Читать книгу Oral Cells and Tissues - Philias R. Garant - Страница 8

Early Tooth Development

Teeth are formed from oral epithelium, in the form of a dental lamina, and neural crest ectomesenchyme of the maxillary and mandibular processes (Fig 1-1). The oral epithelium contributes the enamel component, and the ectomesenchyme contributes the dentin and cementum components of the fully formed tooth. Although the initiating events that trigger downgrowth of the oral epithelium to form a dental lamina are incompletely understood, it is known that neural crest ectomesenchyme is necessary.^1,2 Early reciprocal inductive interactions between the oral epithelium and the underlying ectomesenchyme, and subsequent interactions between the enamel organ and dental papilla, coordinate the sequential events of tooth development.^3–5

Efforts to understand the instructional signals that originate in each of these interacting tissues have been ongoing for more than 50 years.^6,7 Most investigations have been performed with dental tissues obtained from embryonic mice and rats or with the continuously growing incisor teeth of adult mice and rats. Organ culture techniques have been perfected to study the growth of dental tissues in chemically defined media, to observe the results of various epithelial-mesenchymal combinations, and to examine the effects of various growth factors on odontogenesis. Thus, nearly all current insight into the regulatory mechanisms of tooth development has come from studies of animal models, often from tooth buds grown in organ culture.

This chapter contains a discussion of the initiation of tooth formation and the histodifferentiation of the enamel organ and dental papilla. Subsequent chapters will examine the cytodifferentiation of dentin-and enamel-forming cells and the secretion and mineralization of their respective matrices.

Role of the Neural Crest

Early in embryogenesis, soon after the neural tube forms by invagination of the overlying ectoderm, migratory pluripotent neuroepithelial cells, the neural crest cells, migrate from the dorsal midline region of the neural tube.⁸ In exiting from the neural tube, neural crest cells lose their epithelioid characteristics and assume a mesenchymal phenotype capable of directed cell migration. Cranial neural crest cells invade the developing branchial arches and, in a series of reciprocal inductive interactions with early oral epithelium, form tooth primordia (Figs 1-1 and 1-2). When the movement of dye-injected neural crest cells was traced in organ cultures of developing dental arches, it was shown that neural crest cells from the posterior midbrain, and to a lesser extent from the anterior hindbrain, form dental ectomesenchyme.⁹ The failure of neural crest ectomesenchymal cells to migrate normally to appropriate sites during craniofacial development leads to serious developmental defects, including the absence of teeth (anodontia) and underdeveloped jawbones (micrognathia).

Fig 1-1 Stages in the development of a tooth bud. (A) Oral epithelium and the underlying ectomesenchyme and mesenchyme during the development of the dental lamina (DL). (B) The enamel organ arises from a genetically determined site of the dental lamina by cell proliferation. The dental papilla develops from ectomesenchymal cells of neural crest origin.

Fig 1-2 Histologic section of a developing tooth at early bell stage. (DL) Dental lamina; (DP) dental papilla; (DS) dental sac; (EO) enamel organ; (M) mesenchyme; (OE) oral epithelium; (SL) successional lamina. (Hematoxylin-eosin stain. Original magnification × 220.)

Subsets of cranial neural crest cells give rise to chondrocytes, osteoblasts, periodontal ligament fibroblasts, cementoblasts, and odontoblasts. Final phenotype differentiation is regulated by interaction of the ectomesenchymal cells with extrinsic factors, such as growth factors, and substrate adhesion molecules in the local microenvironment.¹⁰ It has been suggested that there may be separate populations of neural crest cells for each tooth type. The molecular code for each tooth type appears to reside in specific sets of homeobox genes.^11,12

Development of the Dental Lamina, Enamel Organ, and Dental Papilla

The first evidence of tooth formation in humans is observed as a thickening of the oral epithelium in the mandibular, maxillary, and medial nasal processes in the 1-month-old fetus (Figs 1-3 to 1-5). It has been suggested that the zone of epithelial thickening (the dental plate or placode) contains the genetic determinants for the initiating signals that regulate the number and position of the future tooth buds. Experiments with epithelial-mesenchymal tissue recombination have shown that early-stage oral epithelium is capable of inducing tooth development in non-oral ectomesenchyme.^13–15 When non-oral epithelium is used in the recombination, only bone and cartilage form in the ectomesenchyme. Mouse oral epithelium has been shown to induce biochemical markers of early tooth development in chick oral ectomesenchyme, a tissue thought to have lost its ability to form teeth.¹⁶ The results of these studies suggest that the oral ectoderm contains instructional signals for tooth development and perhaps the prepattern for the entire dentition. Weiss et al¹⁷ suggested that a very early signaling system (prior to neural crest migration) involving Shh and Pax6 genes might form the basis of epithelial patterning mechanisms for tooth development.

Formation of the dental lamina

At a slightly later stage of development (11- to 14-mm embryos), the epithelium invaginates into the underlying mesenchyme to form the dental lamina. This process begins in the distal (molar) region and later in the midline. In 15- to 20-mm human embryos, the dental lamina shows signs of additional differential growth, reflecting the determination of incisor, canine, and molar domains (see Figs 1-4 and 1-5). Deep notches in the dental lamina are present between the incisor and canine domains, especially in the mandible. Continued site-specific enlargement of the dental lamina, along with condensation of neural crest ectomesenchyme, gives rise to the individual tooth buds.

Fig 1-3 Facial region of a human embryo. (LNP) Lateral nasal process; (MNP) medial nasal process; (MP) maxillary process; (MdP) mandibular process; (CRL) crown-rump length. (Adapted from Ooe⁷⁴ with permission.)

Fig 1-4 Degree of oral epithelial thickening in various human embryos ranging from 10- to 20-mm crown-rump length (CRL). Note the undulating character of the undersurface of the epithelium. (Adapted from Ooe⁷⁴ with permission.)

Fig 1-5 Model of the reconstructed oral epithelium of the mandible in a 16-mm human embryo. The “swellings” correspond to the sites of early development of the future primary central incisor (i1), lateral incisor (i2), canine (c), and molar (m) tooth buds. (Adapted from Ooe⁷⁴ with permission.)

Role of homeobox genes

Recent studies of the role of homeobox genes indicate that the expression of these genes in ectomesenchymal tissues may control the development and ultimate shape of the tooth.^11,18–20 Homeobox genes constitute a large family of genes that specify correct positioning of body parts during embryonic development. These genes are implicated in determining axial patterns, such as the anteroposterior development of limbs. All members of this family share a common code for a 60–amino acid DNA-binding sequence (the homeodomain) that allows the protein to act as a gene regulatory factor. Homeobox genes (Dlx, Pax, Msx, etc) are widely expressed in embryonic craniofacial tissues. Whiting²¹ has reviewed their role in normal development as well as the developmental defects that result from mutations.

Studies of tooth development in mice that have mutant homeobox genes support the idea that regional expression of various homeobox genes may provide the positional information for the type of tooth to be formed.¹¹ The results of these studies indicate that mutations in Dlx1 and Dlx2 genes prevent maxillary molar development but have no negative effect on maxillary incisor development. Incisor development is regulated by Msx1 and Msx2 homeobox genes. Thus, according to Thomas et al,¹¹ the odontogenic pattern (ie, tooth type and position in the arch) is determined by early regional and restricted expression of various combinations of homeobox genes. Once the tooth buds are formed, the homeobox genes are activated in a more generalized pattern. The presence of Msx1 is required for progression of molar tooth development beyond the bud stage.^20,22

Karg et al²³ described the localization of the homeobox gene, S8 (Prx2), in the dental papillae of developing mouse incisor and molar tooth buds. Because the highest level of S8 expression occurs during the growth of the dental papilla, it was suggested that S8 might take part in regulating the overall growth of the developing tooth. At the cap stage of tooth development, epithelial growth centers (enamel knots) regulate the cuspal outline of the developing tooth by coordinating cell proliferation within the enamel organ and dental papilla through the secretion of growth factors.^24,25

Progress in research on gene expression in tooth development can be found on the Internet at http://bite-it.helskini.fi.²⁶

Histogenesis of the tooth

The enamel organ develops by proliferation of cells in the dental lamina. The adjacent ectomesenchymal cells continue to proliferate and concentrate to form the dental papilla and dental sac (see Fig 1-2). During this coordinated growth, various growth factors and regulatory proteins are exchanged between the epithelium and ectomesenchyme.

During the early stage of tooth development, the enamel organ, shaped like a cap, is superimposed over a condensation of ectomesenchymal cells (Figs 1-2, 1-6, and 1-7a). At the cap stage, the enamel organ is subdivided into four regions: the outer enamel epithelium (OEE), the stellate reticulum (SR), the stratum intermedium (SI), and the inner enamel epithelium (IEE) (see Fig 1-6).^27–30 Later in development, the enamel organ is bell shaped, encompassing a well-defined dental papilla along its concave internal surface (Fig 1-7b).

The cells of the OEE are cuboidal and separated from the adjacent dental sac ectomesenchyme by a basement membrane. Along their concave surface, they contact the star-shaped cells of the SR. The cells of the SR are separated by wide intercellular spaces. Adjacent SR cells remain in contact via long cytoplasmic folds joined by numerous desmosomes and gap junctions (see Fig 1-6). The intercellular spaces of the SR contain hyaluronan and chondroitin sulfates that bind large amounts of water.³¹ The SR retains its hydrated state until the initiation of enamel formation; thereafter, the SR and the OEE differentiate into the papillary layer (described in chapter 3).

