Desarrollo Matriz Esmalte

Journal of Structural Biology 126, 270299 (1999) Article ID jsbi.1999.4130, available online at http://www.idealibrary.
com on
The Structural Biology of the Developing Dental Enamel Matrix

A. G. Fincham,* J. Moradian-Oldak,* and J. P. Simmer
*Center for Craniofacial Molecular Biology, School of Dentistry, University of Southern California, Los Angeles, California 90089; and University of Texas School of Dentistry, Health Science Center at San Antonio, Department of Pediatric Dentistry, San Antonio, Texas Received January 21, 1999, and in revised form April 26, 1999
The biomineralization of the dental enamel matrix with a carbonated hydroxyapatite mineral generates one of the most remarkable examples of a vertebrate mineralized tissue. Recent advances in the molecular biology of ameloblast gene products have now revealed the primary structures of the principal proteins involved in this extracellular mineralizing system, amelogenins, tuftelins, ameloblastins, enamelins, and proteinases, but details of their secondary, tertiary, and quaternary structures, their interactions with other matrix and or cell surface proteins, and their functional role in dental enamel matrix mineralization are still largely unknown. This paper reviews our current knowledge of these molecules, the probable molecular structure of the enamel matrix, and the functional role of these extracellular matrix proteins. Recent studies on the major structural role played by the amelogenin proteins are discussed, and some new data on synthetic amelogenin matrices are 1999 Academic Press reviewed. Key Words: amelogenesis; biomineralization; dental enamel proteins; development; function; hydroxyapatite.
INTRODUCTION . . . It seems to me that much which is paradoxical in enamel is tied up with the particular and perhaps unusual relationship which exists between its organic and inorganic components. The organic matrix seems to specically exist for the eventual creation of a structure which is quantitatively almost completely inorganic and yet possesses the largest hydroxyapatite crystals consistent with, rstly, their ability to be nucleated and to grow in a biological system and, secondly, maintenance of sufficient elasticity in the enamel system as a whole to prevent its hard structure from being too brittle. (John E. Eastoe, 1971)
striking example of a highly mineralized structure exquisitely adapted to absorb essential mechanical and abrasive stresses throughout the lifetime of the organism (Fig. 1). Recent studies have shown that secretory stage ameloblasts secrete a number of proteins, some of which appear to be unique to the dental enamel matrix. Dental enamel differs from other vertebrate mineralized tissues (e.g., bone, cartilage, and dentine) in that it is noncollagenous, of epithelial origin, and does not undergo resorption and remodeling. Recent reviews of dental enamel structure and ameloblast biology include those by Simmer and Fincham (1995), Thesleff (1995), Skobe et al. (1995), Shore et al. (1995a,b), Robinson et al. (1995a), Thesleff and berg (1997), Boyde (1997), Sasaki et al. (1997), and Smith (1998). This article will consider current knowledge of the structure and molecular composition of the extracellular matrix of the developing dental enamel, viewed as a biomineralizing system. First, we consider the requirement for an effective extracellular mineralizing matrix and comparisons of the dental enamel matrix with other biomineralizing systems. Second, we review, in some detail, current knowledge of the molecular composition of the enamel matrix. Third, we consider the structure and putative function of the matrix and the interrelations of the matrix proteins with the mineral phase. Fourth, we briey discuss the properties of the enamel matrix proteinases as they relate to the dynamics of amelogenesis. Finally, we consider outstanding issues regarding the dental enamel matrix and its components.
1. ENAMEL AND ENAMELOID . . . I think it quite reasonable to assume that enameloid is the ancestral tooth crown covering . . . (R.W. Fearnhead, 1979)
Dental enamel is the most highly mineralized structure in the vertebrate body and is formed within a unique, extracellular matrix derived through the synthesis and secretion of proteins by the ameloblast cells of the inner enamel epithelium. Mature dental enamel has a complex form, providing a
1047-8477/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
Enamel and enameloid form the highly mineralized outer layer of the vertebrate tooth, providing the hard abrasion-resistant surface essential for the animal to capture and consume its food. While the gross tooth form varies according to function between both order and species, the mineral phase of both of these tooth structures is conserved.
270
DENTAL ENAMEL MATRIX BIOMINERALIZATION
271
FIG. 1. The organization of dental enamel. Scanning electron micrograph of the surface of an acid-etched ground section of mature mouse incisal dental enamel. Ordered arrays of enamel prisms are each constructed of parallel bundles of carbonated hydroxyapatite enamel crystallites. Each prism structure is considered to arise as the product of a single secretory ameloblast cell of the inner enamel epithelium. (Inset) An enlargement of the center section of the images as indicated. [Electron micrograph courtesy of Dr. W. Luo.]
A carbonated calcium apatite comprises the biomineral phase of both mature enamel and mature enameloid structures in vertebrates, where it occurs as oriented crystallites formed within a secreted extracellular protein matrix. In mammalian enamel, the mineral is present as structured arrays of crystals whose boundaries are dened as enamel prisms or prism sheathes (see Fig. 1). In the enameloid of the Chondrichthyes (sharks and rays), Osteichthyes (bony shes), and in some Amphibia (frogs, toads, and newts), a prismatic enamel structure is generally absent and this aprismatic structure appears as one criterium for distinguishing enameloid from mammalian enamel. However, enamel and enameloid structures are functionally comparable and species-specic in form. Enameloid structure and its comparative histology have been the subject of much study (Poole, 1967; Shellis, 1975; Shellis and Miles, 1976; Meinke, 1982; Nanci et al., 1983; Kawasaki and Fearnhead, 1983; Clement, 1984; Kemp, 1985; Uehara and Miyoshi, 1987; Sasagawa and Ishiyama,
1988; Prostak and Skobe, 1988; Prostak et al., 1989; Sasagawa, 1989; Miake et al., 1991). If both enamel and enameloid are functionally comparable and are also mineralized with a common apatitic biomineral phase, how then may these structures be distinguished? Poole (1967) distinguished enameloid from enamel on the basis of its content of collagen, which is not seen as a component of the enamel protein of the higher vertebrates. Beyond the presence of collagens (principally Type I) and other matrix proteins tentatively identied as enamelins (Graham, 1985; Deutsch et al., 1991), little information exists regarding tissue-specic enameloid matrix components. An apparent difference between enameloid matrices and the enamel extracellular matrix is the presence of matrix vesicles in the former, structures which have never been reported for the enamel matrix. Also, a notable difference lies within the mineral phases, which in some species of sharks and other shes contain a high proportion of a carbonated uorapatite (Suga et al., 1978, 1981).
272
FINCHAM, MORADIAN-OLDAK, AND SIMMER
Beyond these differences, however, a key issue appears to be the presence, or absence, of the amelogenin proteins in the enameloid tooth matrix. Herold et al. (1980), employing immunouorescent techniques with antisera obtained from preparations of bovine enamel matrix, reported the presence of amelogenins in the enameloid of teeth and dermal denticles of Chondrichthyes, in the enameloid of Teleostei and Amphibia, and in the enamel of Reptilia. However, the nonspecic character of the antisera employed in this study provides cause for doubt as to the signicance of these ndings. Kawasaki et al. (1980), in an exhaustive biochemical study of the proteins extracted from the dermal denticles of the blue shark (Prionace glauca), failed to identify any material chemically resembling amelogenins. More recently, Slavkin and Diekwisch (1997), employing RT-PCR techniques, have reported evidence for the expression of amelogenin proteins in the nonmineralized keratinized teeth of the agnathan, Pacic hagsh, although the validity of these observations have been challenged by more recent studies (Girondot and Sire, 1998; Ishiyama et al., 1998). Thus, while differences in both the extracellular tooth matrices and the mineral phases of enameloid and enamel have been described, the details and signicance of these observations remain controversial. (For discussions of the evolutionary origins of enamel, enameloid, and the amelogenins, see Huysseune and Sire, 1998; Smith and Coates, 1998; and Girondot and Sire, 1998.) In this paper our further treatment is limited to the biomineralization of dental enamel.
2. ENAMEL AS A BIOMINERALIZING MATRIX It would seem to be an earth united with a portion of animal substance. (John Hunter, 1771)
As Eastoe has noted (Eastoe, 1979), John Hunter, the English surgeon and anatomist, was perhaps the rst person to recognize that dental enamel contained materials other than solely mineral material. Much later it was established that immature dental enamel contained a substantial proportion (circa 19% by weight) of organic material (Deakins, 1942), establishing this structure as a biomineralizing matrix. Partial amino acid analyses of mature enamel (Block and Bolling, 1952) led to the suggestion that the enamel matrix protein was probably a eukeratin, a concept which appeared reasonable at the time in view of the epithelial origins of the structure or, alternatively, by analogy with bone and dentine, possibly a collagen. Lowenstam and Weiner (1989) have described the processes under which organisms form biominerals
in the following steps: (i) delineation of space; (ii) existence of a preformed organic matrix; (iii) creation of a saturated ion solution; (iv) control over crystal nucleation; (v) control over crystal growth; and, nally, (vi) cessation of crystal growth. These steps result in the formation of biological crystals with unique morphologies and highly ordered structures. Enamel biomineralization also involves these processes, but in a unique manner compared to some other mineralizing systems. In enamel, the matrix for biomineralization is not preformed but is continuously secreted and assembled by the ameloblasts, which play active roles in protein synthesis, secretion, ion transport, and eventually protein matrix resorption. In enamel, we envisage these processes as follows: (i) Secretory ameloblasts (Tomes processes) and the DEJ delineate the enamel space. (ii) Secreted amelogenin proteins assemble, forming a supramolecular structural framework. (iii) Calcium and phosphate ions are transported into the extracellular matrix by the ameloblasts, resulting in a supersaturated solution. (iv) The crystals are nucleated either on preexisting dentin matrix or by other nonamelogenin matrix molecules. (v) Crystal growth, morphology, and orientation are controlled by the matrix (probably amelogenin nanospheres). (vi) Cessation of initial crystal growth is indirectly determined by the eventual degradation and removal of the matrix. (vii) Finally, in enamel, there is a maturation step during which a physical hardening occurs through rapid crystal growth concomitant with controlled protein processing, degradation, and loss. This latter stage is perhaps unique to dental enamel. In enamel, the proteins form an organic matrix framework most of which disappears after it completes its function, leaving behind the highly oriented, largely inorganic, unique enamel structure. The organic matrix appears to self-assemble through a relatively low-energy mechanism (hydrophobic interactions between amelogenin molecules) which does not involve proteinprotein covalent crosslinking reactions as occur in bone collagen. Moreover, this self-assembly process results in the formation of thermodynamically stable structures (i.e., spheres), which appear to be the basic building blocks of the enamel extracellular matrix framework. It is the primary structure of the amelogenin molecule that allows the matrix assembly to occur in such an intriguing and simple manner. The sequence of the amelogenin molecule is characterized as being bipolar with two distinct regions. The carboxyterminal telopeptide (1315 amino acid residues) is highly charged, having an isoelectric point (pI) of about 4.2, compared to the bulk of the molecule with
273
a pI of 8.0. This hydrophilic carboxy-terminal is probably externalized on the amelogenin nanosphere surface creating negatively charged structures which will interact with the forming enamel apatite crystals at a very early stage of enamel formation. Whether the enamel extracellular matrix, as a mineralizing tissue, needs the major acidic macromolecules (Lowenstam and Weiner, 1989, Chap. 2, p. 22) for controlling crystal nucleation and growth, or whether amelogenin uniquely provides both an acidic and a hydrophobic structure within the same molecule is a question requiring further examination.
3. COMPONENTS OF THE MATRIX The enamel component of teeth is such a beautifully organized system of orientated hydroxyapatite crystallites that one must surely expect that here, par excellence, is the place to trace out the part played by collagen, if the enamel matrix really is collagen. That is the question, though, for enamel is very compact and there is comparatively little protein left once the inorganic salts have been removed, . . . (William T. Astbury, 1961)
Piez (1960) obtained the rst complete amino acid analysis of the matrix protein of developing enamel from an unerupted human third molar tooth and concluded that the protein was nearly all soluble in the decalcifying solution . . . and it . . . contained no hydroxyproline or hydroxylysine . . . and . . . was unlike any other known protein. These early analyses were further rened by Eastoe (1963), who reported that foetal enamel protein had a unique composition with several characteristic features . . . distinguishing it from collagen, epidermis, other keratins and proteins from mature enamel. This unique composition was characterized by high proportions of proline, glutamyl, histidine, and leucine residues, and subsequently became recognized as that of the amelogenin protein, which is the dominant component of the early (secretory-stage) enamel matrix (Eastoe, 1965). 3.1. Isolation of the Matrix Proteins
It can be assumed that the embryonic enamelin in deaggregating solvents constitutes a reassociating system of proteins centered around four distinct molecular weight aggregates which are multicomponent. These multicomponent aggregates are also in dynamic equilibrium with high and low molecular weight material. (Gerald Mechanic, 1971)
Further studies established that the developing enamel matrix contained a complex of closely related amelogenin proteins ranging in apparent molecular mass from some 527 kDa (Seyer and Glimcher, 1977a; Fincham et al., 1983). From the mid-1960s through the 1970s, several groups of researchers sought to isolate and further characterize these
amelogenin proteins (Glimcher et al., 1964; Burgess and MacLaren, 1965; Nikiforuk and Simmons, 1965; Mechanic, 1971; Eggert et al., 1973; Papas et al., 1977; Seyer and Glimcher, 1977a; Fincham, 1979; Fukae et al., 1979). Generally, standard protein chemistry procedures (size exclusion, ion-exchange chromatography, and preparative electrophoresis) were applied to the problem of fractionation of preparations of developing enamel proteins extracted from porcine or bovine sources. Frequently, these studies led to incomplete separations and confusing data (see review by Eastoe, 1979). Quite early in this research, evidence suggesting that proteinprotein interactions (aggregation effects) were the confounding factor (Katz et al., 1965), even in the presence of dissociative reagents (Mechanic, 1971). Termine et al. (1980) introduced an important two-step sequential dissociative modication to the usual demineralizing extraction procedures. In this procedure, the immature enamel matrix preparation was initially extracted into a buffered guanidine hydrochloride solution, followed by extraction in the same dissociating medium, but now containing EDTA as a demineralizing agent. Termine et al. (1980) showed that the initial guanidine extraction yielded the bulk of the matrix protein ( 90 % by weight) with an amino acid composition characteristic of amelogenin, while the second stage, demineralizing extraction, produced a preparation with a composition typically enriched in aspartyl, serine, and glycine amino acid residues which was termed enamelin, again after the suggestion of Eastoe (1979). SDSPAGE demonstrated that both of these extracts were heterogenous, with the enamelin preparation being predominantly composed of proteins in the range of 4070 kDa, compared to the 5- to 27-kDa-sized amelogenin fraction. This work is important from two aspects. First, it provided a relatively simple, dened, and reproducible procedure for the separation of the amelogenin fraction from the more acidic (mineral bound) enamelins of the matrix and has subsequently been extensively employed for this purpose. Second, however, the separation obtained by this procedure is certainly not quantitative or completely selective (Fig. 2), which has led to difficulties of data interpretation in experiments where the separation has been assumed to be denitive. Current experience suggests that dissociatively extracted amelogenins will certainly include matrix proteins such as the ameloblastins (see below), while comparable enamelin preparations are certainly heterogenous. In addition, the term enamelin, as applied to the dissociatively extracted prepa-
274
3.2. The Amelogenins

Hence, the intact extracellular murine amelogenin is 180 amino acid residues long with a calculated molecular weight of 23,532 Da. (Snead et al., 1985)
FIG. 2. Sequential dissociative extraction of bovine enamel matrix. Fetal bovine enamel scrapings (8.2 g, moist weight; circa 6 months in utero) were sequentially extracted according to the protocol of Termine et al. (1980). Extracts were desalted (Biogel P2 column), lyophilized, and weighed. Assuming 16% water (based on dry-weight estimation) and 70% inorganic content (based on ash-weight determinations), total matrix protein equals 2066 mg. Recovery of protein, 1383 mg (67%). Amelogenin fraction, 97.2%. (A) Recoveries of sequential extracts ( 90% in extract 1). Transition from guanidine HCl to guanidine HCl EDTA is between extracts 5 and 6 (double arrow). (B) Fifteen percent SDSPAGE of selected extracts from the separation shown in A (silver staining). Note the increase in the proportion of higher molecular weight protein (enamelin fraction) in extracts 6 and 8. Also note the apparent presence of amelogenin-sized (26-kDa marker) material still present in the demineralized extracts.
ration, must now be clearly distinguished from the matrix protein of that name (see below). In the sections which follow, an outline of the history and characteristics of the several matrix proteins is presented.
