells of the epidermis to form odontodes. In early gnathostomes, these NCCs colonized the jaw mesenchyme, interacted with stomodeal ectoderm, and gave rise to teeth. This condition (teeth inside and odontodes outside the mouth) has been conserved until now in chondrichthyans. However, the morphology and arrangement of the odontodes in sharks and rays strongly differs from that of teeth. This indicates different ways of evolution for both organs, probably in relationship with different selection pressures, inside and outside the mouth. In osteichthyans, several changes have occurred in the dermal skeleton: outside the mouth the odontogenic ability has either been lost (in the mesenchyme or in the epidermis), or it has been subjected to transformations (giving rise to various types of scales, scutes, etc.). In some unrelated teleost lineages, the premigratory odontogenic NCC population responsible for teeth formation in the mouth has again divided into subpopulations, and these have re-colonized the dermis in various regions of the body. These cells have then interacted in each of these regions with the epidermal basal cells to form extra-oral teeth similar to those in the mouth. The existence of at least four unrelated teleost lineages in which these “new” pathways of odontogenic cells have occurred and have been selected for at various times during evolution illustrates the well-known potential of teleosts to generate new phenotypes. The question of the selective advantage of these dermal denticles can now be addressed. Why have these “accidental” extra-oral structures been selected for during evolution? Can a biological role be attributed to these denticles? A hydrodynamic biological role for extra-oral teeth? Given their location outside the mouth cavity, the denticles obviously do not serve the same role in facilitating ingestion as do teeth. In xiphioids, the denticles are small, oriented upward, and located on the surface of the rostrum only (in larvae and small, young specimens they are identical to teeth in that they are sharp and fixed on the bone surface, whereas in adults they differ in being rounded and ankylosed into the bone surface). In the armored catfish, the denticles are either small or large, pointed backward, and located on all the bony elements of the dermal skeleton (head, fin-rays, and scutes), and in Denticeps and Atherion they are small, pointing backward, and located on the dermal bones of the head only. Moreover, these extra-oral structures are slightly movable due to the presence of unmineralized ligaments. By their shape, size, and orientation, these denticles resemble certain odontodes covering the body of sharks and rays. In sharks, the odontodes have been experimentally demonstrated to be useful devices for improving hydrodynamics, facilitating the

108

EVOLUTION & DEVELOPMENT Vol. 3, No. 2, March–April 2001

penetration in the water by lowering the frictional drag (Burdak 1986). Such a hydrodynamical function can be postulated for the extra-oral denticles in teleosts. We can easily imagine the advantage that represents improvement of swimming performance for a small aquatic organism and why, in these four lineages, the “accidental” presence of teeth outside the mouth has been turned into an advantage during evolution. Acknowledgments I thank Prof. Ann Huysseune (Ghent University), Prof. J. Castanet (Université P. & M. Curie), Dr. M. Whitear (Tavistok, UK), Dr. J. Cubo, and Dr. M. Laurin (UMR 8570) for their comments on a previous draft. I am grateful to the two referees for their constructive remarks. Richard P. Vari and Susan L. Jewett (Smithonian Institution, Washington, DC) are acknowledged for donation of the Atherion specimens (USNM 139865).

REFERENCES Bardack, D. 1961. New Tertiary teleosts from Argentina. Am. Mus. Novit. 2041: 1–27. Burdak, V. D. 1986. Morphologie fonctionnelle du tégument écailleux des poissons. Cybium 10, suppl: 1–147. Carroll, R. L. 1988. Vertebrate Paleontology and Evolution. W.H. Freeman and Cie, New York, p. 698. Chai, Y., Jiang, X., Ito, Y., et al. 2000. Fate of mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 127: 1671–1679. Dollo, L. 1893. Les lois de l’Evolution. Bull. Soc. Belge Geol. 8: 164–166. Fierstine, H. L., and Applegate, S. P. 1974. Xiphiorhynchus kimblalocki, a new billfish from the Eocene of Mississipi with remarks on the systematics of Xiphioid fishes. Bull. Southern Calif. Acad. Sci. 73: 14–22. Graveson, A. C., Smith, M. M., and Hall, B. K. 1997. Neural crest potential

for tooth development in a urodele amphibian: developmental and evolutionary significance. Dev. Biol. 188: 34–42. Greenwood, P. H. 1960. Fossil denticipid fishes from East Africa. Bull. Brit. Mus. Geol. 5: 3–11. Huysseune, A., and Sire, J. Y. 1998. Evolution of patterns and processes in teeth and tooth-related tissues in non-mammalian vertebrates. Eur. J. Oral Sci. 106, suppl: 437–481. Lumsden, A. G. S. 1988. Spatial organization of the epithelium and the role of neural crest cells in the initiation of the mammalian tooth germ. Development 103, suppl: 155–169. Malabarba, M. C. De S. L. 1988. Loricarid dermal plate and pectoral fin spines from the Tertiary of Sao Paulo, Brazil (Osteichthyes, Siluriformes). Comun. Mus. Cienc. PUCRS, Ser. Zool., Porto Alegre 1: 5–12. Ørvig, T. 1977. A survey of odontodes (“dermal teeth”) from developmental, structural, functional, and plyletic points of view. In S. M. Andrews, R. S. Miles, and A. D. Walker (eds.). Problems in Vertebrate Evolution. Linnean Society Symposium Series no. 4, pp. 53–75. Peters, H., and Balling, 1999. Teeth: where and how to make them? Trends in Genet. 15: 59–65. Reif, W. E. 1982. Evolution of dermal skeleton and dentition in vertebrates. Evol. Biol. 15: 287–368. Roux, G. H. 1942. The microscopic anatomy of the Latimeria scale. S. Afr. J. Med. Sci. Suppl. 7: 1–18. Ruch, J. V. 1985. Odontoblast differentiation and the formation of odontoblast layer. J. Dent. Res. 64: 489–498. Sharpe, P. T. 1995. Homeobox genes and orofacial development. Connect. Tissue Res. 32: 17–25. Sire, J.-Y., 1989. The scales in young Polypterus senegalus are elasmoid: new phylogenetic implications. Am. J. Anat. 186: 315–323. Sire, J.-Y., and Huysseune, A. 1996. Structure and development of the odontodes in an armoured catfish, Corydoras aeneus (Callichthyidae). Acta Zool. (Stockholm) 77: 51–72. Sire, J.-Y., Marin, S., and Allizard, F. 1998. A comparison of teeth and dermal denticles (odontodes) in the teleost Denticeps clupeoides (Clupeomorpha). J. Morphol. 237: 237–255. Smith, M. M., and Hall, B. K. 1993. A developmental model for evolution of the vertebrate exoskeleton and teeth. Evol. Biol. 27: 387–448. Thesleff, I., and Nieminen, P. 1996. Tooth morphogenesis and cell differentiation. Curr. Opin. Cell Biol. 8: 844–850.