The SI consists of one or two layers of low cuboidal cells situated between the SR and the IEE (see Fig 1-6). A clearly defined SI is established between the SR and the IEE just prior to the differentiation of the ameloblasts. The cells of the SI and IEE express similar enzyme patterns, suggesting that both cell types have common metabolic functions.

The cells of the IEE are juxtaposed to the ectomesenchymal cells (preodontoblasts) of the dental papilla (Figs 1-6 and 1-8). The basement membrane beneath the IEE consists of a basal lamina densa and many aperiodic fibrils (see Fig 1-8). The nature of these fibrils and their significance in odontoblast differentiation are discussed in chapter 2.

Fig 1-6 Enamel organ and dental papilla. The outer enamel epithelium (OEE) forms the convex surface of the enamel organ and is separated from adjacent dental sac (DS) cells and general mesenchyme (not shown) by a basement membrane. The stellate reticulum (SR) lies between the OEE and the stratum intermedium (SI). The SI cells are closely juxtaposed to the cells of the inner enamel epithelium (IEE). The enamel knot (EK) represents a small group of nondividing cells near the IEE. The IEE is separated from the preodontoblasts (PO) of the dental papilla by a basement membrane (see Fig 1-8). (DL) Remnant of the dental lamina.

Cytodifferentiation of odontoblasts and ameloblasts starts at the tip of the future cusps. Under the influence of stimuli originating from the IEE, the preodontoblasts begin differentiation. In turn, they stimulate the cells of the IEE to undergo differentiation to form a single layer of enamel matrix–secreting cells, the ameloblasts. Preodontoblasts reach maturity as secretory odontoblasts before the preameloblasts mature into secretory ameloblasts. Regulatory control of cell proliferation and the differentiation of ameloblasts and odontoblasts is provided in part by complex sequential interactions involving cell membrane receptors, growth factors, and/or matrix molecules concentrated in the IEE basal lamina. Recent research has begun to define regulatory signals in tooth development at the level of gene activation.^32,33

Figs 1-7a and 1-7b Three-dimensional reconstructions of enamel organs made from serial sections of human embryos. Dental papilla and mesenchyme not shown. (Adapted from Ooe⁷⁴ with permission.)

Fig 1-7a Cap stage.

Fig 1-7b Bell stage.

Fig 1-8 Role of basement membrane components at the junction between the preameloblast (PA) of the inner enamel epithelium (IEE) and the adjacent preodontoblast (PO). A basement membrane consisting of a lamina densa (LD) and aperiodic fibrils (APF) separates the two tissues. The POs extend cell processes toward the APFs. (SR) Stellate reticulum; (SI) stratum intermedium.

Epithelial-Ectomesenchymal Morphogenetic Regulation of Odontogenesis

During the 1930s, the science of experimental embryology developed hand-in-hand with advances in organ culture technology. It soon became possible to grow whole and disassociated tooth buds in vitro. Enamel organs, when separated from the dental papillae by trypsin digestion of the basement membrane, were cultured alone or in various recombination with non-oral mesenchymal tissues (Figs 1-9 and 1-10). Isolated cap stage enamel organ, grown either in vivo as a transplant or in vitro in an organ culture system, failed to produce ameloblasts. Dental papilla cells failed to differentiate into odontoblasts unless grown in contact with the enamel organ. These studies established the need for contact between the epithelium (enamel organ) and the ectomesenchyme (dental papilla) as a preliminary condition for the differentiation of ameloblasts and odontoblasts. It was also observed that the dental papilla, once established, controlled the shape of the tooth and gained the ability to direct the differentiation of overlying epithelium (see Figs 1-9 and 1-10).^34–36

When it was discovered that the odontogenic inductive interaction could take place across a thin, porous filter, the search for diffusible soluble factors responsible for inducing the differentiation of ameloblasts and odontoblasts became the mission of several dental researchers. In the late 1960s and early 1970s, as the science of molecular biology was being developed, it was speculated that the transfer of informational messenger ribonucleic acid (mRNA) across the basement membrane might control the differentiation of odontogenic cells. In the 1970s, electron microscopic studies showed that cell-to-cell contacts were formed between preodontoblasts and preameloblasts during the cytodifferentiation stage of tooth development. It was proposed that such contacts might provide informational clues responsible for initiating differentiation. Because additional evidence in support of these hypotheses was not forthcoming, attention was directed to the extracellular matrix as a potential communication link between the enamel organ and the dental papilla. This premise was supported by the apparent importance of the basal lamina during odontoblast differentiation.

Fig 1-9 Control of tooth shape by the dental papilla (DP). Dissociation of the enamel organ from the dental papilla by low calcium and trypsin digestion of the basement membrane makes it possible to study the development of various recombinations. Organ cultures of recombined tissues demonstrate the controlling influence of ectomesenchyme (dental papilla) on final tooth form. (EO) Enamel organ. (Based on the findings of Kollar and Baird.^34,35)

Fig 1-10 Inductive action of mesenchyme on epithelial differentiation. Organ cultures of dental epithelium recombined with skin mesenchyme develop skin epidermis, complete with skin appendages. When skin epithelium is cultured in contact with dental mesenchyme, a tooth is formed, complete with enamel organ. These results demonstrate the inductive influence of mesenchyme on epithelium. (Based on the findings of Kollar.³⁶)

Role of matrix-mediated signaling

The discovery that enamel organs expressed amelogenin transcripts when cultured on a basement membrane gel, but not when grown on a laminin-coated filter, reinforced the concept that cell-matrix interactions had a permissive effect on gene transcription during tooth development. Research was soon focused on the interactions of cell membrane receptors with specific extracellular matrix ligands as important signaling events that might regulate odontogenic cell differentiation. These findings led Ruch et al to state:

Experimental data demonstrate that dental histomorphogenesis and cytodifferentiation are controlled by an alternative flux of information circulating between ectomesodermal and epithelial cells. They are matrix-mediated signals. The basement membrane is a dynamic, asymmetric interface demonstrating compositional and conformational modulations. The spatial pattern and timing of these changes result from specific activities of adjacent cells.⁴

Based on numerous in vitro experiments, Ruch et al⁴ proposed that basement membrane modifications are causally related to successive steps of odontogenesis. The following are the essential points of this hypothesis:

1. Time- and space-specific information is encoded in the basement membrane constituents.

2. This information is read by cell membrane receptor molecules of adjacent cells.

3. Receptor-ligand interactions act on the cytoskeleton and/or cytoplasmic enzymes, which subsequently influence transcriptional and posttranscriptional events.

To date, fibronectin, fibronectin receptors, tenascin, and syndecan have been implicated as participants in matrix-mediated signaling during Odontogenesis.

The distribution of cell adhesion molecules and substrate adhesion molecules as potential control factors in tooth development has been a subject of increasing interest. Syndecan, a proteoglycan cell adhesion molecule located in the cell membrane, is expressed prior to tooth formation in the ectomesenchymal cells that underlie the dental epithelium.³⁷ Tenascin, a large substrate adhesion molecule, is expressed in the ectomesenchyme during the downgrowth of the dental lamina and during the subsequent condensation of the dental papilla.³⁸ It has been proposed that the binding of membrane-bound syndecan molecules to extracellular tenascin molecules is responsible for the condensation of the ectomesenchymal cells.^37,39

An alternative explanation is that tenascin interferes with cell-to-fibronectin attachment, leading to decreased migration of the ectomesenchymal cells, causing them to aggregate in the form of the dental papilla. Adhesion of fibroblasts is weaker to fibronectin than to tenascin.⁴⁰ It has also been shown that when cells express syndecan they have a reduced ability to invade a collagen gel. Thus, the appearance of syndecan on the cell surface of ectomesenchymal cells may have a direct, negative effect on their ability to migrate, thereby causing them to form aggregates, such as the dental papilla.

Tissue separation and recombination studies have demonstrated that the expression of syndecan and tenascin in tooth ectomesenchyme is induced during specific epithelial-mesenchymal interactions.⁵ In situ hybridization studies indicate that mRNA for tenascin is expressed in high amounts in cells of the inner enamel epithelium and the preodontoblasts. Redundant pathways regulating cell condensation are undoubtedly present, because tooth development has been shown to proceed normally in mice lacking tenascin expression.⁴¹

Role of growth factors

Advances in organ culture technique have made it possible to grow developing teeth in chemically defined culture media. Yamada and coinvestigators⁴² demonstrated that explants of developing teeth could undergo complete cell differentiation and matrix mineralization in a chemically defined medium. They concluded that autocrine and paracrine factors coordinate the sequence of cellular differentiation events during tooth development. This stimulated the search for diffusible growth and regulatory factors that might be involved in odontogenesis.

Using chemically defined culture media, Chai et al⁴³ showed that tooth size and rate of development are regulated in part by transforming growth factor β2 (TGF-β2). When antisense oligonucleotides against TGF-β2 are added to tooth organ cultures, development is accelerated and the tooth buds grow larger than controls.⁴³ Addition of exogenous TGF-β2 reverses the effect of antisense nucleotides, leading to normal growth.

The advent of powerful molecular biologic approaches marked the beginning of a new era by discovery of the regulatory role of growth factors in dental morphogenesis. Thesleff and colleagues^5,33,44,45 have reviewed recent advances in this area of developmental biology. The earliest growth factor signal emanating from the presumptive dental lamina epithelium is bone morphogenetic protein 4 (BMP-4)^5,46 (Fig 1-11). Epithelial cells make BMP-4 until the cap stage, when the production of BMP-4 shifts to the condensed ectomesenchymal cells. Soon thereafter, a new bone morphogenetic protein (BMP-2) appears in the epithelial cells. These shifts in BMP expression may account for the transfer of instructional activity from the epithelium to the dental papilla ectomesenchyme at the cap stage.