Amelogenin has been shown to be critical for normal enamel formation. The human amelogenin gene is expressed primarily from a single gene on the X chromosome, although minor expression ( 10%) from the Y-chromosomal copy of the amelogenin gene has been detected (Lau et al., 1989; Fincham et al., 1991a; Salido et al., 1992). The X-chromosomal copy of the human amelogenin gene has been linked to amelogenesis imperfecta or AI (Lagerstrom et al., 1990), a disease characterized by inherited enamel defects. Eight different amelogenin mutations have been characterized in kindreds suffering from AI (Lagerstrom et al., 1990; Aldred et al., 1992; Lench and Brook, 1997; Lagerstrom-Fermer et al., 1995; Lench and Winter, 1995; Collier et al., 1997). In rodents, amelogenin expression has been specically and articially interrupted using both antisense oligonucleotides (Diekwisch et al., 1993) and ribozyme approaches (Lyngstadaas et al., 1995). In both cases knockout of amelogenin resulted in defective enamel formation. These data strongly support a specic and highly conserved functional role for amelogenins in enamel biomineralization. 3.2.1. Sequences. Amino acid sequences of amelogenins were initially determined by Edman sequencing of the puried proteins and later derived from DNA sequences of amelogenin cDNAs. Protein sequencing has been reported for amelogenins isolated from pigs (Fukae et al., 1979, 1980; Fukae and Shimizu, 1983; Yamakoshi et al., 1989), cattle (Seyer, 1972; Zalut et al., 1980; Fincham et al., 1981; Takagi et al., 1984; Shimokawa et al., 1987), humans (Fincham et al., 1983; Catalano-Sherman et al., 1994), rabbits (Zeichner-David et al., 1988), hamsters (Lyaruu et al., 1998), opossums (Hu et al., 1996a), guinea pigs (Cerny and Hammerstrom, 1998, 1999), and, most recently, platypuses, echidnas, caimans, and toads (Toyasawa et al., 1998). These data have revealed a striking level of homology of sequence between species (Simmer and Snead, 1995), especially in the amino- and carboxy-terminal regions (see Fig. 3). Further, the amelogenins have been shown to be completely tissue specic to the ameloblast cells of the dental enamel organ. Comparison of the known amelogenin sequences has shown that, while variations in sequence are found in the more central region (core) of the molecule, motifs of the primary structure are almost completely conserved at both the carboxy- and the amino-terminals, suggesting the conservation of specic functional motifs. Indeed, a recent study of a case of human X-linked amelogenesis imperfecta has
275
FIG. 3. Comparison of amino acid sequences. The rst 50 residues of the N-terminal region of amelogenins are shown with the conserved TRAP sequences (residues 145) and the conserved proteolytic TRAP cleavage locus -GW-. The six conserved tyrosine residues of TRAP are shown in bold. Sources of data are as follows: cattle (Seyer, 1972; Zalut et al., 1980; Fincham et al., 1981; Takagi et al., 1984; Gibson et al., 1991); pig (Fukae et al., 1979; 1980; Fukae and Shimizu, 1983; Yamakoshi et al., 1989; Hu et al., 1996b); human (Fincham et al., 1983; Salido et al., 1992; Catalano-Sherman et al., 1994); rabbit (Zeichner-David et al., 1988; Fincham, 1999, unpublished data); mouse (Lau et al., 1992; Simmer et al., 1994a; Simmer, 1995; Hu et al., 1997a); rat (Bonass et al., 1994a,b; Li et al., 1995); opossum (Hu et al., 1996a); hamster (Lyaruu et al., 1998); platypus, echidna, caiman, and toads (Toyasawa et al., 1998); guinea pig (Cerny and Hammerstrom, 1998, 1999).
revealed the molecular lesion as being a point mutation in the amelogenin X gene leading to the transformation of proline-40 (Human AMGX) to a threonine residue (Collier et al., 1997). The principal opossum amelogenin isoform of 202 residues (Hu et al., 1996a) includes a striking proline-rich repetive sequence of 27 residues: -QPIQPIQPIQPIQPMQPMQPMQPMQPM- within exon 6, an equivalent of which is not seen in mouse, rat, human, pig, or hamster amelogenins, although similar motifs are present in both bovine and guinea pig amelogenins. Such repetitive proline-rich sequences are suggested to function in reversible proteinprotein interactions (for a review of this topic see Williamson, 1994). The rabbit (Lagomorpha) appears to express an amelogenin (Zeichner-David et al., 1988) which may differ signicantly from others so far examined in containing elevated levels of alanine and a reduced content of glutamyl residues (Levine et al., 1967; Wright and Eastoe, 1989), although denitive amino acid sequence data have yet to be obtained (see Fig. 3).
3.2.2. Alternative splicing. The characterization of amelogenins isolated from developing enamel has revealed that amelogenins are the translation products of alternatively spliced RNA messages and that most amelogenin polypeptides have been processed by proteinases. The alternative splicing of amelogenins was rst proposed to explain the origin of the leucine-rich amelogenin polypeptide (LRAP) which was identical to the major amelogenin protein at its amino and carboxyl-termini but lacked a large segment from the center of the protein (Fincham et al., 1981; Sasaki et al., 1984). The acronym LRAP was rst used to describe two leucine-rich amelogenin polypeptides (LRAP-1 and LRAP-2) isolated from developing bovine enamel (Fincham et al., 1981). These LRAP molecules were sequenced and found to have the same sequence of 42 and 46 amino acid residues (LRAP-2 having 4 additional residues at its carboxyterminal end). Initially, it was presumed that these peptides, as was the case for TRAP (the amino-terminal 44/45 residue sequence of the principal amelogenin), were the
276
amino-terminal fragments of a larger ( 20 kDa) unidentied amelogenin. However, the later study by Gibson et al. (1991) showed that LRAP in fact is translated from a shorter mRNA that has the coding regions from exons 4, 5, and part of 6 deleted during splicing (Bonass et al., 1994a). The sequence determined by Gibson et al. (1991) codes for 59 amino acid residues, including the conserved carboxy-terminal 13 residues: -WPATDKTKREEVD. Thus, LRAP, as originally described, is the proteolytically processed product of the alternatively spliced amelogenin identied by Gibson et al. (1991). For the purposes of this review, LRAP is now used to refer to the full-length (59-residue) amelogenin. Alternatively spliced amelogenin mRNAs have now been isolated from mouse (Lau et al., 1992; Simmer et al., 1994a, 1995; Hu et al., 1997a), rat (Bonass et al., 1994a,b; Li et al., 1995), hamster (Lyaruu et al., 1998), guinea pig (Cerny and Hammarstrom, 1998, 1999), pig (Hu et al., 1996b), cattle (Gibson et al., 1991), opossum (Hu et al., 1996a), and human (Salido et al., 1992). For reviews see Sasaki and Shimokawa (1995), Simmer and Snead (1995), Brookes et al. (1995) and Gibson et al. (1998). In each species investigated there appears to be a major expressed amelogenin isoform ranging in size from 173 (pig) to 210 amino acids (guinea pig) (Cerny and Hammarstrom, 1998, 1999). At the mRNA level, the LRAP message is about 10% as abundant as the message for the major amelogenin isoform, which is probably also a good estimate for the relative abundance of the expressed protein (Yuan et al., 1996). Messages encoding the major amelogenin and LRAP isoforms have been cloned from every organism under investigation, with the exception of LRAP in humans, although the cloning of amelogenin cDNA in humans used a strategy that would not have detected the LRAP message (Salido et al., 1992). Most amelogenins differ from the major expressed amelogenin isoform by lacking an internal (core) region. The most noteworthy exceptions to this are isoforms containing a unique hydrophilic 14-amino acid segment (-KSHSQAINTDRTAL-) encoded by mouse exon 4. Amelogenins containing the exon 4-encoded segment have been detected in developing enamel by immunohistochemistry (Simmer et al., 1994a) and Western blot analyses (Simmer et al., 1994b), although they represent a small fraction (roughly 1%) of total expressed amelogenin protein (Ryu et al., 1996). Other amelogenin isoforms have been detected using antipeptide antibodies, and the expression of some amelogenins appears to be developmentally regulated (Gibson et al., 1995). Recent studies (Li et al., 1998) have reported the presence of two additional 3 amelogenin exons. The functional
signicance of these multiple amelogenin isoforms remains a complete mystery. 3.2.3. Posttranslational modications. (i) Phosphorylation. Takagi et al. (1984) reported that the bovine amelogenin contained a single phosphorylated residue: serine-16. Takagi et al. measured the phosphate content in the intact bovine amelogenin, in each tryptic fragment, and found 1 mol of phosphate in the intact molecule and in a tryptic fragment comprising the rst 24 amino acid residues (Takagi et al., 1984). This observation appeared to be at variance with other early reports, which suggested that the amelogenin polypeptides (E3 and E4) each contained three phosphorylated residues (Seyer and Glimcher, 1977b; Papas et al., 1977). Later studies (Fincham et al., 1981) identied the bovine peptide E4 as being TRAP, the amino terminal section of the principal (Mr, 20 kDa) bovine amelogenin, which might then be presumed to also contain at least three phosphorylated residues. Attempts to verify this nding using direct sequencing methods (Fincham et al., 1992) were not successful and the enigma remained until mass spectrometric methods became available to the protein chemistry world. Fincham and Moradian-Oldak (1993) used HPLC methods to isolate the TRAP and LRAP polypeptides from both bovine and porcine enamel matrices and determined their masses using electrospray ionization. It was found that, in each case, the experimental mass could only be correlated ( 2.0 Da) with the computed mass, based on the known sequence if a single phosphorylated residue was assumed. Further, mass analysis of tryptic peptides derived from cleavage of the bovine LRAP also identied this phosphorylated residue as being contained within the (24-residue) amino-terminal tryptic peptide (MPLPPHPGHPGYINFSYEVLTPLK-), while a fast-atom bombardment mass spectrometric analysis of this peptide identied fragments consistent with the presence of a phosphorylated serine at position 16, observations which were in accord with the earlier report of Takagi et al. (1984). More recently, multiple mass spectrometric analyses of amelogenins from many species have consistently given mass values matching a large number of the alternatively spliced amelogenin isoforms and their cleavage products, but only after adding the mass of a single phosphate group (80 Da) to the molecular mass of amelogenin derived from its cDNA sequence (Fincham and Moradian-Oldak, 1993, 1996; Fincham et al., 1994a; Hu et al., 1996a; Lyaruu et al., 1998; Ryu et al., 1998a). Mammary gland casein kinase recognizes the sequence motif xSxEx (Kemp and Pearson, 1990) and might thus be expected to phosphorylate the amelogenin serine-16 (. . . NFSYEK . . .). However, recent
277
studies have indicated that this amelogenin phosphorylation is affected by a specic enamel matrix protein kinase (Salaih et al., 1998). (ii) Glycosylation. Earlier studies of enamel matrix amelogenin preparations obtained using the guanidine extraction procedure (Termine et al., 1980) have been shown to include glycosylated proteins (Nanci et al., 1989). However, as noted above, this operational procedure is not exclusive for amelogenin extraction and, while the primary sequence of amelogenin displays potential targets sites for both Nlinked and O-linked glycosylations (Asn-14 and Ser16), studies employing digoxigenin immuno-glycan labeling (Fincham et al., 1991b) failed to detect any glycosylations of either murine or bovine amelogenin preparations. Akita et al. (1992) employed a range of lectins to probe for glycosylation of porcine enamel proteins and found such modications to be conned to the nonamelogenin matrix proteins. Further, as noted above, multiple mass spectrometric analyses of amelogenins return exact mass numbers without assuming any glycosylations (Fincham et al., 1994a). 3.2.4. Amelogenin secondary and tertiary structures. Since the discovery (Piez, 1960; Eastoe, 1960) that the developing dental enamel matrix contained an apparently unique proline-rich protein (amelogenin), attempts have been made to elicit secondary and tertiary structural information in the belief that this protein was involved in the enamel biomineralization process, either as a primary nucleator of mineral or as a structural entity. Attempts have been made to obtain secondary and tertiary structural information for amelogenin proteins using classical structural analysis techniques, and, as early as 1965, poorly resolved X-ray diffraction images of developing human tooth enamel matrices were obtained showing diffuse, nonorientated possible - or cross- -patterns (Pautard, 1961, 1963; Bonar et al., 1965; Fincham et al., 1965). CD and NMR studies have led to suggestions that the molecule contains both -structures and -turns (Goto et al., 1993; Zheng et al., 1987; Renugopalakrishnan et al., 1989). Jodaikin et al. (1986, 1987) in an X-ray diffraction study, considered that the -sheet reections obtained from enamel matrix proteins were more likely derived from the nonamelogenin (enamelin) components than the amelogenins. Renugopalakrishnan et al. (1989) proposed that the -spiral structure computed for the remarkable (Pro-Gln-X)n repetitive sequence identied in the bovine amelogenin (Takagi et al., 1984) might be involved in Ca2 transport. However, subsequent amelogenin sequence determinations have shown this structural feature to be absent from human, mouse, rat, hamster, and pig amelogenins, although a similar, even more striking, repetitive feature has
recently been identied in the opossum and guinea pig amelogenins (Hu et al., 1996a; Cerny and Hammarstrom, 1998, 1999). Termine and Torchia (1980), in a solid-state NMR study of a sample of manually dissected bovine developing enamel matrix, reported a high ( 70%) level of rapid and unrestricted mobility. Aoba et al. (1990), in a 1H-NMR study of porcine amelogenins in solution, showed that structural changes appear to accompany the proteolytic processing transformations from the initial full-length 25kDa (173 residue) molecule to the lower molecular weight products. Further, they reported that the carboxy-terminal Trp-161 and some of the tyrosine residues of the amino-terminal TRAP sequence are exposed at the molecule surface. Recent application of small-angle X-ray scattering techniques have suggested that the porcine 20-kDa amelogenin may exist in solution as an elongated bundle structure (Matsushima et al., 1998). 3.2.5. Localization of amelogenin expression. Amelogenin is an enamel-specic protein expressed by cells of the ameloblast lineage beginning just before the initial mineralization of dentin (Inai et al., 1991; Bronckers et al., 1993) and terminating in the early maturation stage (Wakida et al., 1999). The expression of amelogenin during enamel formation has been extensively reported using in situ hybridization (Snead et al., 1988; Wurtz et al., 1996; Wakida et al., 1999) and immunohistochemistry (Inai et al., 1991; Uchida et al., 1991a; Fukae et al., 1993; Nanci et al., 1994, 1998; Simmer et al., 1994a,b; Gibson et al., 1995). This latter study showed that amelogenin signals were generally weak and located close to secretory cell surfaces, while ameloblastin signals (see section 3.3.4) showed the inverse effect, perhaps suggesting (i) that nascent amelogenins are occluded in some manner from immunoprobes and (ii) that ameloblastin molecules are rapidly processed to fragments within a short distance (1.25 m) of the cell surfaces. 3.2.6. Lectin-like properties. A recent study (Ravindranath et al., 1999) has shown that both native and recombinant amelogenins hemagglutinate (HA) mouse red blood cells. This HA is inhibited by monomers, dimers, and tetramers of GlcNAc (N-acetylglucosamine) but not by N-acetylgalactosamine or related sugars. This activity is retained by the TRAP molecule but not by LRAP (which shares 33 N-terminal amino acids of the 45-residue conserved TRAP sequence). It was shown that [14C]GlcNAc binds to the N-terminal sequence of TRAP (-PYPSYGYEPMGGW) but not when the three tyrosyl residues are substituted with phenylalanine or if the third proline is substituted with threonine. Signicantly, this latter modication mimics a point mutation recently identied in a case of human
278
X-linked amelogenesis imperfecta (Collier et al., 1997). It appears possible that this lectin-like activity of the TRAP motif of amelogenins may be functionally involved in interactions with enamel matrix glycoproteins (enamelin, tuftelin, or ameloblastin), perhaps promoting structural stability of the matrix or, alternatively, functioning in a signaling role through recognition by cell surface glycoproteins (Akita et al., 1992). Further studies of the signicance of these interesting observations to enamel matrix biomineralization are needed. 3.2.7. Physicochemical properties of amelogenins. The development of expression systems for recombinant amelogenins (Simmer et al., 1994a,b), has greatly advanced our ability to not only investigate some structural features, such as the putative amelogenin molecular self-assembly process (MoradianOldak et al., 1994b, 1998c); specic cleavage sites on amelogenin molecules (Moradian-Oldak et al., 1994a; Ryu et al., 1999), and mineral binding properties (Simmer et al., 1994b; Moradian-Oldak et al., 1998a; Hunter et al., in press; Ryu et al., 1998b), but also to examine the physico-chemical properties of the amelogenin molecule and those of its partially processed products. Tan et al. (1998) made a detailed study of the solubility properties of both recombinant and in vivo amelogenins at a range of solution pHs. These studies were directed at providing baseline information for the further study of amelogenin structure, crystallization, and proteinmineral interactions. The solubility of amelogenins was found to be strongly dependent on the protein primary structure, solution pH, and ionic strength and, also, to some extent, on the buffer composition. The solubility of the full-length recombinant amelogenin (rM179) and that of the modied amelogenin (rM166) was similar in acidic solutions, but the latter was much less soluble in basic solutions (Fig. 4). This behavior has been attributed to the lack of the hydrophilic carboxy-terminal sequence in the rM166 protein (Moradian-Oldak et al., 1994b). The native porcine 25-kDa amelogenin solubility showed a behavior as a function of pH similar to that seen for rM179. The TRAP was found to be very insoluble at all pH ranges examined, as previously reported (Seyer and Glimcher, 1977b), while the LRAP molecule was even more soluble than rM179. Taking into account the local pH changes reported to occur in the enamel extracellular microenvironment (Sasaki et al., 1991a), the strong dependence of amelogenin solubility on pH may have signicance to the following extracellular matrix events: (i) proteolytic processing activities, (ii) amelogenin quaternary structures, (iii) association of the proteins with the mineral phase, and (iv) the kinetics of enamel biomineralization. In another study, Ryu et al. (1998b) have reported that
FIG. 4. Solubility properties of amelogenins. Comparison of rM179 and rM166 solubility with native porcine (25-kDa) amelogenin, in 0.05 potassium phosphate buffer from pH 4 to 9, IS 0.15 M, 25C. The solubility of all the amelogenins examined was minimum at a pH near their isoelectric point and increased rapidly below and above. The recombinant amelogenin rM166 was found to be less soluble than rM179 at pH 8. The solubility behavior of amelogenins was found to be independent of the presence of divalent metal ions such as calcium and magnesium in the buffer (Tan et al., 1998).