It has been proposed that BMP-4 activates Msx genes in the adjacent ectomesenchymal cells⁴⁶ (see Fig 1-11). The Msx genes are “muscle segment” members of the homeobox genes (regulators of segmentation) that have been implicated as regulators of the mesiodistal axis of tooth bud placement. Msx gene products are believed to be transcription activators that regulate the expression of BMPs, syndecan, and peptide growth factors in the condensing ectomesenchyme (see Fig 1-11). At the bell stage, Msx2 is active in secondary enamel knots (EKs) and in the dental papilla.

Fig 1-11 Proposed model of molecular mechanisms in early tooth bud development, illustrating the role of bone morphogenetic protein 4 (BMP-4) in activating Msx gene expression and a cascade of differentiation within the underlying ectomesenchyme. With the activation of Msx genes, the inductive potential is transferred to the dental ectomesenchyme. Reciprocal interactions involving signaling growth factors, matrix molecules, and cell surface receptors regulate cell differentiation. Enamel knot signaling centers appear in the enamel organ prior to cusp formation. (FGF-8) Fibroblast growth factor 8. (Based on the findings of Vainio et al.^38,46)

Transcription products of Msx1 function during later stages of tooth development, possibly regulating the differentiation of ameloblasts and odontoblasts.²⁰ Animals that lack the Msx1 gene fail to develop teeth.²²

An especially important discovery was the identification of the enamel knot as a signaling center within the enamel organ.^24,25,47 The enamel knot, a component of the enamel organ previously believed to be unimportant, has achieved prominence as a potential regulatory center of cell proliferation involved in cusp formation. The EK is a small group of closely packed, nondividing cells located adjacent to the IEE, and, in a single-cusped tooth, close to the center of the enamel organ (Figs 1-6 and 1-12).

Fig 1-12 Possible role of the enamel knot (EK) in cusp formation. (arrows) Direction of growth. During the cap stage, the epithelium grows laterally around the dental mesenchyme. A single EK coordinates the development of the early cap stage. In multicusped teeth, secondary EKs are formed over future cusps to coordinate development during the late cap stage to the bell stage. (Adapted from Jernvall et al²⁴ with permission.)

The earliest sign of EK formation appears to be the localized expression of BMP-2 and BMP-7 in epithelial cells of the dental lamina and enamel organ. In situ hybridization techniques demonstrate that EK cells produce fibroblast growth factor 4 (FGF-4), several bone morphogenetic proteins (BMP-2, BMP-4, and BMP-7), and sonic hedgehog (Shh) protein.^26,27,48 Fibroblast growth factor 4 is a potent stimulator of epithelial and mesenchymal cell proliferation.²⁵ Epithelial and ectomesenchymal cells adjacent to the EK continue to divide in response to FGF-4, while the EK cells, which produce FGF-4, remain nondividing. The cells of the EK are retained in the G1 phase of the cell cycle by a high level of expression of the cyclindependent kinase inhibitor, p21. Bone morphogenetic protein 4 may regulate EK activity via its ability to sustain high levels of p21 expression.⁴⁷

By secreting growth factors, the EK promotes cell proliferation along a proximodistal axis, leading to the formation of a cusp. In this sense, the EK is akin to the apical ectodermal ridge that controls limb bud development. In establishing coronal form, embryonic dental tissues follow a pattern of polarized growth. Cells in the cervical loop proliferate and move away from older differentiating cells located nearer to the cusp tip.

The best example of polarized growth is found in the developing limb. The specific genes that participate in determining the anteroposterior axis of developing limbs are also expressed in cap to bell stage tooth buds. The Shh gene responsible for polarizing activity in developing limbs is active in the enamel knot (see Fig 1-12) and in differentiating odontoblasts and ameloblasts.⁴⁸ Proof that genes that regulate polarized growth, such as Shh, are active in the tooth bud was obtained when tooth buds were grafted to developing limbs. The grafted tooth buds induced the formation of additional digits, revealing a capacity for polarizing growth in an anteroposterior axis.⁴⁸

In multicusped teeth, secondary EKs are formed over the tips of the future cusps (see Fig 1-12). In mouse molar teeth, the EKs remain active for about 24 hours before undergoing apoptosis.⁴⁹ Programmed cell death is also responsible for the removal of the dental lamina after tooth bud formation.

Growth and Differentiation Factors That Regulate Tooth Formation

Bone morphogenetic factors, Shh, and FGFs are also important during the later stages of tooth development.⁴⁷ Both BMP-2 and BMP-7 are expressed in the IEE across from the differentiating odontoblasts, suggesting that they may have an inductive role. Secretory odontoblasts express BMP-4 and BMP-7, while BMP-5 appears to be restricted to fully differentiated ameloblasts. Bone morphogenetic protein 3 is localized in the cells of the dental follicle.

Activin A, a protein structurally related to BMPs and a member of the TGF-β superfamily of cytokines, has been implicated in signaling during tooth development.⁵⁰ Mice deficient in activin A have craniofacial abnormalities and failure of incisor tooth development.

Vitamin A and its metabolic derivatives, retinol and retinoic acid (RA), are essential regulators of epithelial cell proliferation and differentiation and have special impact on tooth development.^51–54 The importance of vitamin A in the initiation of tooth development was underscored by the observation that when endogenous vitamin A is blocked in vitro, the dental lamina fails to develop in organ cultures of mouse embryonic mandibles.⁵⁵ Early studies of the effect of vitamin A on tooth development showed that a deficiency of the metabolite leads to defective enamel and dentin.⁵⁶ In contrast, excessive vitamin A increases the chance for tooth bud fusion and/or the formation of supernumerary teeth.^57,58

In organ cultures of embryonic mandibular explants, retinol and retinoic acid increase epithelial proliferation and stimulate the formation of extra tooth buds. Retinoic acid exerts its effect by binding to nuclear transcription factors (RA receptors [RARs]) located near retinoid response elements on various target genes, one being the gene that produces epidermal growth factor (EGF) (Figs 1-13a and 1-13b).⁵⁹ Retinoic acid also increases the expression of midkine (MK) protein, a regulator of cell proliferation.

Cellular retinol-binding proteins (CRBPs) and cellular retinoic acid–binding proteins (CRABPs) are involved in the metabolism and storage of vitamin A metabolites in the cytoplasm. Cellular retinol-binding proteins and CRABPs may control the level of free RA available to interact with the nuclear RARs. Because a nuclear RAR and an RA response element control the gene responsible for coding EGF, the ability of RA to increase cell proliferation may be mediated through increased EGF production (see Figs 1-13a and 1-13b). The site-specific increase in epithelial cell division required for the formation of the dental lamina and the subsequent development of tooth buds could be controlled by localized production of EGF in response to RA.^51,53,54,60

Both RARs and CRABPs have been localized in the dental lamina and adjacent ectomesenchyme as well as in dental epithelium and ectomesenchymal components of developing teeth (see Figs 1-13a and 1-13b).^51,53,54 In addition, CRABPs have been localized in the epithelium adjacent to sites of dental lamina formation, suggesting that RA may be bound at such sites. In the dental lamina, where there appears to be fewer CRABPs, the RA molecules are free to interact with their nuclear receptors and thereby increase the expression of EGF.⁵⁴

Figs 1-13a and 1-13b Role of vitamin A during tooth formation.

Fig 1-13a Cellular action. Retinoic acid (RA), the major active metabolite of vitamin A, diffuses into the cell interior, where it binds to cellular retinoic acid–binding protein (CRABP), or, if the level of CRABP is low, may enter the nucleus to interact with its receptor (RAR). Retinoic acid receptors activate retinoic acid response elements (RARE) that regulate gene transcription, thereby stimulating the production of messenger ribonucleic acid (mRNA). The epidermal growth factor gene (Egf) is regulated by a RAR-RARE complex. The increase in cell proliferation effected by vitamin A is believed to be the result of the secretion of epidermal growth factor (EGF), a known mitogen for dental epithelium and ectomesenchyme. (CRBP) Cellular retinol-binding protein.

Fig 1-13b Tissue expression. Proposed model by which vitamin A can set the location of the dental lamina (DL). Cellular retinoic acid–binding proteins (CRABPs) expressed in epithelium adjacent to the DL limit the availability of retinoic acid (RA) for interaction with retinoic acid receptors (RARs), while the level of expression of CRABPs is low in the DL, permitting RA stimulation of epidermal growth factor (Egf) gene trans-cription in the DL and the adjacent ectomesenchyme (EM). (mRNA) Messenger ribonucleic acid; (EGF) epidermal growth factor.

Epidermal growth factor, acting in a paracrine or autocrine manner, appears to control the rate of cell proliferation in the early stages of tooth development. Epithelial cells of the dental lamina and early enamel organ express EGF receptor.⁶¹ When the enamel organ reaches the cap stage of development, the level of binding of EGF decreases in the epithelial cells but increases in the ectomesenchymal cells of the underlying dental papilla. The importance of EGF in tooth development is underscored by the observation that interfering with the synthesis of EGF blocks odontogenesis.⁶²

Another RA-regulated gene expressed during tooth development is midkine (MK).^63,64 This gene codes MK protein, a heparin-binding growth and differentiation factor unrelated to two other heparin-binding molecules, fibroblast growth factor, and hepatocyte growth factor. The MK gene and its product are preferentially located in embryonic tissues undergoing epithelial-mesenchymal interaction. Both MK mRNA and MK protein are preferentially expressed in all stages of developing maxillary and mandibular teeth of embryonic mice. The differential or appositional localization of MK mRNA and MK protein in developing dental ectomesenchyme and its receptor on the cells of the IEE provides an instructive example of epithelial-mesenchymal interaction (Fig 1-14). During the cap stage of tooth development, the MK protein is secreted by the ectomesenchymal cells and concentrated in the basal lamina. The MK protein binds to MK receptor, acting as a paracrine regulator of cellular activity in the IEE (see Fig 1-14).