amelogenins in solution may also act as a proton sink, promoting nascent OCP mineral conversion to apatite. 3.3. The Ameloblastins (Amelin/Sheathlin)
. . . we report on one of these (mRNA) sequences, which we found highly expressed in ameloblasts. . . . The sequence is represented by two mRNA variants with the potential to code for proteins which we have named amelins (ameloblastins). The amelins are likely to constitute components of the extracellular enamel matrix. (Cerny et al., 1996)
3.3.1. Sequences. The amino and carboxyl ends of the protein that would become called ameloblastin are different biochemically and were discovered separately during investigations of pig enamel proteins. The carboxyl end was represented by two polypeptides with apparent molecular weights of 27 and 29 kDa, isolated during a search for enamel proteins that bound calcium by monitoring fractions during a progressive purication process for bands that stained blue with the Stains-All dye (Fukae and Tanabe, 1987a). The amino-terminal end of ameloblastin was discovered during a search for the existence of nonamelogenins like enamelin (Fukae and Tanabe, 1985, 1987b). The amino-terminus of ameloblastin is represented in the enamel matrix by a group of low molecular weight cleavage products in the range 1317 kDa with aggregative properties (Fukae and Tanabe, 1987b). The amino acid composition of ameloblastin was distinctive and suggested that the protein was neither amelogenin nor enamelin (Fukae and Tanabe,
279
1987b). Polyclonal antibodies raised against the 13to 17-kDa nonamelogenins gave a unique prole in both Western blot and immunohistochemical analyses (Uchida et al., 1991a), immunostaining an array of enamel proteins in the ranges 1020 and 3040 kDa (Uchida et al., 1991b). Especially interesting was that the protein immunolocalized throughout the entire thickness of the developing enamel layer and, except for the most supercial enamel (roughly 30 m beneath the ameloblasts), showed a honeycomb staining pattern (Uchida et al., 1991a; Fukae et al., 1993). This honeycomb pattern was due to the concentration of ameloblastin cleavage products in the sheath (or interprismatic) space that partially separates rod and interrod enamel. Lectin binding studies indicated that the 13- to 17-kDa nonamelogenins specically bound MPA (Maclura pomifera) lectin, and histochemistry using uorescent-labeled MPA also showed a honeycomb pattern throughout the developing enamel layer (Akita et al., 1992). In adsorption experiments the 13- to 17-kDa nonamelogenins showed low hydroxyapatite affinity (Akita et al., 1992). Edman degradation revealed the rst amino acid sequence for ameloblastin, and affinitypuried antipeptide antibodies against the aminoterminal sequence showed the same honeycomb pattern as described earlier (Uchida et al., 1995). On the basis of the novel immunostaining pattern, it was proposed that these low molecular weight nonamelogenins and their parent protein constitute a new family of enamel proteins designated as sheath proteins (Uchida et al., 1995). 3.3.2. Cloning of ameloblastin cDNAs and genes. In 1996, two research groups, randomly screening rat tooth-specic cDNA libraries, independently cloned and characterized cDNAs encoding the rat homologue of the protein that had been studied for over 10 years in Japan. The protein was termed ameloblastin (Krebsbach et al., 1996) and amelin (Cerny et al., 1996). Later, the cDNA from pig was cloned and designated as sheathlin (Hu et al., 1997b). A cDNA encoding the mouse homologue was cloned and also called ameloblastin (Simmons et al., 1998). (Note: Ameloblastin is the name used in this paper.) Ameloblastin gene loci are on chromosome 5 in the mouse (Krebsbach et al., 1996) and chromosome 4q12 in humans (MacDougall et al., 1997). The human gene is within a region previously linked to a local hypoplastic form of amelogenesis imperfecta in three Swedish families (Forsman et al., 1994). Two ameloblastin isoforms are synthesized from alternatively spliced mRNA transcripts and differ by the deletion or inclusion of a 15 amino acid segment that is absolutely conserved between the pig and the rat (Hu et al., 1997b).
3.3.3. Bipolarity. Based upon the pig ameloblastin cDNA sequence, the rst 129 residues of the protein, which correspond to a 15-kDa cleavage product, have a calculated pI of 10.6, while the 66 residues at the carboxyl-terminus have a calculated pI of 4.5 (Hu et al., 1997a). The 95 residues at the carboxyl-terminus of ameloblastin correspond to the 27- and 29-kDa calcium binding proteins (Murakami et al., 1997). Based upon a loss of immunoreactivity in the Golgi when using an antipeptide antibody raised against residues 385399 (EMTMDSTATPYSEHT), it has been proposed that this segment contains an O-linked glycosylation site (Uchida et al., 1998). Affinity-puried anti-peptide antibodies specic for a segment near the carboxylterminus demonstrated that intact pig ameloblastin and cleavage products containing the carboxylterminus are highly concentrated within 2 m of the ameloblast Tomes process, and from there to a depth of about 50 m produced a reverse honeycomb pattern (immunostaining the rod and interrod enamel but not the sheath space), and showed no staining in the deeper enamel (Murakami et al., 1997). Therefore, the amino- and carboxyl-termini of ameloblastin show differences in charge (based upon their isoelectric points) and location in the developing enamel matrix. 3.3.4. Localization of ameloblastin expression. Ameloblastin is a tooth-specic protein. It is expressed by cells derived from the inner enamel epithelium, including cells of the epithelial root sheath, presecretory, secretory, and maturation ameloblasts, as well as transiently in preodontoblasts (Cerny et al., 1996; Fong et al., 1996a,b, 1998; Krebsbach et al., 1996; Lee et al., 1996; Hu et al., 1997a; Uchida et al., 1997, 1998; Begue-Kirn et al., ` 1998; Nanci et al., 1998). 3.4. The Enamelins
We can therefore describe non-amelogenins only in terms of molecular size, amino acid composition, modications (e.g. phosphorylation), immunoreactivity and to some extent localization within the tissue. In these terms non-amelogenins vary considerably. (Robinson et al., 1989a)
Enamelin was the name used to designate a class of nonamelogenin enamel proteins that strongly adsorbed to enamel crystals and were not released from the matrix until the crystallites were dissolved (Termine et al., 1980). [Note: In earlier studies (Mechanic, 1971) the term enamelin was applied generally to any developing enamel matrix protein, but this usage was not subsequently retained (see Eastoe, 1979).] The term has since come to be used for a specic enamel protein which, based upon its immunolocalization in developing enamel, has the crystal binding properties originally associated with
280
the enamelin class. Enamelin is not tuftelin, it is not serum albumin, and it is not a proteinase. Enamelin is a novel protein with an amino acid sequence unrelated to any other in the databases. It is the largest known enamel protein, and its expression is highly restricted to developing teeth. In the pig, intact enamelin immunolocalizes within 2 m of the ameloblast Tomes process and has an apparent molecular mass of 186 kDa (Hu et al., 1997c). Enamelin cDNAs have now been cloned and characterized from pigs (Hu et al., 1997c), mice (Hu et al., 1998a), and humans (Hu et al., 1998b). Unlike other enamel proteins, no enamelin isoforms that are translated from alternatively spliced RNA transcripts have been observed. Minus the signal peptide and any posttranslational modications, pig enamelin has 1104 amino acids, a molecular mass of 124.3 kDa, and a pI of 6.5. Mouse enamelin has 1236 amino acids, a molecular mass of 137 kDa, and a pI of 9.4. Mouse enamelin is distinctive in that it contains 14 copies of an 11 amino acid segment tandem repeated in the protein. This segment is not repeated in the pig and human proteins. 3.4.1. Processing. As is true of all enamel proteins, enamelin is processed by proteinases shortly after being secreted. Only its cleavage products have been isolated and characterized. Enamelin cleavage products were rst isolated and characterized in the same investigation that yielded the amino-terminus of ameloblastin (Fukae and Tanabe, 1987b). Like amelogenins, enamelin is processed from its carboxylterminus. Large enamelin cleavage products of 155, 142, and 89 kDa apparent molecular weight have been characterized which retain the original aminoterminus (Fukae et al., 1996). The 89-kDa enamelin accumulates and is further processed to generate the 32- and 25-kDa enamelins (Fukae et al., 1996). The best-studied enamelin cleavage product has an apparent molecular mass of 32 kDa on SDSPAGE (Tanabe et al., 1990; Uchida et al., 1991a; Yamakoshi, 1995; Yamakoshi et al., 1998). The pig 32-kDa enamelin has 106 amino acids (residues 174279), which include two phosphoserines and three glycosylated asparagines (Yamakoshi, 1995; Fukae et al., 1996; Yamakoshi et al., 1998). In the absence of glycosylations, the 32-kDa phosphoprotein has a derived molecular mass of 11.8 kDa and pI of 6.4. Although the protein shows a single band on SDS PAGE (which stains blue with Stains-All dye), twodimensional gel electrophoresis detects seven different constituents in this band, having pI values ranging from 3 to 4.5 (Tanabe et al., 1990). This heterogeneity is attributed to variable glycosylation. The 32-kDa enamelin is easily released from secretory stage enamel by extraction in 50 mM phosphate
(pH 7.4), and may represent about 1% of total enamel protein (Tanabe et al., 1990). 3.4.2. Immunolocalization. Immunostaining the developing enamel matrix with antibodies raised against puried 89- and 32-kDa enamelins, as well as affinity-puried anti-peptide antibodies against the amino-terminus of the 32-kDa enamelin, show that enamelin specically localizes in secretory stage enamel from the dentino-enamel junction (DEJ) to the surface of the tooth and rapidly disappears during the early maturation stage (Uchida et al., 1991a,b; Fukae et al., 1993). Based upon amino acid composition analyses, the nding that enamelin is fully degraded during early maturation casts doubt upon the earlier interpretation that the amelogenins are selectively degraded and that enamelins are retained in the maturation enamel. The pattern of immunostaining in secretory stage enamel was a reverse honeycomb; that is, enamelin was restricted to the rod and interrod enamel and absent from the sheath space. This pattern suggests that these enamelin cleavage products bind enamel crystallites in vivo. This nding is supported by hydroxyapatite binding studies in vitro, but the binding studies are hard to extrapolate back to the in vivo situation because they were performed in buffers containing 4 M guanidine (Tanabe et al., 1990). On the other hand, the observation that the 32-kDa enamelin can be extracted with neutral 50 mM phosphate buffer may suggest weak hydroxyapatite binding in vivo. 3.5. Tuftelin
. . . ameloblasts, being ectodermal cells, might be expected to produce an epithelium-like protein. Tuft protein might therefore be a primitive epithelial component of a more specialized enamel matrix. (Robinson, Lowe and Weatherell, 1975)
Enamel crystals originate at the DEJ and appear to extend uninterrupted to the surface of the tooth. This means that if enamel crystals nucleate on macromolecules and there are grounds to consider the alternative mechanism of homogenous nucleation as being unlikely (Mann, 1989; Simmer and Fincham, 1995), then the enamel crystal nucleator should be found at the DEJ. The requirement for an enamel-specic nucleator is not certain, however, as it is presently unresolved whether enamel crystallites are independently nucleated (Diekwisch et al., 1994) or if they extend from preexisting dentin crystals (Arsenault and Robinson, 1989). Extending from the DEJ into the overlying enamel layer in mature teeth are two types of structures: enamel spindles, which are extensions of odontoblastic processes, and enamel tufts (Weatherell et al., 1968), which may represent residual enamel matrices that include the nucleator of enamel crystals (Palamara et al., 1989; Robinson et al., 1989b). Some data
281
suggested what the enamel nucleator should look like. For instance, the amino acid compositions of the trichloracetic acid (TCA)-soluble and -insoluble tuft protein (total protein from the DEJ) of mature bovine and human teeth were known. Assuming the tuft is essentially made up of a single protein, its amino acid composition was known (Robinson et al., 1975). The nucleator might also be expected to display acidic or phosphorylated amino acid side chains in a region of -sheet secondary structure to register with the array of phosphate groups on the incipient crystal (Addadi and Weiner, 1985, 1989; Jodaikin et al., 1986, 1987; Addadi et al., 1989, 1992; Moradian-Oldak et al., 1992). 3.5.1. Structure and function. A cDNA encoding a protein called tuftelin that may or may not play a role in the nucleation of enamel crystallites has been characterized (Deutsch et al., 1989, 1991). The tuftelin clone was isolated from a cDNA expression library using an affinity-puried polyclonal antibody raised against the major 66-kDa protein in the guanidine/EDTA fraction of developing bovine enamel (Deutsch et al., 1987). The deduced amino acid sequence of tuftelin ts the expected prole for tuft protein. It was somewhat acidic, and secondary structure predictions suggested that it contained -sheet. It was novel and showed no signicant homology to any previously characterized protein. It had an amino acid composition that agreed with that determined for TCA-insoluble tuft protein. Its predicted molecular mass of 43.8 kDa (given the presence of a single potential N-linked glycosylation site) was consistent with that of a cDNA encoding the original 66-kDa targeted protein, and affinitypuried anti-peptide antibodies made against three regions of the deduced tuftelin sequence selectively immunostained the DEJ and proteins in a guanidine/ EDTA extract of enamel proteins (Deutsch et al., 1991, 1995, 1998; Zeichner-David et al., 1997). Given these ndings, tuftelin has been advanced as a prime candidate protein for the nucleator of enamel crystals (Deutsch et al., 1989, 1991). The human tuftelin gene has been cloned and localized to chromosome 1, and efforts are underway to screen pedigrees with inherited enamel defects for linkage to the tuftelin gene (Deutsch et al., 1994). However, additional data concerning the tuftelin gene and protein (Bashir et al., 1998) have not pointed in the same direction as that of the early ndings. The bovine tuftelin gene was cloned and characterized and showed that the original tuftelin cDNA was missing a G, which changed the reading frame for the deduced protein. The shift removed the C-terminal 92 amino acids and replaced them with only 42 (Bashir et al., 1997). The revised reading frame was also found in mouse and pig tuftelin cDNA
(MacDougall et al., 1998b). It is clear that the original bovine tuftelin cDNA clone was sequenced correctly (Deutsch et al., 1998), but the origin of the deletion is unclear. Was a nucleotide deleted during reverse transcription when making the expression library, or could this nucleotide be deleted during normal transcription and lead to A and B isoforms of tuftelin? The original assignment of the bovine tuftelin start codon was tentatively made, recognizing that other start codons were possible (Deutsch et al., 1989). Information gained from the mouse tuftelin cDNA makes the original start codon assignment appear unlikely. Revising the start codon to a more 5 position adds 52 amino acids to the tuftelin aminoterminus (MacDougall et al., 1998b). Changes in the deduced amino acid sequence of tuftelin makes it more dissimilar to the determined tuft protein amino acid composition. There is no obvious signal peptide at the amino-terminus of tuftelin (regardless of which start codon you assign), and Northern blot analyses have detected tuftelin mRNA in many nonmineralizing tissues (MacDougall et al., 1998b). To resolve these controversies surrounding tuftelin and its role in enamel biomineralization, the tuftelin protein should be isolated and characterized. This may not be as easy as it would seem. Tuftelin has never been isolated from the enamel matrix. It is apparent that the protein encoded by the tuftelin cDNA is not the major 66-kDa enamelin that was originally used to make antibodies (Deutsch et al., 1987). Anti-peptide antibodies raised against the deduced tuftelin sequence immunostain primarily the DEJ and only show faint staining of proteins in enamelin fractions on Western blots (Deutsch et al., 1991; Zeichner-David et al., 1997). 3.6. Tuft Protein
Judging from amino acid composition, the so-called tuft protein . . . seems to have been present in every type of tooth examined. . . . (Robinson et al., 1975)
Enamel tufts are a histological feature which appear as sinuous bundles of bers at the DEJ in ground sections of mature human enamel (Osborn, 1969). Weatherell et al. (1968) demonstrated the presence of an insoluble proteinaceous material at the tooth DEJ. Robinson et al. (1975) have shown that complete demineralization of mature human (and bovine) enamel yields an insoluble tuft-like protein residue which is identied with the structures observed histologically and may account for some 0.010.06% by weight of the mature enamel. Amino acid analyses of this material showed it to have a relationship to the nonamelogenin proteins (Robinson et al., 1989a). Interestingly, immuno-crossreactivity was shown to exist between the tuft protein and a polyclonal antikeratin antibody (Robinson
282
et al., 1975, 1989c). Amizuka and Ozawa (1989) and Amizuka et al. (1992) showed that polyclonal antibodies raised to an enamelin preparation also crossreacted with tuft proteins. Most recently, Robinson et al. (1998a), using demineralized human molars, have shown that tuft protein contains partial sequences of ameloblastin. Currently, the relationship between the rather insoluble tuft protein and other nonamelogenin proteins tuftelin, ameloblastin, or enamelinsis obscure. 3.7. The Sulfated Proteins
It is very likely that sulfated enamel proteins . . . have been part of certain fractions previously classied as enamelins (non-amelogenins). (Smith et al., 1995a)
Employing 35S in vivo labeling, Smith et al. (1995a) have demonstrated the presence of a group of shortlived sulfated matrix proteins in developing rat enamel. These components (49 and 25 kDa) are apparently proteolytically generated from a 65-kDa precursor protein and rapidly degraded after secretion. Smith et al. (1995a) have suggested that these sulfated enamel glycoproteins may interact functionally with nascent amelogenins, a suggestion which may have greater signicance in view of the recently identied, lectin-like properties of the amelogenins (section 3.2.6). However, the function of these transient sulfated proteins in enamel biomineralization is presently unknown. 3.8. The Proteinases
This method showed that the whole layer of forming enamel matrix was associated with a well dened clear area produced by the proteolytic digestion of gelatin substrate . . . (Suga, 1970)
The nature and sequence of proteolytic activities during enamel biomineralization are critical and somehow unique to this mineralizing tissue. First, these activities cause changes in structural and physicochemical properties of the parent enamel matrix proteins (i.e., cleavage of the hydrophilic carboxy-terminal of amelogenin (Moradian-Oldak et al., 1994a; Tanabe et al., 1992). Second, the hardening of enamel, resulting from rapid crystal growth during the maturation stage, appears to be dependent on the complete degradation of the extracellular organic matrix (Eastoe, 1979; Robinson and Kirkham, 1984, 1985). Metalloproteinases and serine proteinase activities have been detected in the enamel extracellular matrix by various investigators (Moe and Birkedal-Hansen, 1979; Overall and Limeback, 1988; Smith et al., 1989a, 1995b, 1996a; Sasaki et al., 1991b; Tanabe et al., 1992, 1994; MoradianOldak et al., 1994a, 1996; Fincham and MoradianOldak, 1996; DenBesten et al., 1998). While the presence of proteinases within the extracellular
enamel matrix has been well known from the early observations of Suga (1970), the isolation and characterization of these enzymes from the matrix has been elusive. Recent cloning and determination of the expression pattern of the metalloproteinase enamelysin (MMP-20) (Bartlett et al., 1996; Llano et al., 1997) and of the serine proteinase EMSP-1 (Simmer et al., 1998) have helped to clarify the confusing picture of the enamel proteinase occurrence and function. Northern analysis has shown that enamelysin mRNA displays a developmentally dened pattern of expression in the enamel organ when the message is expressed at relatively high levels during the presecretory and early transition stages of development (Bartlett et al., 1998). The message is, however, reduced during the maturation stage. On the other hand, the amount of EMSP-1 mRNA was demonstrated to sharply increase during the transition and early maturation stages (Scully et al., 1998). Incubation of a recombinant catalytic domain of enamelysin (rpMMP-20) with recombinant amelogenin has shown that this metalloproteinase not only processes the full-length amelogenin at its carboxy-terminal side at least six different cleavage sites, but it also cleaves the TRAP polypeptide (Ryu et al., 1999). In vitro incubation experiments using a fraction rich in EMSP-1 activity have demonstrated complete degradation of the recombinant amelogenin substrate to peptides as small as 147 Da by the serine proteinase (Moradian-Oldak et al., 1998b). Both enamelysin and EMSP-1 are thought to be active against other enamel proteins such as enamelins, ameloblastin, and the dentin sialo-phospho protein. Experimental evidence for such activities, however, is still lacking. Based on the above evidence, it can be assumed that enamelysin is the proteinase responsible for the gradual processing of amelogenin from the carboxy-terminal, resulting in critical structural changes of the enamel extracellular matrix and generating the TRAP molecule, which by itself may have an important function during enamel development (Ravindranath et al., 1999). The serine proteinase, on the other hand, has a digestive function, an activity which allows the hydroxyapatite crystals to grow rapidly during the maturation stage. For more details regarding the enamel proteinases, see Bartlett and Simmer (in press). 3.9. Other Matrix Components
The ability of [125I]-calcitonin and [125I]-insulin to enter enamel matrix adjacent to smooth-ended ameloblasts as early as 10 min after injection shows that their intercellular junctions are permeable to proteins with molecular weight as great as approx. 5700. (McKee et al., 1986)
3.9.1. Serum albumin. Albumin has been found to be a major component of the demineralized (guani-
283
dine/EDTA) fraction of both porcine (Limeback et al., 1989; Limeback and Simic, 1989) and bovine enamel matrix preparations (Strawich and Glimcher, 1989, 1990; Strawich et al., 1993). In a pattern typical of virtually all enamel proteins, albumin was detected in developing rat enamel throughout the secretory and transition stages, but was lost during early maturation (Robinson et al., 1996). Although albumin has affinity for enamel crystallites (Menanteau et al., 1987), it appears to be blocked from binding to crystallites in vivo by the presence of intact (21 kDa) amelogenins (Robinson et al., 1996). Albumin was also detected by immunohistochemistry during the secretory stage, at the enamel surface, but not in deeper layers (Okamura, 1983). However, albumin is not synthesized or secreted by ameloblasts (Yuan et al., 1996), and radiolabeled serum albumin injected into rabbits did not incorporate into the enamel layer, suggesting a physiological barrier between the extravascular uid and the enamel matrix (Kinoshita, 1979). Furthermore, ingress of albumin into enamel from dentin is restricted, particularly during the secretory stage (Shore et al., 1995a). How does albumin get into the enamel matrix? Kinoshita (1979) injected 125I-labeled serum albumin into rabbits and demonstrated, by autoradiography, that while the isotope was present within the dentine and predentine, none was present within the enamel matrix. Shapiro and Amdur (1965) demonstrated that red pigmentation observed in extracted developing bovine teeth (6 months in utero) was probably derived from hemoglobin adsorbed post mortem from the dental sac uid, an observation which suggests that albumin would be likely to exhibit the same behavior. Robinson et al. (1992) have suggested that albumin inclusion into the developing matrix may result in a localized defect of enamel mineralization, perhaps resulting in a white spot lesion. However, the participation of albumin in dental enamel formation, as distinct from being derived solely by post mortum infusion of the protein into the matrix, remains an open question. 3.9.2. Lipids. Wuthier and Irving (1964) suggested that the intense sudanophilia (Irving, 1963a,b) demonstrated in the developing enamel matrix was due to acidic lipid components. Shapiro et al. (1966) analyzed matrix phospholipids of both dentine and enamel matrix and reported a total phospholipid content of 0.06% by weight of the tissue. Subsequent studies (Fincham et al., 1972) demonstrated that the neutral lipids of bovine enamel matrix represented some 0.1% of the (lyophilized) sample weight. Thus, the total lipid content (neutral acidic) of developing enamel matrix is some 0.2% by weight, not dissimilar to levels reported for other mineralized
tissues. Is this lipid component (phospholipids, cholesterol, mono- and diglycerides, cholesteryl esters, and free fatty acids) functionally involved in the mineralization process or in enamel matrix structure? Perhaps the only pointer of signicance is the observation (Shapiro et al., 1966) that only a fraction (40%) of the phospholipid component is extractable without prior demineralization. Again, Goldberg et al., (1984, 1998) have explored the possibility that lipid-rich membrane fragments of the Tomes processes may become entombed within the matrix, contributing to the overall lipid content. Beyond the issue of a functional role for lipids in the enamel matrix, it is important to recognize that some of the amelogenin proteins themselves, as a result of their hydrophobic nature, exhibit lipophilic properties, such as being soluble in typical lipid solvents (e.g., chloroformmethanol, 2:1, v/v) (Burgess and MacLaren, 1965; Fincham et al., 1972).
4. ENAMEL MATRIX STRUCTURE AND FUNCTION Its main functions appear to be rstly to nurture the rapid growth of the large apatite crystals, by some mechanisms where growth in specic directions is either directed, or more probably, assisted by the matrix, and secondly to be readily removed during enamel maturation to permit the ultimate formation of a tissue which is almost entirely inorganic. (John E. Eastoe, 1965)
As Eastoe has noted the matrix nurtures the growth of the enamel crystallites, allowing them to develop as an oriented array of extremely long structures, perhaps extending through the full thickness of the secreted matrix. It is commonly accepted that these early crystallites arise at the developing DEJ and then extend through the matrix by progressive accretion of ions, transported by the ameloblast, onto the (001) face of the crystallites. How these enamel crystallites may be nucleated is presently controversial. One view is that nucleation occurs on preexisting dentine mineral (Arsenault and Robinson, 1989) or alternatively that the crystals are nucleated de novo by acidic matrix proteins within the enamel matrix and are not continuous with the dentine mineral (Diekwisch et al., 1994). Enamelins, tuftelin, amelogenins, ameloblastin, and, most recently, dentine sialophosphoprotein (MacDougall et al., 1998a) have all been cited as candidate proteins to act as mineral nucleators (Deutsch et al., 1998, 1991; Zeichner-David et al., 1997; Robinson et al., 1998b; Termine et al., 1980; Fong et al., 1996a; Krebsbach et al., 1996). Convincing evidence of a nucleating role for any of these proteins is, however, currently lacking. It is of interest to note that, depending on the angle of the section, high resolution electron microscopic images of early-stage enamel matrices do appear to show crystallites
284
impinging very closely to the ameloblast Tomes process cell membranes (Diekwisch, personal communication), prompting the question as to whether nucleation might be ameloblast-membrane initiated. Also associated with the ameloblast membrane are structures commonly identied as stippled material consisting of elds of particles some 1020 nm in diameter (Watson, 1960). These structures are also seen within ameloblast secretory vesicles, suggesting that they are an early stage of matrix protein secretion. It has been suggested (Fincham et al., 1994b) that the stippled material may represent the initial stage of amelogenin nanosphere assemblies as discussed below. A recent review of the enamel matrix properties and the initiation of the mineral phase has been provided by Robinson et al. (1998b). 4.1. Gross Matrix Composition
Both amelogenins and enamelins were found at all stages of enamel development, with higher levels of enamelins being present both at later stages where tissue maturation becomes advanced and at early stages where active matrix secretion is predominant. (Termine et al. 1980)
As has already been noted, Termine et al. (1980) reported the use of a dissociative extraction procedure for isolating enamel matrix proteins from preparations of the soft cheese-like scrapings obtained from developing bovine teeth. This procedure resulted in the fractionation of the total matrix protein into two classes: amelogenins and nonamelogenins (or enamelins). On the basis of this procedure Termine et al. (1980) showed that in bovine enamel the proportion of amelogenin in the matrix varied with fetal age, rising from some 70% in the youngest (24 months) to 8590% at 56 months and then falling to some 50% as maturation of the enamel takes place. By implication, the difference of these values was made up by the enamelin (or nonamelogenin) fraction. Thus, at the maximum of the secretory stage of matrix deposition some 10% was accounted for as nonamelogenins. This fraction would now appear to comprise the ameloblastins, tuftelins, and enamelins. How realistic is this picture? Other data (Fincham, unpublished observations) obtained by employing the dissociative extraction of 8 g of developing bovine enamel matrix scrapings, using ve sequential steps of guanidine extraction before using the guanidineEDTA step to demineralize the matrix, suggest that (i) secretory stage enamel matrix may contain only some 3% or less of nonamelogenins, and (ii), unsurprisingly, the extraction procedure is not quantitative, the guanidineEDTA extract (23% of total protein) still apparently including some amelogenin (see Fig. 2). This issue appears important in view of the putative role of the nonamelogenin proteins (e.g., amelo-
blastin) as structural components of the matrix (Uchida et al., 1995). Based on the data of Fig. 2, it appears that the amelogenin content of the secretory stage matrix may in fact be greater than previously suggested (Termine et al., 1980). However, this is difficult to substantiate as quantitative data for the proportions of nonamelogenin proteins (ameloblastins, enamelins), based on direct measurement, is presently lacking, the presence of these molecules within the matrix being generally inferred either from mRNA levels or by immunohistological observations. The difficulty of quantitatively identifying the proportions of matrix proteins is further complicated by the temporal changes in both protein secretion and protein resorption, which occur as the matrix mineralizes. These changes in gross composition of the enamel matrix are complex functions, comprising sequential changes in total protein, mineral, and aqueous compartments. Such changes have been determined by Robinson et al. (1988) on a volumetric basis, and the data have recently been reviewed by Smith (1998). 4.2. Spherical Substructures
Globular structures or spheres of 300500 in diameter seem to have occupied virtually the entire volume of the tissue. (Robinson et al., 1981)
A number of earlier studies of enamel matrix ultrastructure suggested that a periodic substructure existed either between or around the earliest enamel crystallites (Ronnholm, 1962; Travis and Glimcher, 1964). In contrast, Eastoe (1963) proposed that the matrix was essentially an unstructured gel with thixotropic properties (i.e., a matrix which modied its structure in response to local pressures exerted by the developing mineral crystallites), while Warshawsky (1985) took the view that the amelogenin protein acted to . . . separate the crystallites to keep them from fusing and provide an extracellular environment for rapid diffusion of minerals to the calcifying sites on growing crystallites. In 1975, Smales, working with developing rat enamel, described the presence of helical structures showing a pitch varying from 30 to 90 nm. Robinson et al. (1981), in a scanning electron microscope freeze fracture study of developing rat enamel, demonstrated colinear spheres of organic material. . . and suggested that they might be mineralprotein aggregates. Bai and Warshawsky (1985) noted globular particles in freezefracture replicas of fresh and xed enamel samples which were removed by 4.0 M guanidine extraction. They concluded that amelogenins were the nonstructural, heterodispersed par-
285
ticulate material in the intercrystallite space, although Warshawsky (1985) considered these globular particles to be artifactual. 4.2.1. Nanospheres. The recent availability of recombinant amelogenin together with the development of HPLC chromatographic purication procedures for in vivo amelogenin proteins has greatly facilitated studies of this principal matrix component (Simmer et al., 1994b). In 1994, we reported that high resolution transmission electron micrographs showed evidence for beaded rows of structures (some 20 nm in diameter) aligned with and between the initial-stage developing enamel crystallites (Moradian-Oldak, 1994b, 1995; Fincham et al., 1994b, 1995). It was also demonstrated that recombinant mouse amelogenin protein (rM179) spontaneously formed such structures when a solution was sprayed onto electron-microscope grids or alternatively adsorbed onto mica and imaged by AFM. It was proposed that amelogenin molecules underwent a spontaneous self-assembly process generating nanosphere structures which, it was argued, functioned to organize the microstructure of the matrix, spacing the initial crystallites at some 20-nm distances and inhibiting early crystalcrystal lateral fusions (Fincham et al., 1995; Fincham and Simmer, 1997). Further, current studies (Moradian-Oldak et al., 1998a) have shown that amelogenin nanospheres can act to adhere synthetic apatite crystals through a polymer-bridging mechanism (Nancollas and Budz, 1990) (see Fig. 4). Studies in this laboratory employing dynamic light scattering and imaging techniques with both recombinant and in vivo amelogenins have now rmly established that amelogenins undergo a selfassembly process to generate spherical structures variously ranging in size from 5 to 100 nm, depending on the local conditions (Moradian-Oldak et al., 1994b, 1998c; Fincham et al., 1998). Landis et al. (1998) have used 3D-TEM tomographic reconstruction techniques to image xed preparations of developing mouse enamel matrix and observed rows of nanospheres apparently spiraling around the initial mineral crystallites. These observations have been independently conrmed using cryo-transmission electron microscopy, a direct imaging technique requiring no prior tissue xation (Leytin et al., 1998). 4.3. The Matrix Assembly Process
The precipitate, when separated from the supernatant by centrifugation and stored at 4C, formed a clear gel, which, when heated to 18C, reverted to a white opaque mass. (Nikiforuk and Simmons, 1965)
Beyond studies of amelogenin secondary and tertiary structures (noted above), it is pertinent here to note the early work of Nikiforuk and Simmons
(1965), who rst demonstrated that amelogenin preparations undergo a reversible temperaturesensitive hydrophobic aggregation (coacervation effect) passing from a translucent gel at 4C to an opaque material at room temperature. Recent studies (Wen et al., 1999) in which these early temperature-sensitive matrix properties were further explored have shown that a synthetic matrix prepared from porcine amelogenins and rendered translucent at 4C may be xed in this condition using conventional glutaraldehyde (Karnovsky) xation. Likewise, a preparation held at room temperature (opaque) is xed in this opaque condition. Scanning electron micrographs of these preparations have now shown that the differing optical properties (opaque to clear) are apparently due to the room-temperature preparations containing numerous globular voids, compared to the 4C preparations (Wen et al., 1999). Figure 5 shows examples of the structure of a porcine amelogenin matrix prepared at room temperature, and Fig. 6 uses AFM imaging to illustrate the difference in the surface features of recombinant (rM179) amelogenin matrix gels formed both at room temperature and at 4C. While these experiments provide some explanation of the phenomenon originally described by Nikiforuk and Simmons (1965), they have also shown that these synthetic matrices are composed of assemblies of spherical structures (nanospheres) ranging in size from 20 to 200 nm in diameter (Fig. 5). The change in the distribution of these structures with temperature is interpreted to result from a changed distribution of hydrophobic interaction forces between these nanospheres, as temperature is varied. Also addressing the problem of amelogenin amelogenin interactions, Paine and Snead (1997), employing the yeast two-hybrid system (Fields and Song, 1989), have found that amelogenin proteins self-associate and that this process appears to depend on the amino-terminal 42 residues interacting with a 17-residue domain in the carboxyl region of the protein. This twin locus interaction provides the potential for amelogeninamelogenin interactions to generate the supramolecular structures which we have described. However, the situation is almost certainly more complex, since AFM imaging studies of a synthetic TRAP (the N-terminal 45 residues of murine amelogenin; M180) show that this molecule alone has the ability to form supramolecular structures (unpublished observation). Further, it has been shown that the tuftelin protein also has a self-assembly potential, which is, however, apparently not manifest in ameloblastin (Paine et al., 1998a,b).