Fig 1-14 Appositional pattern of the expression of the midkine (MK) gene in the ectomesenchyme (EM) and the localization of the MK protein (MKp) to the surface of the inner enamel epithelial cells adjacent to the basement membrane (BM) of a cap stage tooth bud. The diffusible MK protein is concentrated in the BM and is bound to cell surface receptors (MK-R) on epithelial cells, where it may act as a paracrine-signaling molecule. Although EM cells make MK protein, they appear to lack receptors. (IEE) Inner enamel epithelium; (EO) enamel organ; (DP) dental papilla. (Adapted from Mitsiadis et al⁶³ with permission from The Company of Biologists.)

Midkine appears to regulate cell proliferation, possibly by inhibiting cell division in preparation for cell differentiation. The significance of MK in tooth development is confirmed by the observation that antibodies to MK inhibit odontogenesis.^63,64 The highest levels of MK are observed in the IEE, its basal lamina, the dental papilla, and especially in differentiating odontoblasts. With the onset of dentin secretion, MK is no longer detectable in odontoblasts or in the differentiating preameloblasts.

Evidence continues to accumulate that reciprocal interaction via diffusible signaling molecules, as exemplified by MK, regulates epithelial-mesenchymal differentiation. A similar pattern of expression and localization has been reported for TGF-β, hepatocyte growth factor, and BMP during tooth development.

Neurotrophins and neurotrophin receptors are expressed in developing teeth in association with differentiating preameloblasts and preodontoblasts.^65,66 They are also expressed in the subodontoblastic layer. Neurotrophins play a central role in the development and maintenance of nerves. Recent studies suggest that neurotrophins are expressed in early dental epithelium before the developing teeth are innervated.⁶⁷ The presence of neurotrophins and their receptors in developing teeth, and their changing spatiotemporal distribution, suggest that, in addition to a role in dental neuronal development, they may have other non-neuronal regulatory functions. Evidence obtained in other developing organ systems has indicated that neurotrophin receptors also bind matrix molecules and could act in an adhesive capacity during cell migration and/or condensation.

Nerve growth factor is a ligand for the tyrosine kinase receptor A member of the neurotrophin receptor family. Nerve growth factor produced in the developing tooth may act locally to control the number of cell cycles in the IEE and dental papilla proliferation compartments. The expression of nerve growth factor receptor decreases as cell division in the IEE ceases prior to ameloblast differentiation.⁶⁸

Growth hormone, growth hormone–binding protein, and growth hormone receptor have been localized in developing teeth. Cells of the enamel organ and dental papilla appear to be targets for growth hormone. Increased staining for growth hormone and its receptor was observed in differentiating cells of the IEE and the preodontoblastic layer of the dental papilla.⁶⁹ Likewise, insulin-like growth factor is concentrated in the IEE and dental papilla during ameloblast and odontoblast differentiation.⁷⁰ Hepatocyte growth factor and its receptor are expressed in the dental papilla.⁷¹ Hepatocyte growth factor acts as a mitogen in regulating cell proliferation in the enamel organ and dental papilla. Antisense nucleotides to hepatocyte growth factor reduce mitotic activity in the IEE and dental papilla, leading to abnormal tooth development.

The neurotransmitter serotonin (5-hydroxytryptamine) is another potential morphogenetic signaling molecule. Specific uptake of serotonin occurs transiently in oral epithelium and developing teeth.⁷² Tooth buds grown in the presence of inhibitors of serotonin uptake fail to develop beyond the bud stage.

Continued research of the signaling events initiated by growth factors and matrix molecules will soon lead to a more complete understanding of tooth development. According to Slavkin,⁷³ “Recent advances towards identifying epigenetic signals such as growth factors, regulatory or homeotic genes, and the significant advances towards understanding how cis- and trans-regulating elements control differential gene expression during development provide enormous optimism for future research in craniofacial genetics and developmental biology.”

Establishing Coronal Form (Cusp Formation)

As noted earlier, the three-dimensional plane of the IEE basal lamina sets the position of the dentinoenamel junction and thus the anatomic shape of the crown. From the cap stage, the enamel organ continues to increase in size until it assumes a bell-shaped structure, almost completely enclosing the dental papilla (see Fig 1-1). The three-dimensional shape of the enamel organ, at various stages of its development, has been precisely reconstructed from serial sections of human embryos. In extensive studies of human embryos, Ooe⁷⁴ has demonstrated that secretion and mineralization of dentin and enamel matrices begin only after the shape of the crown has been determined in soft tissues.

Numerous factors under genetic control, including rates of cell division, assembly of cytoplasmic contractile filaments in differentiating preameloblasts, and the osmotic pressure of the surrounding tissues, act to shape the three-dimensional topography of the basement membrane between the IEE and the dental papilla. Cusp outline is set by the three-dimensional folding of the IEE basement membrane, setting the position of the future dentinoenamel junction. Cells in both the preameloblast and preodontoblast compartments must stop dividing to differentiate into matrix-producing ameloblasts (enamel) and odontoblasts (dentin) (Fig 1-15). Proliferation is controlled from primary and secondary enamel knots established over the tips of the future cusps. The FGF-4 and EGF produced by the nondividing cells of the EK may diffuse laterally to regulate cell proliferation in the IEE and the underlying preodontoblasts (see Fig 1-12).

Fig 1-15 Proliferation of preodontoblasts and preameloblasts from undifferentiated precursors in the dental papilla and inner enamel epithelium located in the cervical loop area. Cell cohorts leave the proliferation compartment and differentiate into mature secretory cells. Odontoblast differentiation and dentin deposition occur slightly in advance of ameloblast differentiation and enamel matrix secretion.

Apoptosis of epithelial cells in the EK terminates cusp growth.⁴⁹ As the enamel knot begins its apoptotic decline, its function is transferred to the stratum intermedium. Progressing away from the tip of the cusp, in the proximodistal direction, a wave of signaling activity occurs in the cells of the stratum inter-medium that promotes the cell proliferation necessary to complete the morphodifferentiation of the bell-shaped crown.

Cell division at the cervical loop extends the size of the enamel organ until it reaches its mature state as a bell-shaped organ almost encompassing the dental papilla. Harada et al⁷⁵ have demonstrated the presence of stem cells in the stellate reticulum of the cervical loop. Each division of a stem cell creates two daughter cells; one remains within the stem cell pool while the other cell enters the transit-amplifying pool (preameloblasts) within the IEE. A signaling pathway involving Notch and its ligand (Lunatic fringe) plays a central role in determining daughter cell entry into the differentiation pathway.^75,76 Odontoblasts differentiate slightly in advance of ameloblasts, forming a thin layer of predentin prior to the start of enamel secretion.

Basic Science Correlations

Cell migration

Embryonic development involves orderly and precisely timed cell migrations. In many cases, cells must move over long distances. Some migrations contain large cohorts of cells moving over relatively long distances, as in the migration of neural crest cells from specific sites in the neural tube of the head region to their final destination in the developing face and jaws. Another example is the migration of pigment cells from the neural crest to sites throughout the epidermis. Tooth development requires the migration of neural crest ectomesenchyme to appropriate locations in the developing jaw. During root development, cells of the dental sac migrate toward the newly deposited dentin surface prior to cementogenesis.

For decades, developmental biologists sought answers to the following questions: What is the basis of cell motility? What guides a migrating cell to its ultimate destination? Although the answers to these questions are still incomplete, rapid progress is being made in understanding the molecular basis of cell migration. Directed cell locomotion is a complex process. It requires plasma membrane cycling or flow, the interaction of cell surface integrins with components of the extracellular matrix as well as the cytoskeleton, and the contraction of actin and myosin filaments.^77,78 It also requires receptor-ligand signaling systems to detect and respond to gradients of chemotactic molecules.

Some cells types are relatively stationary, while other types engage in locomotion (neutrophils and lymphocytes).⁷⁹ Transmigration through the extracellular matrix is a result of the cell’s capacity to explore its immediate environment. It does this through the extension of probing cytoplasmic processes (lamellae and filopodia).⁸⁰ Lamellae are flat folds of cytoplasm sent out across a broad area, while filopodia are narrow fingerlike protrusions (Fig 1-16).

The extension and retraction of lamellae and filopodia are, in part, responses to two fundamental properties of the cell: the continuous turnover of the plasma membrane, and the contractility of cytoplasmic microfilaments. When cell processes from a region of the cell boundary make adhesive contact with a substrate, cytoplasmic polarity is established toward the substrate, and new membrane is transported toward that surface. This region of the cell surface has the potential of becoming the leading edge if there is no impediment to prevent the cell from moving forward in that direction. New membrane is added to the leading edge of the cell and retrieved toward the center of the cell.