286
FIG. 5. Structure of a synthetic amelogenin gel matrix. Scanning electron micrographs of features of a synthetic amelogenin gel matrix xed at room temperature (Wen et al., 1999). Porcine amelogenins (principally composed of 7% P173 and 40% P148 amelogenins) were precipitated and sedimented by centrifugation. The resulting gel matrix was xed with glutaraldehyde and imaged by scanning EM (original magnication, 5000). (A) Surface feature of the matrix shows multiple globular pits distributed over the surface. Comparable matrices xed at 4C (data not shown) show an essentially pit-free surface. (B and C) Details of individual pits (rotated 90 for convenience) show them to be nanosphere-lined structures (original magnication, 30 000). See text for discussion.
4.4. Amelogenin Mineral Interactions

Changes in the secondary structure upon adsorption, as well as transfer of water from the molecule and the adsorbent surface to the bulk solution, may provide the increase in entropy required for adsorption to take place. . . . Unfortunately, there is insufficient information about the molecular conformation of amelogenins in solution and in the adsorbed state, . . . (Aoba et al., 1989)
Inhibition of crystal growth by macromolecules in solution is a general phenomenon which may not be directly related to the function of the investigated
macromolecule but determines whether the macromolecules interact with the crystals. The adsorption behavior of amelogenins in solution has been related to their inhibitory activity in crystal growth experiments (Doi et al., 1984; Aoba et al., 1987a,b). The apatite binding properties of amelogenins have previously been investigated, mainly in terms of adsorption parameters (Aoba et al., 1987a; Aoba and Moreno, 1989). Kinetic studies also reported the inhibition of apatite crystal growth by amelogenins in solution
287
FIG. 6. Tapping mode, atomic force micrographs of the surface features of a synthetic enamel matrix formed by puried recombinant amelogenin (rM179). The puried amelogenin was dissolved in dilute formic acid, and the pH was adjusted to 6.8 with NaOH additions. The precipitated protein was pelleted by centrifugation, creating an amelogenin gel matrix which was found to exhibit a temperaturesensitive reversible coacervation (clear & opaque) as originally described (Nikiforuk and Simmons, 1965). Samples of this matrix were xed with glutaraldehyde as for conventional electron microscopy. (A) Matrix sample xed at room temperature (23C). A eld of essentially uniform-sized nanospheres (4050 nm diameter) contains major irregularities and pits. (Compare Fig. 5A.) (B) Matrix sample xed at 4C. Nanospheres of sizes similar to those seen in A form a fairly uniform surface without major pit structures. We speculate that this effect results from the increased hydrophobic interactions, generated by the breakdown of hydration shells in the room-temperature sample, causing distortion of even-force distributions in the surface, compared to the 4C sample.
(Doi et al., 1984). The aggregation properties of amelogenins in solution have limited the conditions under which such experiments may be performed. Aoba et al. (1989) and Simmer et al. (1994b) showed that cleavage of the hydrophilic carboxy-terminal of amelogenin decreases its apatite binding affinity. In another recent study, puried recombinant amelogenin rM179 did not have any signicant effect on
the nucleation of hydroxyapatite crystals (Hunter et al., in press). Our current studies have shown that cleavage of the amelogenin carboxy-terminal motif lowers the inhibitory potential of the molecule to apatite growth kinetics, compared to the intact rM179 amelogenin (Moradian-Oldak et al., 1998a). In an attempt to investigate interactions of amelogenin nanospheres
288
with growing hydroxyapatite crystals, MoradianOldak et al. (1998a) have investigated the inuence of the two recombinant amelogenins (rM179 and rM166) and also the synthetic polypeptide, LRAP, compared to two other macromolecules [poly(Lproline) and phosvitin], on the kinetics of seeded growth of apatite crystals. It was shown that the rM179 amelogenin had some inhibitory effect, but was not as destructive as that of the other macromolecules examined. The degree of inhibition of the macromolecules occurred in the following order: rM179 phosvitin LRAP poly(L-proline) rM166. In addition, using transmission electron microscopy, the effect of these macromolecules on the aggregation of growing apatite crystals was evaluated based on the previously reported polymer bridging studies (Nancollas and Budz, 1990). Analysis of the particle size distribution of apatite crystal aggregates has indicated that the full-length amelogenin protein causes aggregation of growing apatite crystals more effectively than the other macromolecules tested. Two major parameters were recognized as being critical for the apparent adherence effect: (i) the binding affinity of the amelogenin nanospheres, depending on the acidic carboxy-terminal telopeptide, and (ii) the self-assembly of amelogenin molecules through hydrophobic interactions. It was therefore proposed that during the formation of hydroxyapatite crystal clusters, the highly packed and tightly associated amelogenin nanospheres interact with apatite crystal planes through their hydrophilic surfaces bridging the crystals together (Fig. 7). The biological implication of this effect has been attributed to the ability of (full-length) amelogenin nanospheres to interact with the extremely thin enamel crystallites at the early stage of enamel formation and to protect them from premature crystalcrystal fusions. During the secretory stage, enamel apatite crystals with thicknesses of approximately 15 have large surface areas, resulting in their tendency to interact with each other through electrostatic interactions. Such interactions may favor premature crystalcrystal fusions of the crystallites (Warshawsky, 1985; Moradian-Oldak et al., 1998a). 4.5 Proteolysis and Maturation
No one has demonstrated convincingly, however, the actual physical movement of formerly extracellular material from outside to inside the cell. (C. E. Smith, 1979)
FIG. 7. Aggregation of synthetic apatite crystals by recombinant amelogenin in solution. Transmission electron micrographs of calcium hydroxyapatite crystals collected at the end of crystal growth experiment and resuspended in TrisHCl, pH 7.4, (A) in the absence of any macromolecule and (b) in the presence of 20 g/ml rM179 in the crystallization solution (Moradian-Oldak et al., 1998a). Scale bar applies to both images.
Immunohistochemistry and Western blot analyses of proteins extracted from supercial and deeper enamel layers have demonstrated that the carboxylterminus of amelogenin is restricted in its localization to within 40 m of the surface of developing enamel, while its proteolytic cleavage products are
observed throughout the enamel layer down to the DEJ (Uchida et al., 1991a; Tanabe et al., 1992; Fukae et al., 1993; Fukae and Tanabe, 1998; Smith et al., 1989b, 1996b). Different sublocalizations for amelogenin and amelogenin polypeptides suggest separate and distinct functions for the intact protein and its cleavage products. The mass spectrometric analysis of enamel proteins isolated from secretory stage enamel has permitted the identication of many of the amelogenin proteolytic cleavage sites. The 173residue amelogenin in the pig (also known as the 25-kDa amelogenin) appears to experience a succession of rapid cleavages near the carboxyl-terminus. Pig amelogenin cleavage products extending from the amino-terminus to residues 171, 162, 160, 157,
289
153, 148, 147, 136, 107, 105, 45, and 43 have been identied by mass spectrometry (Fincham and Moradian-Oldak, 1993). Because of their distinctive amino acid compositions, the smaller polypeptides (45 and 43 residues) are known as the tyrosine-rich amelogenin polypeptides (Fincham et al., 1981). A number of postTRAP cleavage products have also been identied by mass spectrometric analysis, these extended from Leu 46 to 148, 147, 136, 107, and 105 (Ryu et al., 1999). The current interpretation of the data for the pig is that successive cleavages from the carboxylterminus reduce the parent 173-residue secreted protein to the 148-residue polypeptide (known as the 20-kDa amelogenin), the most abundant amelogenin in the enamel matrix (Fukae and Shimizu, 1983). This 20-kDa porcine amelogenin is the most stable of all the amelogenin cleavage products. It accumulates in the matrix and is essentially the starting point for subsequent cleavages. Comparison with the known processing products identied in other species (Fincham et al., 1994a) suggests that this initial proteolytic transformation, from the full-length hydrophilic molecule to a less transient, more hydrophobic protein molecule lacking some 25 or so residues from the carboxy-terminal, is a general feature of early amelogenesis. The functional signicance of the processing of amelogenins during the secretory stage of amelogenesis is unknown, but it is evident that this removal of the amelogenin carboxyl-terminus reduces its binding affinity for hydroxyapatite (Aoba et al., 1987a, 1989; Aoba and Moreno, 1989; Bronckers et al., 1995; Brookes et al., 1998), and also its solubility (Tan et al., 1998). Further, as suggested in the model of Fig. 8, removal of this hydrophilic motif will function to further promote hydrophobic amelogeninamelogenin interactions.
5. MATRIX-MEDIATED ENAMEL MINERALIZATION While molecular approaches are able to dene the mutations responsible for amelogenesis imperfecta disorders, further protein chemistry and ultrastructural analyses are required to elucidate the mineralprotein interactions and relations in both normal and pathological enamel development. (Wright et al., 1997)
The data presented above serve to emphasize a number of key issues which have a bearing on our understanding of the mechanism of dental enamel biomineralization: (i) Enamel biomineralization is an extracellular event. (ii) Evidence for the nucleation of mineral crystallites de novo within the matrix is inconclusive. (iii) Amelogenin proteins are established to be the major ameloblast-specic gene products.
(iv) The secretory stage matrix is composed predominantly of amelogenin proteins. (v) The proportions of nonamelogenin proteins (tuftelin, enamelin, and ameloblastin) within the secretory matrix are currently unknown. (vi) Ameloblastin immunoreactivity within the matrix is localized to putative prism boundaries. (vii) Tuftelin has never been isolated from the enamel matrix and is expressed in other soft tissues (e.g., spleen, liver; MacDougall et al., 1998a). (viii) Amelogenins (both recombinant and native) spontaneously assemble to generate nanosphere supramolecular structures. (ix) A range of imaging techniques have identied amelogenin nanospheres as the principal structural component of the secretory-stage enamel matrix. (x) Native and recombinant full-length amelogenins interact with apatite mineral surfaces. (xi) Amelogenins have lectin-like properties in the TRAP sequence motif: -PYPSYGYEPMGGW- with a specicity for N-acetylglucosamine residues (Ravindranath et al., 1999). (xii) Amelogeninamelogenin and amelogenin mineral interactions promote crystalcrystal binding (bridging). (xiii) Amelogenins are initially proteolytically processed from the anionic carboxy-terminus, predictably altering the hydrophilicity of the molecules and their assemblies. (xiv) Characterization of the genetic lesions leading to cases of X-linked amelogenesis imperfecta establish the requirement for conserved amelogenin structures in normal enamel biomineralization. Fincham and Simmer (1997) have advanced a model for enamel matrix biomineralization in which amelogenin nanospheres are formed with the hydrophilic (pI 4.2) carboxy-terminus externalized. These assemblies act initially to space enamel crystallites through ionic interactions with the mineral. (Alternative models for the matrix-mediated biomineralization of enamel include those of Aoba and Moreno, 1989; Fincham et al., 1991b; and Limeback, 1991.) We may then speculate that a progressive proteolytic processing of the exposed carboxy-terminals will provide a modulated change in the hydrophilicity of the nanosphere surfaces from an anionic, mineralbinding structure to an essentially hydrophobic selfassociating matrix. An updated version of this model is presented in Fig. 8.
6. PROBLEMS AND OUTSTANDING QUESTIONS I believe that octacalcium phosphate (OCP) acts as a precursor in the formation of apatite and therefore greatly affects the crystallite morphology, chemical composition, and mechanical properties of the enamel mineral. (W. E. Brown, 1979)
290
FIG. 8. Schematic diagram for enamel biomineralization. (1) Amelogenins are synthesized by the ameloblasts and secreted extracellularly. (2) Amelogenin monomers assemble to generate nanosphere structures of some 20 nm diameter with the hydrophilic (anionic) carboxy-terminals externalized. (3) Anionic nanospheres initially interact electrostatically with the crystallite faces parallel to the c axis, preventing crystalcrystal fusions and acting as 20-nm spacers. Enamelysin (proteinase-1) then processes the exposed amelogenin carboxy-terminals progressively reducing their anionic character. Hydrophobic nanospheres further assemble, stabilizing the matrix containing the initial enamel crystallites, which continue to grow by ion accretion on their exposed 001 faces. (4) EMSP-1 (proteinase-2) action degrades the hydrophobic nanosphere amelogenins generating smaller fragments and other unidentied products, which are eventually resorbed by ameloblasts. (5) As amelogenin nanosphere protection is removed, crystallites thicken and eventually may fuse to generate the mature enamel.
There are a number of key questions which relate to the mechanism of dental enamel biomineralization and the functional role of the enamel matrix. A selection of these issues might include the following: (i) What are the relationships between the enamel matrix, the DEJ, and the dentin matrix proteins? What proteinprotein interactions exist? (ii) What are the temporal controls of enamel
matrix secretion and resorption? What signals the ameloblast to cease matrix synthesis and secretion? What is the pathway of matrix protein resorption? (iii) What are the phylogenetic relationships between the enamel matrix proteins and those of enameloid structures? (iv) How does amelogenin nanosphere assembly take place? What is the nanosphere substructure?
291
(v) How (where) are enamel crystals nucleated? Does this event occur de novo within the matrix, or is it mediated by proteins associated with the DEJ (enamelins, tuftelin)? Alternatively, does it arise from preexisting dentine mineral, or could it occur the other way around, at the ameloblast cell membrane? (vi) What is the chemical nature of the initial enamel crystallite ribbons (OCP or HAP)? What factors control the crystallite mineral phase and growth? (vii) Where are the enamel proteinases located within the matrix? Are they secreted in an inactive form? If so, how are they activated? (viii) What is the signicance of the multiple alternatively spliced isoforms of amelogenin? (ix) Of the several nonamelogenin proteins (ameloblastin, enamelin, and tuftelin), which of these are enamel specic and what may be their functional role?