Fig 1-16 Changes in shape and cell-to-substrate contacts made by chick heart fibroblasts explanted onto plastic culture dishes. (A) In the early phase of migration, the cells exhibit a clear leading lamella devoid of dense focal contacts. Only close contacts are made at this stage. (B) With time, the cells establish filopodia and focal contacts at the leading edge. A tail of trailing cytoplasm is characteristically found on migrating fibroblasts. (C) After 3 days in culture, most cells no longer have the migratory phenotype, no leading lamella is observed, and many well-developed focal adhesions are present in many regions of the cells. (Adapted from Couchman and Rees⁸¹ with permission from The Company of Biologists.)

It has been calculated that the lipid phase of the plasma membrane of a fibroblast turns over in about 50 minutes. Some intramembrane proteins are caught up in this flow, while others remain in place because of their association with the internal cytoskeleton or with extracellular substrates.

Protrusion of lamellae and filopodia at the leading edge is driven by rapid polymerization of actin filaments (see chapter 11 for a discussion of actin filament formation). Assembly of linear actin bundles may push the membrane outward or cause an increase in local hydrostatic pressure to deform the membrane outward at the leading edge. Because calcium triggers actin polymerization, it has been proposed that filopodial formation at the leading edge might be regulated by the entry of calcium ions through cell membrane channels.

Another explanation for the forward extension of the plasma membrane is the assembly of new membrane via exocytosis at the leading edge and the simultaneous endocytosis toward the middle and rear of a migrating cell. Polarized exocytosis-endocytosis cycles have been observed in migrating fibroblasts and neurite growth cones.

To develop traction and forward movement, cells must form attachments between their leading edge and the substratum. Cells migrating in vitro on glass cover slips make close contacts and focal contacts with the surface of the glass.⁸² At close contacts, the cell membrane is separated from the substratum by a space of 20 to 30 nm. Close contacts represent the initial association of specific cell membrane attachment proteins to the extracellular matrix. Close contacts are typically found at the very leading edge of lamellae and filopodia.

In contrast, focal contacts typically occur just distal to the outer zone of the leading edge (Figs 1-16 and 1-17). In focal contacts, the cell membrane is only 10 to 15 nm from the surface of the substrate. The focal contact is the product of the maturation of the close contact by recruitment of integrin receptors and other membrane-associated proteins. Along with the integrins, actin, vinculin, and talin are rapidly associated with the initial site of attachment to form a focal contact or focal adhesion. Thus, the integrins mediate transmembrane linkage of the cytoskeletal proteins to the extracellular matrix.⁸³

Fig 1-17 Hypothesis proposed by Harris (1973) to explain how the forward movement of cells is coordinated to the development of stable cell-to-matrix contacts associated with actin and myosin filament bundles. (A) Focal contacts (1 and 2) established at the leading edge remain in position as (B) new membrane and cytosol advance in the continued protrusion of the leading lamella. (C) With time, the focal contacts, first established at A, become located at the trailing end of the cell, and will eventually be ruptured as the tail is pulled forward. The detached focal contacts with bits of cytoplasm remain attached to the substratum. Contraction of actin and myosin in the cell body propels cytosol forward to the leading lamella. In the process, matrix molecules become aligned parallel to the direction of cell migration. (Adapted from Hay.⁸⁸)

The integrin dimer α5β1 represents one type of integrin fibronectin receptor. Fibronectin participates as the extracellular component of the close contact in migrating fibroblasts and neural crest cells.⁸⁴ Motile cells make cell-to-matrix attachment interactions of a transient nature (close contacts). Fibronectin receptors tend to be more dispersed over the surface of migrating cells.

Cell-to-cell attachments and stable cell-to-matrix adhesions (focal adhesions) assume greater importance in stabilizing nonmotile cells at their final destination. In stationary cells, the fibronectin receptors are clustered in alignment with extracellular fibronectin fibrils.^85,86 When cells are attached to matrix fibrils, which are under tension, the cells develop large focal adhesions (fibronexus) associated with cytoplasmic actin and myosin bundles (stress fibers). The fibronexus junction is described in chapter 6.

Specific extracellular matrix molecules, organized into three-dimensional scaffolds, provide pathways for the selective migration of certain cell types. Neural crest cells migrate in defined tracks rich in fibronectin and hyaluronic acid. The same is true for the migration of fibroblasts into the primary corneal stroma. The basal lamina, or substances associated with it, can also act as a substrate for the preferential migration of cells in vivo. Certain types of neural crest cells end their migration when they encounter regions rich in tenascin, a large extracellular attachment molecule.

Several environmental stimuli cause cells to undergo directed migration. Cells can move along a concentration gradient of an extracellular matrix molecule (haptotaxis). In an electrical field, cells migrate toward the cathode (galvanotaxis). Fibronectin fragments induce directed migration of fibroblasts, a stimulus likely to be important in wound healing.⁸⁷

Cells also tend to move outward from a cell mass. Cells on the perimeter of the cell mass continue to form leading lamellae and filopodia along their free surface and thus are able to move away from the cell mass. Within the cell mass, however, cells are contact inhibited; a state of reduced membrane ruffling and filopodial extension occurs along the adjacent surfaces of juxtaposed cells. Directed migrations of neural crest cells within the extracellular matrix scaffold proceed from areas of high to low cell density because of contact inhibition.

Extracellular matrix molecules may undergo reorganization following interaction with the cell surface of a migrating cell (see Fig 1-17).^84,88–90 Traction transmitted to the extracellular matrix by migrating (contracting) cells also exerts an organizational influence over matrix molecules. As fibroblasts migrate through a collagen gel in vitro, they cause the extracellular collagen fibrils to become aligned parallel to the long axis of the fibroblasts and the gel to contract. Fibronectin fibrils increase in size and organization toward the trailing edge of migrating fibroblasts. The role of cell polarity and migration in determining the organization of collagen in the periodontal ligament is discussed in chapter 6.

Cell and substrate adhesion molecules

Calcium-dependent cadherins, integrins, selectins, plasma membrane proteoglycans, and members of the immunoglobulin superfamily, such as neural cell adhesion molecule, participate in forming cell-to-cell and cell-to-matrix adhesions.⁹¹ Members of these transmembrane proteins play essential roles in the cellular organization of tissues and organs and in the migration of cells in embryonic and adult tissues.^91–93 The cadherins, components of desmosomes, are discussed in chapter 4, and the selectins, adhesion molecules that regulate leukocyte emigration from blood vessels, are described in chapters 13 and 14.

Integrins The integrins are a family of cell surface transmembrane proteins that developed very early in evolution^91–94 (Figs 1-18 and 1-19). Integrins are heterodimers made up of α and β subunits. At least 14 α and eight β subunits have been identified. Figure 1-19 contains a chart of the subunits and ligands of the very late activation–type integrins.

Fig 1-18 Integrin-type receptors. The α and β integrin transmembrane proteins form a dimer with a shared ligand-binding site. Metal-binding sites on the α subunit are needed for receptor function.

Fig 1-19 Integrin molecules of the very late activation subfamily. Heterodimers of β and α subunits form cell surface receptors interacting with various extracellular matrix adhesion molecules. (Co) Collagen; (FN) fibronectin; (LM) laminin; (VCAM-1) vascular cell adhesion molecule 1; (VN) vitronectin. (Adapted from Arnaout⁹¹ with permission from Elsevier Science.)

Both integrin subunits are transmembrane proteins. The extracellular globular domains are larger than the cytoplasmic and intramembrane segments (see Fig 1-18). The extracellular portion of the α subunit contains metal-binding sites necessary for receptor function. The combined external globular domains of the α and β subunits form the ligand-binding site. Some integrins bind more than one type of ligand; for example, the α1β1 integrin binds to both collagen and laminin (see Fig 1-19). It is also apparent that individual ligands, such as fibronectin, are recognized by several integrins.

Cells use integrins to adhere to a variety of extracellular matrix molecules and to communicate chemically in a bidirectional way with their environment. Information from the extracellular matrix is gathered when ligands bind to the extracellular portion of the integrins, producing conformational changes in the cytoplasmic portion of the molecules and thereby altering their interaction with adjacent cytoplasmic molecules. Ligand binding to integrins can also exert an intracellular effect through the activation of tyrosine kinases.

Conversely, the binding of certain cytoplasmic proteins to the cytoplasmic domain can induce conformational changes in the external part of the integrin molecules, affecting their affinity for extracellular ligands. Through this process, the cell can interact with its environment, creating adhesive contacts and/or activating specific differentiation cascades.

The expression of integrin receptors for laminin has been shown to oscillate between IEE and dental papilla ectomesenchyme during tooth formation.⁹⁵ Whether integrin-laminin signaling pathways have a significant role in ameloblast differentiation remains to be determined. Additional discussions of the role of integrins in cell activation and muscle differentiation are contained in chapters 11, 13, and 14.

Syndecan Syndecans are integral membrane proteoglycans. Four types have been identified based on differences in the core protein. Each syndecan molecule consists of a short cytoplasmic domain, a helical hydrophobic domain inserted into the plasma membrane, and a large extracellular domain containing several glycosaminoglycan side chains.

Syndecan 1 is typically located in epithelia and in embryonic mesenchymal tissues, especially in areas of epithelial-mesenchymal interaction, such as in developing teeth.⁹⁶ Because of its binding interaction with tenascin, it may play a role in condensation of ectomesenchymal cells to form the dental papilla.³⁷ In addition to binding tenascin, syndecan 1 also binds fibronectin, and collagen types I, III, and V.

Syndecan 4 is the smallest and most widely distributed type of syndecan. It colocalizes with integrins in focal adhesions to extracellular fibronectin. Syndecans are not only matrix receptors but also coreceptors for growth factors and cytokines, capable of potentiating signal transduction events.