We thank our colleagues Dr. Hai-Bo Wen and Dr. Wen Luo for the use of the SEM and AFM images reproduced in this paper. Wendy Leung and Ismael Jimenez provided expert technical support for some of the studies noted in this paper and Dr. Ravindranath contributed insights into the lectin-like properties of the amelogenins. This work has been supported by NIH NIDCR Research Grants DE-02848 (A.G.F. and J.M.O.), DE-12350 (J.M.O.), and DE-10721 (J.P.S.). REFERENCES Addadi, L., and Weiner, S. (1985) Interactions between acidic proteins and crystals: Stereochemical requirements in biomineralization, Proc. Natl. Acad. Sci. USA 82, 41104114. Addadi, L., and Weiner, S. (1989) Stereochemical and structural relations between macromolecules and crystals in biomineralization, in Mann, S., Webb, J., and Williams, R. J. P. (Eds.), Biomineralization, Chemical and Biochemical Perspectives, pp. 133152, Weinheim, New York. Addadi, L., Berman, A., Moradian-Oldak, J. M., and Weiner, S. (1989) Structural and stereochemical relations between acidic macromolecules of organic matrices and crystals, Connect. Tissue Res. 21, 127134. Addadi, L., Moradian-Oldak, J., Furedi-Milhofer, H., Weiner, S., and Veis, A. (1992) Stereochemical aspects of crystal regulation in calcium phosphate-associated mineralized tissues, in Slavkin, H. C., and Price, P. (Eds.), Chemistry and Biology of Mineralized Tissues, pp. 153162, Excerpta Medica, New York. Akita, H., Fukae, M., Shimoda, S., and Aoba, T. (1992) Localization of glycosylated matrix proteins in secretory porcine enamel and their possible functional roles in enamel mineralization, Arch. Oral Biol. 37, 953962. Aldred, M. J., Crawford, P. J. M., Roberts, E., and Thomas, N. S. T. (1992) Identication of a nonsense mutation in the amelogenin gene (AMELX) in a family with X-linked amelogenesis imperfecta (AIH1), Hum. Genet. 90, 413416. Amizuka, N., and Ozawa, H. (1989) Ultrastructural observation of enamel tufts in human permanent teeth, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 410414, Florence, Yokohama. Amizuka, N., Uchida, T., Fukae, M., Yamada, M., and Ozawa, H. (1992) Ultrastructural and immunocytochemical studies of
enamel tufts in human permanent teeth, Arch. Histol. Cytol. 55, 179190. Aoba, T., and Moreno, E. C. (1989) Mechanism of amelogenetic mineralisation in minipig secretory enamel, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 163167, Florence, Yokohama. Aoba, T., Fukae, M., Tanabe, T., Shimizu, M., and Moreno, E. C. (1987a) Selective adsorption of porcine-amelogenins onto hydroxyapatite and their inhibitory activity on hydroxyapatite growth in supersaturated solutions, Calcif. Tissue Int. 41, 281289. Aoba, T., Tanabe, T., and Moreno, E. C. (1987b) Function of amelogenins in porcine enamel mineralization during the secretory stage of amelogenesis, Adv. Dent. Res. 1, 252260. Aoba, T., Moreno, E. C., Kresak, M., and Tanabe, T. (1989) Possible roles of partial sequences at N- and C-termini of amelogenin in proteinenamel mineral interaction, J. Dent. Res. 68, 13311336. Aoba, T., Kawano, K., and Moreno, E. C. (1990) Molecular conformation of porcine amelogenins and its signicance in proteinmineral interaction: 1H-NMR photo-CIDNP study, J. Biol. Buccale 18, 189194. Arsenault, A. L., and Robinson, B. W. (1989) The dentinoenamel junction: A structural and microanalytical study of early mineralization, Calcif. Tissue Int. 45, 111121. Astbury, W. T. (1961) The structure of the bres of the collagen group and related matters twenty-one years after. Procter Memorial Lecture, September 23rd., 1960, J. Soc. Leather Trades Chem. 45, 186214. Bai, P., and Warshawsky, H. (1985) Morphological studies on the distribution of enamel matrix proteins using routine electron microscopy and freeze-fracture replicas in the rat incisor, Anat. Rec. 212, 116. Bartlett, J. D., Simmer, J. P., Xue, J., Margolis, H. C., and Moreno, E. C. (1996) Molecular cloning and mRNA tissue distribution of a novel matrix metalloproteinase isolated from porcine enamel organ, Gene 183, 123128. Bartlett, J. D., Ryu, O. H., Xue, J., Simmer, J. P., and Margolis, H. C. (1998) Enamelysin mRNA displays a developmentally dened pattern of expression and encodes a protein which degrades amelogenin, Connect. Tissue Res. 39, 101109. Bartlett, J. D., and Simmer, J. P. (1999) Proteinases in developing dental enamel, Crit. Rev. Oral Biol. Med., in press. Bashir, M. M., Abrams, W. R., and Rosenbloom, J. (1997) Molecular cloning and characterization of the bovine tuftelin gene, Arch. Oral Biol. 42, 489496. Bashir, M. M., Abrams, W. R., and Rosenbloom, J. (1998) Molecular cloning and characterization of the bovine and human tuftelin genes, Connect. Tissue Res. 39, 1324. Begue-Kirn, C., Krebsbach, P. H., Bartlett, J. D., and Butler, W. T. ` (1998) Dentin sialoprotein, dentin phosphoprotein, enamelysin and ameloblastin: Tooth-specic molecules that are distinctively expressed during murine dental differentiations, Eur. J. Oral Sci. 106, 963970. Block, R. J., and Bolling, D. (1952) The Amino Acid Composition of Proteins and Foods, Thomas, Springeld, IL. Bonar, L. C., Glimcher, M. J., and Mechanic, G. L. (1965) Molecular structure of the neutral soluble proteins of embryonic bovine enamel in the solid state, J. Ultrastruct. Res. 13, 308317. Bonass, W. A., Kirkham, J., Brookes, S. J., Shore, R. C., and Robinson, C. (1994a) Isolation and characterisation of an alternatively-spliced rat amelogenin cDNA: LRAP a highly conserved, functional alternatively-spliced amelogenin? Biochim. Biophys. Acta 1219, 690692.
292
FINCHAM, MORADIAN-OLDAK, AND SIMMER Deutsch, D., Palmon, A., Fisher, L. W., Kolodny, N., Termine, J. D., and Young, M. F. (1991) Sequencing of bovine enamelin (tuftelin), a novel acidic enamel protein, J. Biol. Chem. 266, 16021 16028. Deutsch, D., Palmon, A., Dafni, L., Shenkman, A., Sherman, J., Fisher, L., Termine, J. D., and Young, M. (1991) Enamelin and enameloid, in Suga, S. (Ed.), Mechanisms and Phylogeny of Mineralization in Biological Systems, pp. 7377, Springer, Tokyo. Deutsch, D., Palmon, A., Young, M. F., Selig, S., Kearns, W. G., and Fisher, L. W. (1994) Mapping of the human tuftelin (TUFT1) gene to chromosome 1 by uorescence in situ hybridization, Mamm. Genome 5, 461462. Deutsch, D., Palmon, A., Dafni, L., Catalano-Sherman, J., Young, M. F., and Fisher, L. W. (1995) The enamelin (tuftelin) gene, Int. J. Dev. Biol. 39, 135143. Deutsch, D., Palmon, A., Dafni, L., Mao, Z., Leytin, V., Young, M., and Fisher, L. W. (1998) Tuftelinaspects of protein and gene structure, Eur. J. Oral Sci. 106, 315323. Diekwisch, T., David, S., Bringas, P., Santos, V., and Slavkin, H. C. (1993) Antisense inhibition of AMEL translation demonstrates supramolecular controls for enamel HAP crystal growth during embryonic mouse molar development, Development 117, 471448. Diekwisch, T. G. H., Berman, B. J., Genter, S., and Slavkin, H. C. (1994) Initial enamel crystals are not spatially associated with mineralized dentine, Cell Tissue Res. 279, 149167. Doi, Y., Eanes, E. D., Shimokawa, H., and Termine, J. D. (1984) Inhibition of seeded growth of enamel apatite crystals by amelogenin and enamelin proteins in vitro, J. Dent. Res. 63, 98105. Eastoe, J. E. (1960) Organic matrix of tooth enamel, Nature 187, 411412. Eastoe, J. E. (1963) The amino acid composition of proteins from the oral tissues. II. The matrix proteins in dentine and enamel from developing deciduous human teeth, Arch. Oral Biol. 8, 633652. Eastoe, J. E. (1965) The chemical composition of bone and tooth, Adv. Fluorine Res. Dent. Caries Prevent. 3, 516. Eastoe, J. E. (1971) Session introduction, in Fearnhead, R. W., and Stack, M. V. (Eds.), Tooth Enamel II, p. 65, Wright Bros., Bristol. Eastoe, J. E. (1979) Enamel protein chemistryPast, present and future, J. Dent. Res. 58B, 753764. Eggert, G. M., Allen, G. A., and Burgess, R. C. (1973) Amelogenins: Purication and partial characterization of proteins from developing bovine enamel, Biochem. J. 131, 471484. Fearnhead, R. W. (1979) Matrixmineral relationships in enamel tissues, J. Dent. Res. 58B, 909916. Fields, S., and Song, O. (1989) A novel genetic system to detect proteinprotein interactions, Nature 340, 245246. Fincham, A. G., Graham, G. N., and Pautard, F. G. E. (1965) The matrix of enamel and related mineralized keratins, in Fearnhead, R. W., and Stack, M. V. (Eds.), Tooth Enamel, pp. 117123, Wright Bros., Bristol. Fincham, A. G., Burkland, G., and Shapiro, I. M. (1972) Lipophilia of enamel matrix; a chemical investigation of the neutral lipids and lipophilic proteins of bovine foetal enamel matrix, Calcif. Tissue Res. 9, 247259. Fincham, A. G. (1979) The amelogenin problem; a comparison of puried enamel matrix proteins, Calcif. Tissue Int. 26, 6573. Fincham, A. G., Belcourt, A. B., Termine, J. D., Butler, W. T., and Cothran, W. C. (1981) Dental enamel matrix: Sequences of two amelogenin polypeptides, Biosci. Rep. 1, 771778.
Bonass, W. A., Robinson, P. A., Kirkham, J., Shore, R. C., and Robinson, C. (1994b) Molecular cloning and DNA sequence of rat amelogenin and a comparative analysis of mammalian amelogenin protein sequence divergence, Biochem. Biophys. Res. Commun. 198, 755763. Boyde, A. (1997) Microstructure of enamel, Ciba Found. Symp. 205, 1827. Bronckers, A. L., DSouza, R. N., Butler, W. T., Lyaruu, D. M., van Dijk, S., Gay, S., and Woltgens, J. H. (1993) Dentin sialoprotein: Biosynthesis and developmental appearance in rat tooth germs in comparison with amelogenins, osteocalcin and collagen type-I, Cell Tissue Res. 272, 237247. Bronckers, A. L., Bervoets, T. J., Lyaruu, D. M., and Woltgens, J. H. (1995) Degradation of hamster amelogenins during secretory stage enamel formation in organ culture, Matrix Biol. 14, 533541. Brookes, S. J., Robinson, C., Kirkham, J., and Bonass, W. A. (1995) Biochemistry and molecular biology of amelogenin proteins of developing dental enamel, Arch. Oral Biol. 40, 114. Brookes, S. J., Kirkham, J., Shore, R. C., Bonass, W. A., and Robinson, C. (1998) Enzyme compartmentalization during biphasic enamel matrix processing, Connect. Tissue Res. 39, 8999. Brown, W. E. (1979) General comment, Proceedings of Third International Symposium on Tooth Enamel, J. Dent. Res. 58B, 857. Burgess, R. C., and MacLaren, C. (1965) Proteins in developing bovine enamel, in Fearnhead, R. W., and Stack, M. V. (Eds.), Tooth Enamel, Its Composition, Properties and Fundamental Structure, pp. 7482, Wright Bros., Bristol. Catalano-Sherman, J., Palmon, A., Burstein, Y., and Deutsch, D. (1994) Amino acid sequence of a major human amelogenin protein employing Edman degradation and cDNA sequencing, J. Dent. Res. 72, 15661572. Cerny, R., and Hammarstrom, L. (1998) Cloning, cDNA sequence, and alternative splicing of guinea-pig amelogenin mRNAs, in Robinson, C., and Goldberg, M. (Ed.), Proceedings Sixth International Conference on the Chemistry and Biology of Mineralized Tissues, Abstract No. 38, Vittel, France. Cerny, R., and Hammarstrom, L. (1999) NCBI Protein Database, Accession No. AJ012200. Cerny, R., Slaby, I., Hammarstrom, L., and Wurtz, T. (1996) A novel gene expressed in rat ameloblasts codes for proteins with cell binding domains, J. Bone Miner. Res. 11, 883891. Clement, J. G. (1984) Changes to structure and chemistry of Chondrichthyan enameloid during development, in Fearnhead, R. W., and Suga, S. (Eds.), Tooth Enamel IV, pp. 422426, Academic Press, New York. Collier, P. M., Sauk, J. J., Rosenbloom, J., Yuan, Z. A., and Gibson, C. W. (1997) An amelogenin gene defect associated with human X-linked amelogenesis imperfecta. Arch. Oral Biol. 42, 235242. Deakins, M. (1942) Changes in ash, water and organic content of pig enamel during calcication, J. Dent. Res. 21, 429. DenBesten, P. K., Punzi, J. S., and Li, W. (1998) Purication and sequencing of a 21 kDa and 25 kDa bovine enamel metalloproteinase, Eur. J. Oral Sci. 106, 345349. Deutsch, D., Palmon, A., Catalano-Sherman, J., and Laskov, R. (1987) Production of monoclonal antibodies against enamelin and against amelogenin proteins of developing enamel matrix, Adv. Dent. Res. 1, 282288. Deutsch, D., Palmon, A., Fisher, L., Termine, J. D., and Young, M. (1989) Cloning of bovine enamelin, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 308312, Florence, Yokohama.
DENTAL ENAMEL MATRIX BIOMINERALIZATION Fincham, A. G., Belcourt, A. B., Termine, J. D., Butler, W. T., and Cothran, W. C. (1983) Amelogenins: Sequence homologies in enamel-matrix proteins from three mammalian species, Biochem. J. 211, 149154. Fincham, A. G., Bessem, C. C., Lau, E. C., Pavlova, Z., Shuler, C., Slavkin, H. C., and Snead, M. L. (1991a) Human developing enamel proteins exhibit a sex-linked dimorphism, Calcif. Tissue Int. 48, 288290. Fincham, A. G., Hu, Y., Lau, E. C., Slavkin, H. C., and Snead, M. L. (1991b) Amelogenin post-secretory processing during biomineralization in the postnatal mouse molar tooth, Arch. Oral Biol. 36, 305317. Fincham, A. G., Lau, E. C., Simmer, J., and Zeichner-David, M. (1992) Amelogenin biochemistryForm and function, in Slavkin, H. C., and Price, P. (Eds.), Chemistry and Biology of Mineralized Tissues, pp. 187201, Elsevier, Amsterdam. Fincham, A. G., and Moradian-Oldak, J. (1993) Amelogenin post-translational modications: Carboxy-terminal processing and the phosphorylation of bovine and porcine TRAP and LRAP amelogenins, Biochem. Biophys. Res. Commun. 197, 248255. Fincham, A. G., Moradian-Oldak, J., and Sarte, P. E. (1994a) Mass-spectrographic analysis of a porcine amelogenin identies a single phosphorylated locus, Calcif. Tissue Int. 55, 398400. Fincham, A. G., Moradian-Oldak, J., Simmer, J. P., Sarte, P. E., Lau, E. C., Diekwisch, T., and Slavkin, H. C. (1994b) Selfassembly of a recombinant amelogenin protein generates supramolecular structures, J. Struct. Biol. 112, 103109. Fincham, A. G., Moradian-Oldak, J., Diekwisch, T. G. H., Lyaruu, D. M., Wright, J. T., Bringas, P., Jr., and Slavkin, H. C. (1995) Evidence for amelogenin nanospheres as functional components of secretory-stage enamel matrix, J. Struct. Biol. 115, 5059. Fincham, A. G., and Moradian-Oldak, J. (1996) Comparative mass spectrometric analyses of enamel matrix proteins from ve species suggest a common pathway of post-secretory proteolytic processing, Connect. Tissue Res. 35, 151156. Fincham, A. G., and Simmer, J. P. (1997) Amelogenin proteins of developing dental enamel, Ciba Found. Symp. 205, 118130. Fincham, A. G., Leung, W., Tan, J., and Moradian-Oldak, J. (1998) Does amelogenin nanosphere assembly proceed through intermediary-sized structures? Connect. Tissue Res. 38, 237240. Fong, C. D., Hammarstrom, L., Lundmark, C., Wurtz, T., and Slaby, I. (1996a) Expression patterns of RNAs for amelin and amelogenin in developing rat molars and incisors, Adv. Dent. Res. 10, 195200. Fong, C. D., Slaby, I., and Hammarstrom, L. (1996b) Amelin: An enamel-related protein, transcribed in the cells of epithelial root sheath, J. Bone Miner. Res. 11, 892898. Fong, C. D., Cerny, R., Hammarstrom, L., and Slaby, I. (1998) Sequential expression of an amelin gene in mesenchymal and epithelial cells during odontogenesis in rats, Eur. J. Oral Sci. 106, 324330. Forsman, K., Lind, L., Backman, B., Westermark, E., and Holmgren, G. (1994) Localization of a gene for autosomal dominant amelogenesis imperfecta (ADAI) to chromosome 4q, Hum. Mol. Genet. 3, 16211625. Fukae, M., Ijiri, H., Tanabe, T., and Shimizu, M. (1979) Partial amino acid sequences of two proteins in developing porcine enamel, J. Dent. Res. 58(B), 10001001. Fukae, M., Tanabe, T., Ijiri, H., and Shimizu, M. (1980) Studies on porcine enamel proteins: A possible original enamel protein, Tsurumi Univ. Dent. J. 6, 8794. Fukae, M., and Shimizu, M. (1983) Amino acid sequence of the main component of porcine enamel proteins, Jpn. J. Oral Biol. 25(Suppl.), 29.
293
Fukae, M., and Tanabe, T. (1985) Separation of non-amelogenin component from puried amelogenin preparation of immature porcine enamel, Jpn. J. Oral Biol. 27, 12491251. Fukae, M., and Tanabe, T. (1987a) 45Ca-labeled proteins found in porcine developing dental enamel at an early stage of development, Adv. Dent. Res. 1, 261266. Fukae, M., and Tanabe, T. (1987b) Nonamelogenin components of porcine enamel in the protein fraction free from the enamel crystals, Calcif. Tissue Int. 40, 286293. Fukae, M., Tanabe, T., Uchida, T., Yamakoshi, Y., and Shimizu, M. (1993) Enamelins in the newly formed bovine enamel, Calcif. Tissue Int. 53, 257261. Fukae, M., Tanabe, T., Murakami, C., Dohi, N., Uchida, T., and Shimizu, M. (1996) Primary structure of porcine 89 kDa enamelin, Adv. Dent. Res. 10, 111118. Fukae, M., and Tanabe, T. (1998) Degradation of enamel matrix proteins in porcine secretory enamel, Connect. Tissue Res. 39, 123129. Gibson, C. W., Golub, E. E., Ding, W., Shimokawa, H., Young, M., Termine, J. D., and Rosenbloom, J. (1991) Identication of the leucine-rich amelogenin peptide (LRAP) as the translation product of an alternatively spliced transcript, Biochem. Biophys. Res. Comm. 174, 13061312. Gibson, C. W., Kucich, U., Collier, P., Shen, G., Decker, S., Bashir, M., and Rosenbloom, J. (1995) Analysis of amelogenin proteins using monospecic antibodies to dened sequences, Connect. Tissue Res. 32, 109114. Gibson, C. W., Collier, P. M., Yuan, Z. A., and Chen, E. (1998) DNA sequences of amelogenin genes provide clues to regulation of expression, Eur. J. Oral Sci. 106, 292298. Girondot, M., and Sire, J-Y. (1998a) Evolution of the amelogenin gene in toothed and toothless vertebrates, Eur. J. Oral Sci. 106, 501508. Girondot, M., Delgado, S., and Laurin, M. (1998b) Evolutionary analysis of hagsh amelogenin, Anat. Rec. 252, 608611. Glimcher, M. J., Mechanic, G. L., and Friberg, U. A. (1964) The amino acid composition of the organic matrix and the neutral soluble and acid soluble components of embryonic bovine enamel, Biochem. J. 93, 198202. Goldberg, M., Escaig, F., and Septier, D. (1984) Membrane and matrix-associated lipids in the developing enamel of the rat incisor, in Fearnhead, R. W., and Suga, S. (Eds.), Tooth Enamel IV, pp. 125130, Academic Press, New York. Goldberg, M., Vermelin, L., Mostermans, P., Lecolle, S., Septier, D., Godeau, C., and LeGeros, R. Z. (1998) Fragmentation of the distal portion of Tomes processes of secretory ameloblasts in the forming enamel of rat incisors, Connect. Tissue Res. 38, 159169. Goto, Y., Kogure, E., Takagi, T., Aimoto, S., and Aoba, T. (1993) Molecular conformation of porcine amelogenin in solution: Three folding units at the N-terminal, central, and C-terminal regions, J. Biochem. 113, 5560. Graham, E. E. (1985) Isolation of enamelin-like proteins from blue shark (Prionace glauca) enameloid, J. Exp. Zool. 234, 185191. Herold, R. C., Graver, H. T., and Christner, P. J. (1980) Immunohistochemical localization of amelogenins in enameloid of lower vertebrate teeth, Science 207, 13571358. Hu, C-C., Zhang, C. H., Qian, Q., Ryu, O. H., Moradian-Oldak, J., Fincham, A. G., and Simmer, J. P. (1996a) Cloning, DNA sequence and alternative splicing of opossum amelogenin mRNAs, J. Dent. Res. 75, 17281734. Hu, C. C., Bartlett, J. D., Zhang, C. H., Qian, Q., Ryu, O. H., and Simmer, J. P. (1996b) Cloning, cDNA sequence, and alternative splicing of porcine amelogenin mRNAs, J. Dent. Res. 75, 17351741.
294
FINCHAM, MORADIAN-OLDAK, AND SIMMER ogy of tooth enamel and enameloid, in Mechanisms of Tooth Enamel Formation, pp. 229237. Kemp, B. E., and Pearson, R. B. (1990) Protein kinase recognition sequence motifs, TIBS Sept., 342346. Kemp, N. E. (1985) Ameloblast secretion and calcication of the enamel layer in shark teeth, J. Morphol. 184, 215230. Kinoshita, Y. (1979) Incorporation of serum albumin into the developing dentine and enamel matrix in the rabbit incisor, Calcif. Tissue Int. 29, 4146. Krebsbach, P. H., Lee, S. K., Matsuki, Y., Kozak, C. A., Yamada, K., and Yamada, Y. (1996) Full-length sequence, localization, and chromosomal mapping of ameloblastin: A novel tooth-specic gene, J. Biol. Chem. 271, 44314435. Lagerstrom, M., Dahl, N., Iselius, L., Backman, B., and Petters son, U. (1990) Mapping of the gene for X-linked amelogenesis imperfecta by linkage analysis, Am. J. Hum. Genet. 46, 120125. Lagerstrom-Fermer, M., Nilsson, M., Backman, B., Salido, E., Shapiro, L., Pettersson, U., and Lasndegren, U. (1995) Amelogenin signal peptide mutation: Correlation between mutations in the amelogenin gene (AMGX) and manifestations of the X-linked amelogenesis imperfecta, Genomics 26, 159162. Landis, W. J., Heagle, A., Marko, M., Buttle, K., Fincham, A. G., Moradian-Oldak, M., and Bringas, P. (1998) Ultrastructure of forming enamel determined by tomographic (3D) image reconstruction, in Proceedings of Sixth International Conference on the Chemistry and Biology of Mineral Tissues, Abstract P92, Vittel, France. Lau, E. C., Mohandas, T. K., Shapiro, L. J., Slavkin, H. C., and Snead, M. L. (1989) Human and mouse amelogenin gene loci are on the sex chromosomes, Genomics 4, 162168. Lau, E. C., Simmer, J. P., Bringas, P., Hsu, D., Hu, C. C., Zeichner-David, M., Thiemann, F., Snead, M. L., Slavkin, H. C., and Fincham, A. G. (1992) Alternative splicing of the mouse amelogenin primary RNA transcript contributes to amelogenin heterogeneity, Biochem. Biophys. Res. Comm. 188, 12531260. Lee, S. K., Krebsbach, P. H., Matsuki, Y., Nanci, A., Yamada, K. M., and Yamada, Y. (1996) Ameloblastin expression in rat incisors and human tooth germs, Int. J. Dev. Biol. 40, 11411150. Lench, N. J., and Winter, G. B. (1995) Characterisation of molecular defects in X-linked amelogenesis imperfecta (AIH1), Hum. Mutat. 5, 251259. Lench, N. J., and Brook, A. H. (1997) DNA diagnosis of X-linked amelogenesis imperfecta (AIH1), J. Oral Pathol. Med. 26, 135137. Levine, P. T., Seyer, J., Huddleston, J., and Glimcher, M. J. (1967) The comparative biochemistry of the organic matrix proteins of developing enamel. I. Amino acid composition. Arch. Oral Biol. 12, 407410. Leytin, V., Arad, T., Matlis, S., Daphni, L., Mao, Z., Shay, B., Weiner, S., and Deutsch, D. (1998) Cryo-transmission electron microscopy and atomic force microscopy of the recombinant human amelogenin protein, self assembled into nanospheres, in Workshop on the Structure and Development of Dental Enamel, Univ. of Oslo, Norway. Li, R., Li, W., and DenBesten, P. K. (1995) Alternative splicing of amelogenin mRNA from rat incisor ameloblasts, J. Dent. Res. 74, 18801885. Li, W., Mathews, C., Gao, C., and DenBesten, P. (1998) Identication of two additional exons at the 3 end of the amelogenin gene, Arch. Oral Biol. 43, 497504. Limeback, H. (1991) Molecular mechanisms in dental hard tissue mineralization, Curr. Opin. Dent. 1, 826835. Limeback, H., and Simic, A. (1989) Porcine high molecular weight enamel proteins are primarily stable amelogenin aggregates
Hu, C-C., Ryu, O. H., Qian, Q., Zhang, C. H., and Simmer, J. P. (1997a) Cloning, characterization and heterologous expression of murine exon 4-containing amelogenin mRNAs, J. Dent. Res. 76, 641647. Hu, C-C., Fukae, M., Uchida, T., Qian, Q., Zhang, C. H., Ryu, O. H., Tanabe, T., Yamakoshi, Y., Murakami, C., Dohi, N., Shimizu, M., and Simmer, J. P. (1997b) Sheathlin: Cloning, cDNA/polypeptide sequences, and immunolocalization of porcine enamel proteins concentrated in the sheath space, J. Dent. Res. 76, 648657. Hu, C-C., Fukae, M., Uchida, T., Qian, Q., Zhang, C. H., Ryu, O. H., Tanabe, T., Yamakoshi, Y., Murakami, C., Dohi, N., Shimizu, M., and Simmer, J. P. (1997c) Cloning and characterization of porcine enamelin mRNAs, J. Dent. Res. 76, 1720 1729. Hu, C.-C., Simmer, J. P., Bartlett, J. D., Qian, Q., Zhang, C., Ryu, O. H., Xue, J., Fukae, M., Uchida, T., and MacDougall, M. (1998a) Murine enamelin: cDNA and derived protein sequences, Connect. Tissue Res. 39, 4762. Hu, C-C., Qian, Q., Zhang, C., and Simmer, J. P. (1998b) Cloning of human enamelin, in Robinson, C., and Goldberg, M. (Eds.), Proceedings of Sixth International Conference on the Chemistry and Biology of Mineralized Tissues, Abstract 82, Vittel, France. Hunter, G. K., Curtis, H. A., Grynpas, M. D., Simmer, J. P., and Fincham, A. G. (in press) Effects of recombinant amelogenin on hydroxyapatite formation in vitro. Calcif. Tissue Int. Hunter, J. (1771) The Natural History of the Human Teeth: Explaining their structure, use, formation, growth and diseases, J. Johnson, London. 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, 437481. Iijima, M., Moriwaki, Y., and Kuboki, Y. (1998) Effect of some physico-chemical properties of matrix on lengthwise and oriented growth of mineral crystals, Connect. Tissue Res. 38, 171180. Inai, T., Kukita, T., Ohsaki, Y., Nagata, K., Kukita, A., and Kurisu, K. (1991) Immunohistochemical demonstration of amelogenin penetration towards the dental pulp in the early stages of ameloblast development in rat molar tooth germs, Anat. Rec. 229, 259270. Irving, J. T. (1963a) The sudanophil material at sites of calcication, Arch. Oral Biol. 8, 735743. Irving, J. T. (1963b) Calcication of the organic matrix of enamel, Arch. Oral Biol. 8, 773774. Ishiyama, M., Mikami, M., Shimokawa, H., and Oida, S. (1998) Amelogenin protein in tooth germs of the snake Elaphe quadrivirgata, immuno-histochemistry, cloning and cDNA sequence, Arch. Histol. Cytol. 61, 467474. Jodaikin, A., Traub, W., and Weiner, S. (1986) Protein conformation in rat tooth enamel, Arch. Oral Biol. 31, 685689. Jodaikin, A., Weiner, S., Perl-Treves, D., Traub, W., and Termine, J. D. (1987) Developing enamel matrix proteins: A conformation study of enamelins and amelogenins, Int. J. Biol. Macromol. 9, 166168. Katz, E. P., Mechanic, G. L., and Glimcher, M. J. (1965) The ultracentrifugal and free zone electrophoretic characterization of the neutral soluble proteins of embryonic bovine enamel, Biochim. Biophys. Acta 107, 471484. Kawasaki, H., Kawaguchi, T., Yano, T., Fujimura, S., and Yago, M. (1980) Chemical nature of proteins in the placoid scale of the blue shark, Prionace glauca L, Arch. Oral Biol. 25, 313320. Kawasaki, K., and Fearnhead, R. W. (1983) Comparative histol-
DENTAL ENAMEL MATRIX BIOMINERALIZATION and serum albumin-derived proteins, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 269273, Florence, Yokohama. Limeback, H., Sakarya, H., Chu, W., and MacKinnon, M. (1989) Serum albumin and its acid hydrolysis peptides dominate preparations of mineral-bound enamel proteins, J. Bone Miner. Res. 4, 235241. Llano, E., Pendas, A. M., Knauper, V., Sorsa, T., Salo, T., Salido, E., Murphy, G., Simmer, J. P., Bartlett, J. D., and Lopez-Ortin, C. (1997) Identication and structural and functional characterization of human enamelysin (MMP-20), Biochemistry 36, 15101 15108. Lowenstam, H. A., and Weiner, S. (1989) On Biomineralization, Oxford Univ. Press, New York. Lyaruu, D. M., Hu, C.-C., Zhang, C., Qian, Q., Ryu, O. H., Moradian-Oldak, J., Woltgens, J. H. M., Fincham, A. G., and Simmer, J. P. (1998) Derived protein and cDNA sequences of hamster amelogenin, Eur. J. Oral Sci. 106, 299307. Lyngstadaas, S. P., Risnes, S., Sproat, B. S., Thrane, P., and Prydz, H. P. (1995) A synthetic, chemically modied ribozyme eliminates amelogenin, the major translation product in developing mouse enamel in vivo. EMBO J. 14, 52245229. MacDougall, M., DuPont, B. R., Simmons, D., Reus, B., Krebsbach, P., Karrman, C., Holmgren, G., Leach, R. J., and Forsman, K. (1997) Ameloblastin gene (AMBN) maps within the critical region for autosomal dominant amelogenesis imperfecta at chromosome 4q21, Genomics 41, 115118. MacDougall, M., Gu, T. T., Simmons, D., Luan, X., Cavender, A., and DSouza, R. D. (1998a) Developmental regulation of dentin sialophosphoprotein during ameloblast differentiation: A potential enamel matrix nucleator, Connect. Tissue Res. 39, 2537. MacDougall, M., Simmons, D., Dodds, A., Knight, C., Luan, X., Zeichner-David, M., Zhang, C., Ryu, O. H., Qian, Q., Simmer, J. P., and Hu, C-C. (1998b) Cloning, characterization and tissue expression pattern of mouse tuftelin cDNA, J. Dent. Res. 77, 19701978. Mann, S. (1989) Crystallochemical strategies in biomineralization, in Mann, S., Webb, J., and Williams, R., (Eds.), On Biomineralization: Chemical and Biochemical Perspectives, pp. 3562, Weinheim, New York. Matsushima, N., Izumi, Y., and Aoba, T. (1998) Small-angle X-ray scattering and computer-aided molecular modeling studies of 20 kDa fragment of porcine amelogenin: Does amelogenin adopt an elongated bundle structure? J. Biochem. 123, 150156. McKee, M. D., Martineau-Doize, B., and Warshawsky, H. (1986) Penetration of various molecular-weight proteins into the enamel organ and enamel of the rat incisor, Arch. Oral Biol. 31, 287296. Mechanic, G. L. (1971) The multicomponent re-equilibrating protein system of bovine embryonic enamelin (dental enamel protein): Chromatography in deaggregating solvents, in Fearnhead, R. W. and Stack, M. V. (Eds.), Tooth Enamel II, pp. 8892, Wright Bros., Bristol. Meinke, D. K. (1982) A histological and histochemical study of developing teeth in Polypterus (Pisces, Actinopterygii), Arch. Oral Biol. 27, 197206. Menanteau, J., Gregoire, M., Daculsi, G., and Jans, I. (1987) In vitro albumin binding on apatite crystals from developing enamel, J. Bone Miner. Res. 3, 137141. Miake, Y., Aoba, T., Moreno, E. C., Shimoda, S., Prostak, K., and Suga, S. (1991) Ultrastructural studies on crystal growth of enameloid minerals in Elasmobranch and Teleost sh, Calcif. Tissue Int. 48, 204217. Moe, D., and Birkedal-Hansen, H. (1979) Proteolytic activity in developing bovine enamel, J. Dent. Res. 58B, 10121013. Moradian-Oldak, J., Frolow, F., Addadi, L., and Weiner, S. (1992)
295
Interactions between acidic matrix macromolecules and calcium phosphate ester crystals: Relevance to carbonate apatite formation in biomineralization, Proc. R. Soc. London Biol. 247, 4755. Moradian-Oldak, J., Simmer, P. J., Sarte, P. E., Zeichner-David, M., and Fincham, A. G. (1994a) Specic cleavage of a recombinant murine amelogenin by a protease fraction isolated from bovine tooth enamel, Arch. Oral Biol. 39, 647656. Moradian-Oldak, J., Simmer, P. J., Lau, E. C., Sarte, P. E., Slavkin, H. C., and Fincham, A. G. (1994b) Detection of monodisperse aggregates of a recombinant amelogenin by dynamic light scattering, Biopolymers 34, 13391347. Moradian-Oldak, J., Lau, E. C., Diekwisch, T., Slavkin, H. C., and Fincham, A. G. (1995) A review of the aggregation properties of a recombinant amelogenin, Connect. Tissue Res. 32, 125130. Moradian-Oldak, J., Sarte, P. E., and Fincham, A. G. (1996) Description of two classes of proteinases from enamel extracellular matrix cleaving a recombinant amelogenin, Connect. Tissue Res. 35, 231238. Moradian-Oldak, J., Tan, J., and Fincham, A. G. (1998a) Interaction of amelogenin with hydroxyapatite crystals: An adherence effect through amelogenin molecular self-association, Biopolymers 46, 225238. Moradian-Oldak, J., Leung, W., Tan, J., and Fincham, A. G. (1998b) Effect of apatite crystals on the activity of amelogenin degrading enzymes in vitro, Calcif. Tissue Int. 39, 131140. Moradian-Oldak, J., Leung, W., and Fincham, A.G. (1998c) temperature and pH-dependent supramolecular self-assembly of amelogenin molecules: a dynamic light-scattering analysis, J. Struct. Biol., 122, 320327. Murakami, C., Dohi, N., Fukae, M., Tanabe, T., Yamakoshi, Y., Wakida, K., Satoda, T., Takahashi, O., Shimizu, M., Ryu, O. H., Simmer, J. P., and Uchida, T. (1997) Immunochemical and immunohistochemical study of 27 and 29 kDa calcium binding proteins and related proteins in the porcine tooth germ, Histochem. Cell Biol. 107, 485494. Nanci, A., Bringas, P. J., Samuel, N., and Slavkin, H. C. (1983) Selachian tooth development. III. Ultrastructural features of secretory amelogenesis in Squalus acanthias, J. Craniofac. Genet. Dev. Biol. 3, 5373. Nanci, A., Ahluwalia, J. P., Zalzal, S., and Smith, C. E. (1989) Cytochemical and biochemical characterization of glycoproteins in forming and maturing enamel of the rat incisor, J. Histochem. Cytochem. 37, 16191633. Nanci, A., Kawaguchi, H., and Kogaya, Y. (1994) Ultrastructural studies and immunolocalization of enamel proteins in rodent secretory stage ameloblasts processed by various cryoxation methods, Anat. Rec. 238, 425436. Nanci, A., Zalzal, S., Lavoie, P., Kunikata, M., Chen, W-Y., Krebsbach, P. H., Yamada, Y., Hammarstorm, L., Simmer, J. P., Fincham, A. G., Snead, M. L., and Smith, C. E. (1998) Comparative immunochemical analyses of the developmental expression and distribution of ameloblastin and amelogenin in rat incisors, J. Histochem. Cytochem. 46, 911934. Nanci, A., Zalzal, S., Hashimoto, J., Lavoie, P., Chen, W-Y., Simmer, J. P., Krebsbach, P. H., Yamada, Y., Slaby, I., Hammarstrom, L., and Smith, C. E. (1998). Protein synthesis by ameloblasts and selective accumulation of some nonamelogenins at enamel growth sites, Connect. Tissue Res. 38, 12. Nancollas, G. H., and Budz, J. A. (1990) Analysis of particle size distribution of hydroxyapatite crystallites in the presence of synthetic and natural polymers, J. Dent. Res. 69, 16781685. Nikiforuk, G., and Simmons, N. S. (1965) Purication and properties of protein from embryonic bovine enamel, J. Dent. Res. 44, 11191122.