Fibronectin Fibronectin is a large extracellular glycoprotein with multiple binding sites capable of forming attachments to cells, collagen, heparin, fibrin, tenascin, bacteria, and other proteoglycans.^97,98 Fibronectin has a dimeric structure composed of two equal polypeptide chains joined near their C-terminal by disulfide bonds. Binding sites on each chain have been identified for cell membrane integrins and a variety of extracellular matrix molecules (Fig 1-20). Fibronectin is a significant component of basement membranes in developing organ systems, where it stabilizes cells and thereby permits them to establish polarity and to undergo further differentiation. A good example of this type of interaction occurs during the differentiation of the preodontoblasts.

Fig 1-20 The elongated fibronectin molecule is made up of two similar subunits. Each consists of globular domains joined by flexible polypeptide sections. Specific binding sites have been mapped on the molecule for various cells and molecules as shown.

The interaction of cells with fibronectin is important not only during embryonic development but also in the migration and stabilization of cells in the adult organism. Fibronectin plays an important role in wound healing by interacting with fibrin to create a scaffold for the migration of fibroblasts. Fibronectin stimulates fibroblast invasion of collagen gels. The gelatin-binding domain of the fibronectin molecule is essential to this migratory action. The gelatin-binding domain segment interacts with a fibroblast surface integrin protein to induce a transition to the migratory phenotype.

The recognition site of the cell-binding domain of fibronectin has been identified to consist of the tripeptide, arginine-glycine–aspartic acid (the RGD sequence). This sequence binds to the cell membrane integrins (fibronectin receptors). The α5β1 integrin is the main fibronectin receptor. The association of integrin fibronectin receptors to extracellular fibronectin triggers the recruitment of cytoskeletal and signaling molecules to the plasma membrane site of attachment to form focal adhesions. Fibronectin is concentrated at the IEE basal lamina and along the cytoplasmic surface of preodontoblasts.^4,99–101 The role of fibronectin and its receptor in odontoblast differentiation is discussed in chapter 2.

Laminin Laminin is a major constituent of the basal lamina complex. It is a large glycoprotein with a molecular weight of about 800,000 d. The laminin molecule is a heterotrimer of β1, β2, and α subunits. The three chains assemble to form a cross-shaped molecule (Fig 1-21).¹⁰² Laminin binds to type IV collagen, to heparan sulfate proteoglycans (perlecan) of the basal lamina, and to receptors in the cell membrane of various cells, especially epithelial cells. Laminin 5 subunits are expressed in the enamel organ, and the protein is localized in the basal lamina beneath the IEE.^99,103

Fig 1-21 Structure of the laminin molecule.

Immunocytochemical studies reveal temporospatial changes in laminin subunit expression during odontoblast and ameloblast differentiation.¹⁰³ The results of tissue recombination experiments suggest that the dental ectomesenchyme controls the expression of laminin in the dental epithelium.¹⁰⁴ Laminin is discussed further in chapter 4.

Tenascin Tenascin, a large extracellular matrix molecule, also known as cytotactin and hexabrachion, is made up of six polypeptide chains assembled to form a sixarm structure capable of interacting with a variety of cells and extracellular matrix molecules. Because the six polypeptide chains appear to represent separate gene products, it has been suggested that tenascin molecules may have tissue specificity.

Tenascin binds to cell surface proteoglycan (syndecan). Expression of tenascin in dental ectomesenchyme coincides with the concentration of the dental papilla.^100,105 It has been demonstrated that tenascin prevents the migration of certain neural crest cells, causing them to assume a round shape characteristic of stationary cells.

Nidogen Nidogen (also called entactin) is a rod-shaped protein consisting of a single polypeptide chain, approximately 30 nm long, with globular domains at each end and one centrally located domain.^106,107 Because nidogen binds with high affinity to collagen IV and laminin, it has an organizing role in assembly of the basal lamina. Nidogen also binds perlecan, the large heparan sulfate proteoglycan of the basal lamina.

The coexpression of laminin 1 and nidogen results in a relatively stable basal lamina. In general, laminin is produced by epithelial cells and nidogen by mesenchymal cells. Temporospatial differences in the expression of laminin and nidogen are thought to have significance in epithelial-mesenchymal tissue remodeling because of resulting changes in the stability of the basement membranes.¹⁰⁸

Basal lamina The basal lamina is a supramolecular aggregate of type IV collagen, laminin, fibronectin, nidogen, and perlecan. They form a macromolecular network with the dual function of supporting epithelial cells and providing for a permeability barrier or filter. Meyer et al¹⁰⁹ have reviewed the role of the basal lamina in tooth development and odontoblast differentiation. The basal lamina is discussed in detail in chapter 4.

Clinical Correlation: The Human Dentition

The primary (deciduous) dentition consists of 20 teeth, five in each quadrant (Fig 1-22).^74,110 The permanent incisors, canines, and premolars form from successional laminae that extend lingually from the primary precursors toward the midline (see Fig 1-22). The permanent molars develop from a distal extension of the dental lamina, the accessional lamina (Fig 1-23). Some dental embryologists consider the permanent molars to be members of the first dentition. Their microscopic successors undergo an aborted development.

Fig 1-22 Developing primary teeth and the primordia of the permanent teeth in a 28-week human fetus. Maxillary quadrant. (i1) Primary central incisor; (i2) primary lateral incisor; (c) primary canine; (m1) primary first molar; (m2) primary second molar; (I1) permanent central incisor; (I2) permanent lateral incisor; (C) permanent canine; (P1) permanent first premolar; (P2) permanent second premolar; (M1) permanent first molar. (Adapted from Ooe⁷⁴ with permission.)

Fig 1-23 Mandibular molar region in a 159-mm fetus (at 20 weeks old), depicting the formation of the permanent first molar (M1) from a distal extension of the primordia of the primary second molar (m2). (m1) Primary first molar. (Adapted from Ooe⁷⁴ with permission.)

During the development of primary teeth, the central incisor and canine are positioned labial to the lateral incisor (Fig 1-24). This arrangement is noted very early in the formation of the enamel organ from the dental lamina. The buds of the permanent teeth have a similar position, so that the lateral incisor is positioned lingual to the central incisor and canine. During normal postnatal development, space is created in the dental arch for the alignment of all anterior teeth. Often, the space created is insufficient, and the central incisor and the canine crowd out the lateral incisor.

Fig 1-24 Epithelial portion of the anterior tooth germs and adjacent structures in a 144-mm fetus. (i1) Primary central incisor; (i2) primary lateral incisor; (c) primary canine. (Adapted from Ooe⁷⁴ with permission.)

References

1. Graveson AC, Smith MM, Hall BK. Neural crest potential for tooth development in a urodele amphibian: Developmental and evolutionary significance. Dev Biol 1997;188:34–42.

2. Lumsden AGS. The neural crest contribution to tooth development in the mammalian embryo. In: Maderson PFA (ed). Developmental and Evolutionary Aspects of the Neural Crest. New York: Wiley, 1987:261–300.

3. Thesleff I, Partaanen AM, Vainio S. Epithelial-mesenchymal interactions in tooth morphogenesis: The roles of extracellular matrix, growth factors, and cell surface receptors. J Craniofac Genet Dev Biol 1991;11:229–237.

4. Ruch JV, Lesot H, Karcher-Djuricic V, Meyer JM, Mark M. Epithelial-mesenchymal interactions in tooth germs: Mechanisms of differentiation. J Biol Buccale 1983;11:173–193.

5. Thesleff I, Vaahtokari A, Vainio S, Jowett A. Molecular mechanisms of cell and tissue interactions during early tooth development. Anat Rec 1996;245:151–161.

6. Slavkin HC, Diekwisch T. Evolution in tooth developmental biology: Of morphology and molecules. Anat Rec 1996;245: 131–150.

7. Slavkin HC. Molecular determinants during dental morphogenesis and cytodifferentiation: A review. J Craniofac Genet Dev Biol 1991;11:338–349.

8. Bronner-Fraser M. Origins and developmental potential of the neural crest. Exp Cell Res 1995;218:405–417.

9. Imai H, Osumi-Yamashita N, Ninomiya Y, Eto K. Contribution of early-emigrating midbrain crest cells to the dental mesenchyme of mandibular molar teeth in rat embryos. Dev Biol 1996;176:151–165.

10. LeDouarin NM, Dupin E, Ziller C. Genetic and epigenetic control in neural crest development. Curr Opin Gen Dev 1994;4:685–695.

11. Thomas BL, Tucker AS, Ferguson C, Qiu M, Rubenstein JLR, Sharpe PT. Molecular control of odontogenic patterning: Positional dependent initiation and morphogenesis. Eur J Oral Sci 1998;106:44–47.

12. Thomas BL, Tucker AS, Qiu M, Ferguson C, Hardcastle Z, Rubenstein JLR, Sharpe PT. Role of Dlx-1 and Dlx-2 genes in patterning of the murine dentition. Development 1997;124: 4811–4818.

13. Mina M, Kollar E. The induction of odontogenesis in non-dental mesenchyme combined with early murine mandibular arch epithelium. Arch Oral Biol 1987;32:123–127.

14. Kollar E, Mina M. Role of the early epithelium in the patterning of the teeth and Meckle’s cartilage. J Craniofac Genet Dev Biol 1991;11:223–228.

15. Lumsden AGS. Spatial organization of the epithelium and the role of neural crest cells in the initiation of the mammalian tooth germ. Development 1988;103(suppl):155–169.

16. Wang Y-S, Upholt WB, Sharpe PT, Kollar E, Mina M. Odontogenic epithelium induces similar molecular responses in chick and mouse mandibular mesenchyme. Dev Dyn 1998; 213:386–397.

17. Weiss K, Stock D, Zhao Z, Buchanan A, Ruddle F, Shashikant C. Perspectives on genetic aspects of dental patterning. Eur J Oral Sci 1998;106:55–63.

18. Qiu M, Bulfone A, Ghattas I, Meneses JJ, Christensen L, Sharpe PT, Presley R, Pedersen RA, Rubenstein JLR. Role of the Dlx homeobox genes in proximodistal patterning of the branchial arches: Mutations of Dlx-1, Dlx-2, and Dlx-1 and -2 alter morphogenesis of proximal skeletal and soft tissue structures derived from the first and second arches. Dev Biol 1997;185:165–184.

19. Sharpe PT. Homeobox genes and orofacial development. Connect Tissue Res 1995;31:1–9.

20. Chen Y, Bei M, Woo I, Satokata I, Mass R. Msx1 controls inductive signaling in mammalian tooth morphogenesis. Development 1996;122:3035–3044.

21. Whiting J. Craniofacial abnormalities induced by the ectopic expression of homeobox genes. Mutation Res 1997;396: 97–112.

22. Satokata I, Mass R. Msx-1-deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat Genet 1994;6:348–356.

23. Karg H, Burger EH, Lyaruu DM, Bronckers ALJJ, Woltgens JHM. Spatiotemporal expression of the homeobox gene S8 during mouse tooth development. Arch Oral Biol 1997; 42:625–631.

24. Jernvall J, Kettunen P, Karavanova I, Martin LB, Thesleff T. Evidence for the role of the enamel knot as a control center in mammalian tooth cusp formation: Non-dividing cells express growth stimulating Fgf-4 gene. Int J Dev Biol 1994;38:463–469.

25. Vaahtokari A, Aberg T, Jernvall J, Keränen S, Thesleff I. The enamel knot as a signaling center in the developing mouse tooth. Mech Dev 1996;54:39–43.

26. Nieminen P, Pekkanen M, Aberg T, Thesleff I. A graphical WWW-database on gene expression in tooth. Eur J Oral Sci 1998;106:7–11.

27. Decker JD. A light and electron microscopic study of the rat molar enamel organ. Arch Oral Biol 1963;8:301–310.

28. Pannese E. Observations on the ultrastructure of the enamel organ. I. Stellate reticulum and stratum intermedium. J Ultrastruc Res 1960;4:372–400.

29. Pannese E. Observations on the ultrastructure of the enamel organ. II. Involution of the stellate reticulum. J Ultrastruc Res 1961;5:328–342.

30. Pannese E. Observations on the ultrastructure of the enamel organ. III. Internal and external enamel epithelia. J Ultrastruc Res 1962;6:186–204.

31. Matthiessen ME, Garbarsch C, Olsen BE, Hellström S, Engström-Laurent A. Hyaluronan in human deciduous tooth germs in the bell stage—Histochemistry and immunohistochemistry. Acta Anat (Basel) 1997;159:1–7.

32. Cobourne MT. The genetic control of early odontogenesis. Br J Orthod 1999;26:21–28.

33. Thesleff I, Sharpe PT. Signalling networks regulating dental development. Mech Dev 1997;67:111–123.

34. Kollar E, Baird GR. The influence of the dental papilla on the development of tooth shape in embryonic mouse tooth germs. J Embryol Exp Morphol 1969;21:131–148.

35. Kollar EJ, Baird GR. Tissue interactions in embryonic mouse tooth germs. II. The inductive role of the dental papilla. J Embryol Exp Morphol 1970;24:173–186.

36. Kollar EJ. Tissue interactions in development of teeth and related ectodermal derivatives. Dev Biol 1986;4:297–313.

37. Thesleff I, Vainio S, Salmivirta K, Jalkanen M. Syndecan and tenascin: Induction during early tooth morphogenesis and possible interactions. Cell Differ Dev 1990;32:383–390.

38. Vainio S, Thesleff I. Sequential induction of syndecan, tenascin and cell proliferation associated with mesenchymal cell condensation during early tooth development. Differentiation 1992;50:97–105.

39. Salmivirta K, Elenius K, Vainio S, Hofer V, Chiquet-Ehrismann R, Thesleff I, Jalkanen M. Syndecan from embryonic tooth mesenchyme binds tenascin. J Biol Chem 1991;266: 7733–7739.

40. Lotz MM, Burdsal CA, Erickson HP, McClay DR. Cell adhesion to fibronectin and tenascin: Quantitative measurements of initial binding and subsequent strengthening response. J Cell Biol 1989;109:1795–1805.

41. Saga Y, Yagi T, Ikawa Y, Sakakura T, Aizawa S. Mice develop normally without tenascin. Genes Dev 1992;6:1821–1831.

42. Yamada M, Bringas P, Grodin M, MacDougall M, Slavkin HC. Developmental comparisons of murine secretory amelogenesis in vivo, as xenografts on the chick chorio-allantoic membrane, and in vitro. Calcif Tissue Int 1980;31:161–171.

43. Chai Y, Mah A, Crohin C, Groff S, Bringas P Jr, Le T, Santos V, Slavkin HC. Specific transforming growth factor- β subtypes regulate embryonic mouse Meckel’s cartilage and tooth development. Dev Biol 1994;162:85–103.

44. Thesleff I, Vaahtokari A, Kettunen P, Åberg T. Epithelial-mesenchymal signaling during tooth development. Connect Tissue Res 1995;32:9–15.

45. Thesleff I, Aberg T. Tooth morphogenesis and the differentiation of ameloblasts. In: Chadwick DJ, Cardew G (eds). Dental Enamel. New York: Wiley, 1997:3–12.

46. Vainio S, Karavanova I, Jowett A, Thesleff I. Identification of BMP-4 as a signal mediating secondary induction between epithelial and mesenchymal tissues during early tooth development. Cell 1993;75:45–58.

47. Jernvall J, Aberg T, Kettunen P, Keränen S, Thesleff I. The life history of an embryonic signaling center: BMP-4 induces p21 and is associated with apoptosis in the mouse tooth enamel knot. Development 1998;125:161–169.

48. Koyama E, Yamaai T, Iseki S, Ohuchi H, Nohno T, Yoshioka H, Hayashi Y, Leatherman JL, Golden EB, Noji S, Pacifici M. Polarizing activity, Sonic hedgehog, and tooth development in embryonic and postnatal mouse. Dev Dyn 1996;206: 59–72.

49. Vaahtokari A, Aberg T, Thesleff I. Apoptosis in the developing tooth: Association with an embryonic signaling center and suppression by EGF and FGF-4. Development 1996;122: 121–129.

50. Heikinheimo K, Bègue-Kirn C, Ritvos O, Tuuri T, Ruch JV. Activin and bone morphogenetic protein (BMP) signalling during tooth development. Eur J Oral Sci 1998;106:167–173.

51. Bloch-Zupan A, Décimo D, Loriot M, Mark MP, Ruch JV. Expression of nuclear retinoic acid receptors during mouse odontogenesis. Differentiation 1994;57:195–203.

52. Bloch-Zupan A, Mark MP, Weber B, Ruch JV. In vitro effects of retinoic acid on mouse incisor development. Arch Oral Biol 1994;39:891–900.

53. Mark MP, Bloch-Zupan A, Wolf C, Ruberte E, Ruch J-V. Involvement of cellular retinoic acid-binding proteins I and II (CRABPI and CRABPII) and of the cellular retinol-binding protein (CRBPI) in odontogenesis in the mouse. Differentiation 1991;48:89–98.

54. Kronmiller JE, Nguyen T, Berndt W. Instruction by retinoic acid of incisor morphology in the mouse embryonic mandible. Arch Oral Biol 1995;40:589–595.

55. Kronmiller JE, Beeman CS, Nguyen T, Berndt W. Blockade of the initiation of murine odontogenesis in vitro by citral, an inhibitor of endogenous retinoic acid synthesis. Arch Oral Biol 1995;40:645–652.

56. Mellanby H. The effect of maternal dietary deficiency of vitamin A on dental tissues in rats. J Dent Res 1941;20:489–503.

57. Knudsen PA. Congenital malformations of lower incisors and molars in exencephalic mouse embryos induced by hypervitaminosis A. Acta Odontol Scand 1967;25:669–691.

58. Kronmiller JE, Upholt WB, Kollar EJ. Alteration of murine odontogenic patterning and prolongation of expression of epidermal growth factor mRNA by retinol in vitro. Arch Oral Biol 1992;37:129–138.

59. Hashimoto Y, Shudo K. Retinoids and their nuclear receptors. Cell Biol Rev 1991;25:209–230.

60. Kronmiller JE. Spatial distribution of epidermal growth-factor transcripts and effects of exogenous epidermal growth factor on the pattern of the mouse dental lamina. Arch Oral Biol 1995;40:137–143.

61. Partanen A-M, Thesleff I. Localization and quantitation of 125I-epidermal growth factor binding in mouse embryonic tooth and other embryonic tissues at different developmental stages. Dev Biol 1987;120:186–197.

62. Kronmiller JE, Upholt WB, Kollar EJ. EGF antisense oligonucleotides block murine odontogenesis in vitro. Dev Biol 1991;147:485–488.

63. Mitsiadis TA, Muramatsu T, Muramatsu H, Thesleff I. Midkine (MK), a heparin-binding growth/differentiation factor, is regulated by retinoic acid and epithelial-mesenchymal interactions in the developing mouse tooth, and affects cell proliferation and morphogenesis. J Cell Biol 1995;129:267–281.

64. Mitsiadis TA, Salmivirta M, Muramatsu T, Muramatsu H, Rauvala H, Lehtonen E, Jalkanen M, Thesleff I. Expression of the heparin-binding cytokines, midkine (MK) and HB-GAM (pleiotrophin) is associated with epithelial-mesenchymal interactions during fetal development and organogenesis. Development 1995;121:37–51.

65. Luukko K, Moshnyakov M, Sainio K, Saarma M, Sariola H, Thesleff I. Expression of neurotrophin receptors during rat tooth development is developmentally regulated, independent of innervation, and suggests functions in the regulation of morphogenesis and innervation. Dev Dyn 1996;206:87–99.

66. Nosrat CA, Fried K, Lindskog S, Olson L. Cellular expression of neurotrophin mRNAs during tooth development. Cell Tissue Res 1997;290:569–580.

67. Nosrat CA, Fried K, Ebendal T, Olson L. NGF, BDNF, NT3, NT4, and GDNF in tooth development. Eur J Oral Sci 1998;106:94–99.

68. Christensen LR, Mollgard K, Kjaer I, Janas MS. Immunocytochemical demonstration of nerve growth factor receptor (NGF-R) in developing human fetal teeth. Anat Embryol 1993;188:247–255.

69. Zhang CZ, Li H, Young WG, Bartold PM, Chen C, Waters MJ. Evidence for a local action of growth hormone in embryonic tooth development in the rat. Growth Factors 1997;14: 131–143.

70. Joseph BK, Savage NW, Young WG, Waters MJ. Prenatal expression of growth hormone receptor/binding protein and insulin-like growth factor-I (IGF-I) in the enamel organ. Role for growth hormone and IGF-I in cellular differentiation during early tooth formation. Anat Embryol (Berl) 1994;189: 489–494.

71. Tabata MJ, Kim K, Liu JG, Yamashita K, Matsumura T, Kato J, Iwamoto M, Wakisaka S, Matsumoto K, Nakamura T, Kumegawa M, Kurisu K. Hepatocyte growth factor is involved in the morphogenesis of tooth germ in murine molars. Development 1996;122:1243–1251.

72. Moiseiwitsch JRD, Lauder JM. Stimulation of murine tooth development in organotypic culture by the neurotransmitter serotonin. Arch Oral Biol 1996;41:161–165.

73. Slavkin HC. Molecular biology of dental development: A review. In: Davidovitch Z (ed). The Biology of Tooth Eruption and Root Resorption. Birmingham, AL: EMBCO Media, 1988:107–116.

74. Ooe T. Human tooth and dental arch development. Tokyo: Ishiyaku, 1981.

75. Harada H, Kettunen P, Jung H-S, Mustonen T, Wang YA, Thesleff I. Localization of putative stem cells in dental epithelium and their association with Notch and FGF signaling. J Cell Biol 1999;147:105–120.

76. Mitsiadis TA, Henrique D, Thesleff I, Lendahl U. Mouse serrate-1 (jagged-1): Expression in the developing tooth is regulated by epithelial-mesenchymal interactions and fibroblast growth factor-4. Development 1997;124:1473–1483.

77. Bretscher MS. Getting membrane flow and the cytoskeleton to cooperate in moving cells. Cell 1996;87:601–606.

78. Mitchison TJ, Cramer LP. Actin-based cell motility and cell locomotion. Cell 1996;84:371–379.

79. Haemmerli G. Principles of cell motility and their morphologic manifestations. Exp Biol Med 1985;10:89–117.

80. Abercrombie M, Heaysman JEM, Pegrum SM. The locomotion of fibroblasts in culture. IV. Electron microscopy of the leading lamella. Exp Cell Res 1971;67:359–367.

81. Couchman JR, Rees DA. The behavior of fibroblasts migrating from chick heart explants: Changes in adhesion, locomotion and growth, and in the distribution of actomyosin and fibronectin. J Cell Sci 2002;39:149–165.

82. Izzard CS, Izzard SL, DePasquale JA. Molecular basis of cell-substrate adhesions. Exp Biol Med 1985;10:1–22.

83. Turner CE, Burridge K. Transmembrane molecular assemblies in cell-extracellular matrix interactions. Curr Opin Cell Biol 1991;3:849–853.

84. Brown MJ, Loew LM. Graded fibronectin receptor aggregation in migrating cells. Cell Motil Cytoskeleton 1996;34: 185–193.

85. Duband JL, Nuckolls GH, Ishihara A, Hasegawa T, Yamada KM, Thiery JP, Jacobson K. Fibronectin receptor exhibits high lateral mobility in embryonic locomoting cells but is immobile in focal contacts and fibrillar streaks in stationary cells. J Cell Biol 1988;107:1385–1396.

86. Couchman JR, Blencowe S. Adhesion and cell surface relationships during fibroblast and epithelial migration in vitro. Exp Biol Med 1985;10:23–38.

87. Schor SL, Ellis I, Dolman C, Banyard J, Humphries MJ, Mosher DF, Grey AM, Mould AP, Sottile J, Schor AM. Substratum-dependent stimulation of fibroblast migration by the gelatin-binding domain of fibronectin. J Cell Sci 1996;109: 2581–2590.

88. Hay ED. Interaction of migrating embryonic cells with extracellular matrix. Exp Biol Med 1985;10:174–193.

89. Bernanke DH, Markwald RR. Migratory behavior of cardiac cushion tissue cells in a collagen-lattice culture system. Dev Biol 1982;91:235–245.

90. Harris AK, Stopack D, Wild P. Fibroblast traction as a mechanism for collagen morphogenesis. Nature 1981;290:249–251.

91. Arnaout MA. Cell adhesion molecules. In: Kelley WN, Harris ED, Ruddy S, Sledge CB (eds). Textbook of Rheumatology, ed 4. Philadelphia: Saunders, 1993:213–226.

92. Obara N, Takeda M. Expression of the neural cell adhesion molecule (NCAM) during second- and third-molar development in the mouse. Anat Embryol 1993;188:13–20.

93. Gumbiner BM. Cell adhesion: The molecular basis of tissue architecture and morphogenesis. Cell 1996;84:345–357.

94. Hynes RO. Integrins: Versatility, modulation, and signalling in cell adhesion. Cell 1992;69:11–25.

95. Salmivirta K, Gullberg D, Hirsch E, Altruda F, Ekblom P. Integrin subunit expression associated with epithelial-mesenchymal interactions during murine tooth development. Dev Dyn 1996;205:104–113.

96. Bai XM, Van der Schueren B, Cassiman J-J, Van den Berghe H, David G. Differential expression of multiple cell-surface heparan sulfate proteoglycans during embryonic tooth development. J Histochem Cytochem 1994;42:1043–1054.

97. Hynes RO, Yamada KM. Fibronectins: Multifunctional modular glycoproteins. J Cell Biol 1982;95:369–377.

98. Yamada KM, Hayashi M, Hirano H, Akiyama SK. Fibronectin and cell surface interactions. In: Trelstad RL (ed). The Role of Extracellular Matrix in Development. New York: Liss, 1984: 89–121.

99. Garbarsch C, Matthiessen ME, Olsen BE, Moe D, Kirkeby S. Immunohistochemistry of the intercellular matrix components and the epithelio-mesenchymal junction of the human tooth germ. Histochem J 1994;26:110–118.

100. Nagai N, Yamachika E, Nishijima K, Inoue M, Shin HI, Suh MS, Nagatsuka H. Immunohistochemical demonstration of tenascin and fibronectin in odontogenic tumours and human fetal tooth germs. Eur J Cancer B Oral Oncol 1994;30B: 191–195.

101. Sawada T. Expression of basement membrane components in the dental papilla mesenchyme of monkey tooth germs— An immunohistochemical study. Connect Tissue Res 1995; 32:55–61.

102. Timpl R, Brown JC. The laminins. Matrix Biol 1994;14: 275–281.

103. Salmivirta K, Sorokin LM, Ekblom P. Differential expression of laminin α chains during murine tooth development. Dev Dyn 1997;210:206–215.

104. Yoshiba K, Yoshiba N, Aberdam D, Meneguzzi G, Perrin-Schmitt F, Stoetzel C, et al. Expression and localization of laminin-5 subunits during mouse tooth development. Dev Dyn 1998;211:164–176.

105. Tucker RP, Moiseiwitsch JRD, Lauder JM. In situ localization of tenascin mRNA in developing mouse teeth. Arch Oral Biol 1993;38:1025–1029.

106. Paulsson M. Basement membrane proteins: Structure, assembly, and cellular interactions. Crit Rev Biochem Mol Biol 1992;27:93–127.

107. Timpl R, Dziadek M, Fujiwara S, Nowack H, Wick G. Nidogen: A new, self-aggregating basement membrane protein. Eur J Biochem 1983;137:455–465.

108. Dziadek M. Role of laminin-nidogen complexes in basement membrane formation during embryonic development. Experientia 1995;51:901–913.

109. Meyer J-M, Ruch JV, Kubler MD, Kupferle C, Lesot H. Cultured incisors display major modifications in basal lamina deposition without further effect on odontoblast differentiation. Cell Tissue Res 1995;279:135–147.

110. Kitamura H. Embryology of the Mouth and Related Structures. Tokyo: Maruzen, 1989:12–34.