296
FINCHAM, MORADIAN-OLDAK, AND SIMMER (Ed.), The Chemistry and Biology of Mineralized Tissues, pp. 248263, EBSCO Media, Birmingham, Ala. Robinson, C., Kirkham, J., and Hallsworth, A. S. (1988) Volume distribution and concentration of protein, mineral and water in developing bovine enamel, Archs. Oral Biol. 33, 159162. Robinson, C., Kirkham, J., and Fincham, A. G. (1989a) The enamelin/non-amelogenin problem. A brief review, Connect. Tissue Res. 22, 93100. Robinson, C., Shore, R. C., and Kirkham, J. (1989b) Tuft protein: its relationship with the keratins and the developing enamel matrix, Calcif. Tissue Int. 44, 393398. Robinson, C., Kirkham, J., Stonehouse, N. J., and Shore, R. C. (1989c) Extracellular processing of enamel matrix and origin and function of tuft protein, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 5963, Florence Publishers, Yokohama. Robinson, C., Kirkham, J., Brookes, S. J., and Shore, R. C. (1992) The role of albumin in developing rodent dental enamel: A possible explanation for white spot hypoplasia, J. Dent. Res. 71, 12701274. Robinson, C., Kirkham, J., Brookes, S. J., and Shore, R. C. (1995) Chemistry of mature enamel, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 167191, CRC Press, Boca Raton, Fl. Robinson, C., Brookes, S. J., Kirkham, J., Bonass, W. A., and Shore, R. C. (1996) Crystal growth in dental enamel: the role of amelogenins and albumin, Adv. Dent. Res. 10, 179180. Robinson, C., Kirkham, J., Shore, R. C., Brookes, S. J., and Wood, S. R. (1998a) Matrix function and the tuft enigma: a role in directing enamel microarchitecture, in Robinson, C., Goldberg, M. (Eds.), Proceedings of Sixth International Conference on the Chemistry and Biology of Mineralized Tissues, Vittel, France. [Abstract] Robinson, C., Brookes, S. J., Shore, R. C., and Kirkham, J. (1998b) The developing enamel matrix: Nature and function, Eur. J. Oral Sci. 106, 282291. Ronnholm, E. (1962) The amelogenesis of human teeth as re vealed by electron microscopy. II, J. Ultrastruct. Res. 6, 249 303. Ryu, O. H., Hu, C-C., and Simmer, J. P. (1996) Comparative HPLC, SDS-PAGE, and immunoblot analyses of dental enamel proteins, Adv. Dent. Res. 10, 150158. Ryu, O. H., Hu, C. C., Zhang, C., Quian, Q., Moradian-Oldak, J., Fincham, A. G., and Simmer, J. P. (1998a) Proteolytic activity of opossum tooth extracts, Eur. J. Oral Sci. 106, 337344. Ryu, O. H., Hu, C.-C., and Simmer, J. P. (1998b) Biochemical characterization of recombinant mouse amelogenins: Protein quantitation, proton absorption, and relative affinity for enamel crystals, Connect. Tissue Res. 38, 207214. Ryu, O. H., Fincham, A. G., Hu, C-C., Zhang, C., Qian, Q., Bartlett, J. D., and Simmer, J. P. (1999) Characterization of recombinant enamelysin activity and cleavage of recombinant pig and mouse amelogenin, J. Dent. Res. 78, 743750. Salaih, E., Huang, J., Strawich, E., Gouverneur, M., and Glimcher, M. J. (1998) Enamel specic protein kinases and state of phosphorylation of puried amelogenins, Connect Tissue Res. 38, 225236. Salido, E. C., Yen, P. H., Koprivnikar, K., Yu, L-C., and Shapiro, L. J. (1992) The human enamel protein gene amelogenin is expressed from both the X and Y chromosomes, Am. J. Hum. Genet. 50, 303316. Sasagawa, I., and Ishiyama, M. (1988) The structure and development of the collar enameloid in two teleost shes, Halichoeres poecilopterus and Pagrus major, Anat. Embryol. Berlin 178, 499511.
Okamura, K. (1983) Localization of serum albumin in dentin and enamel, J. Dent. Res. 62, 100104. Osborn, J. W. (1969) The three-dimensional morphology of the tufts in human enamel, Acta Anat. 73, 481. Overall, C. M., and Limeback, H. (1988) Identication and characterization of enamel proteinases isolated from developing enamel. Amelogeninolytic serine proteinases are associated with enamel maturation in pig, Biochem. J. 256, 965972. Paine, M. L., and Snead, M. L. (1997) Protein interactions during assembly of the enamel organic extracellular matrix, J. Bone Miner. Res. 12, 221227. Paine, C. T., Paine, M. L., and Snead, M. L. (1998a) Identication of tuftelin- and amelogenin-interacting proteins using the yeast two-hybrid system, Correct. Tissue Res. 38, 257267. Paine, M. L., Krebsbach, P. H., Chen, L. S., Paine, C. T., Yamada, Y., Deutsch, D., and Snead, M. L. (1998b) Protein-to-protein interactions: Criteria dening the assembly of the enamel organic matrix, J. Dent. Res. 77, 496502. Palamara, J., Phakey, P. P., Rachinger, W. A., and Orams, H. J. (1989) Ultrastructure of spindles and tufts in human dental enamel, Adv. Dent. Res. 3, 249257. Papas, A., Seyer, J. M., and Glimcher, M. J. (1977) Isolation from embryonic bovine dental enamel of a polypeptide (E3) containing as its only phosphorylated sequence, Glu-O-phosphoserineLeu, FEBS Lett. 79, 276280. Pautard, F. G. E. (1961) An X-ray diffraction pattern from human enamel matrix, Arch. Oral Biol. 3, 217220. Pautard, F. G. E. (1963) Mineralization of keratin and its comparison with the enamel matrix, Nature 199, 531535. Piez, K. A. (1960) The nature of the protein matrix of human enamel, J. Dent. Res. 39, 712. Poole, D. F. (1967) Phylogeny of tooth tissues: Enameloid and enamel in recent vertebrates with a note on the history of cementum, in Miles, A. E. W. (Ed.), Structural and Chemical Organization of Teeth, pp. 111150, Academic Press, New York. Prostak, K. S., and Skobe, Z. (1988) Ultrastructure of odontogenic cells during enameloid matrix synthesis in tooth buds from an Elasmobranch, Raja erinacae, Am. J. Anat. 182, 5972. Prostak, K., Seifert, P., and Skobe, Z. (1989) Ultrastructure of developing teeth in the gar pike, (Lepisosteus), in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 188192, Florence, Yokohama. Ravindranath, M. H. R., Moradian-Oldak, J., and Fincham, A. G. (1999) Tyrosyl motif in amelogenins binds N-acetyl-D-glucosamine, J. Biol. Chem. 274, 24642471. Renugopalakrishnan, V., Pattabiraman, N., Prabhakaran, M., Strawich, E., and Glimcher, M. J. (1989) Tooth enamel protein, amelogenin, has a probable -spiral internal channel within a single polypeptide chain: Preliminary molecular mechanics and dynamics studies, Biopolymers 28, 597603. Robinson, C., Lowe, N. R., and Weatherell, J. A. (1975) The amino acid composition, distribution and origin of tuft protein in human and bovine dental enamel. Arch. Oral Biol. 20, 2942. Robinson, C., Fuchs, P., and Weatherell, J. A. (1981) The appearance of developing rat incisor enamel using a freeze fracturing technique, J. Cryst. Growth 53, 160165. Robinson, C., and Kirkham, K. (1984) Enamel matrix components, alterations during development and possible interactions with the mineral phase, in Fearnhead, R. W., and Suga, S. (Eds.), Tooth Enamel IV, pp. 261265, Acad. Press, New York. Robinson, C., and Kirkham, J. (1985) Dynamics of amelogenesis as revealed by protein compositional studies, in Butler, W. T.
DENTAL ENAMEL MATRIX BIOMINERALIZATION Sasagawa, I. (1989) The ne structure of initial mineralisation during tooth development in the gummy shark, Mustelus manazo, Elasmobranchia., J. Anat., 164, 175187. Sasaki, S., and Shimokawa, H. (1995) The amelogenin gene, Int. J. Dev. Biol. 39, 127133. Sasaki, S., Takagi, T., Suzuki, M., Baba, T., and Minegishi, K. (1984) Amino acid sequence of developing bovine enamel protein, in Fearnhead, R. W., and Suga, S. (Eds.), Tooth Enamel IV, pp. 151155, Academic Press, New York. Sasaki, S., Takagi, T., and Susuki, M. (1991a) Cyclical changes in pH in bovine developing enamel as sequential bands, Arch. Oral Biol. 36, 227231. Sasaki, S., Takagi, T., and Suzuki, M. (1991b) Amelogenin degradation by an enzyme having acidic pH optimum and the presence of acidic zones in developing bovine enamel, in Suga, S., and Nakahara, H. (Eds.), Mechanisms and Phylogeny of Mineralization in Biological Systems, pp. 7981, Springer, Tokyo. Sasaki, T., Takagi, M., and Yanagisawa, T. (1997) Structure and function of secretory ameloblasts in enamel formation, Ciba Found. Symp. 205, 3246. Sculley, J. L., Bartlett, J. D., Chaparian, M. G., Fukae, M., Uchida, T., Xue, J., Hu, C.-C., and Simmer, J. P. (1998) Enamel matrix serine proteinase 1: Stage specic expression and molecular modeling, Connect. Tissue Res. 39, 111122. Seyer, J. M. (1972) Bovine enamel proteins, in Slavkin, H. C. (Ed.), The Comparative Molecular Biology of Extracellular Matrices, pp. 273295, Academic Press, New York. Seyer, J. M., and Glimcher, M. J. (1977a) Evidence for the presence of numerous protein components in immature bovine dental enamel, Calcif. Tissue Res. 24, 253257. Seyer, J. M., and Glimcher, M. J. (1977b) Isolation, characterization and partial amino acid sequence of a phosphorylated polypeptide (E4) from bovine embryonic dental enamel, Biochim. Biophys. Acta 493, 44151. Shapiro, I. M., and Amdur, B. H. (1965) Enamel matrix pigmentation in the developing bovine tooth, Arch. Oral Biol. 10, 10151018. Shapiro, I. M., Wuthier, R. E., and Irving, J. T. (1966) A study of the phospholipids of bovine dental tissues. I. Enamel matrix and dentine, Arch. Oral Biol. 11, 501512. Shellis, R. P. (1975) A histological and histochemical study of the matrices of enameloid and dentine in teleost shes, Arch. Oral Biol. 20, 183187. Shellis, R. P., and Miles, A. E. W. (1976) Observations with the electron microscope on enameloid formation in the common eel (Anguilla anguilla: Teleosti), Proc. R. Soc. London 194, 253 269. Shimokawa, H., Ogata, Y., Sasaki, S., Sobel, M. E., McQuillan, C. I., Termine, J. D., and Young, M. F. (1987) Molecular cloning of bovine amelogenin cDNA, Adv. Dent. Res. 1, 293297. Shore, R. C., Robinson, C., Kirkham, J., and Brookes, S. J. (1995a) Structure of developing enamel, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 135150, CRC Press, Boca Raton, FL. Shore, R. C., Robinson, C., Kirkham, J., and Brookes, S. J. (1995b) Structure of mature enamel, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 151166, CRC Press, Boca Raton, FL. Simmer, J. P. (1995) Alternative splicing of amelogenins, Connect. Tissue Res. 32, 131136. Simmer, J. P., and Fincham, A. G. (1995) Molecular mechanisms
297
of dental enamel formation, Crit. Rev. Oral Biol. Med. 6, 84108. Simmer, J. P., and Snead, M. L. (1995) Molecular biology of the amelogenin gene, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 5984, CRC Press, Boca Raton, FL. Simmer, J. P., Hu, C. C., Lau, E. C., Sarte, P. E., Moradian-Oldak, J., Slavkin, H. C., and Fincham, A. G. (1994a) Alternative splicing of the mouse amelogenin primary RNA transcript, Calcif. Tissue Int. 55, 302310. Simmer, J. P., Lau, E. C., Hu, C-C., Aoba, T., Lacey, M., Nelson, D., Zeichner-David, M., Snead, M. L., Slavkin, H. C., and Fincham, A. G. (1994b) Isolation and characterization of a mouse amelogenin expressed in Escherichia coli, Calcif. Tissue Int. 54, 312319. Simmer, J. P., Fukae, M., Tanabe, T., Yamakoshi, Y., Uchida, T., Xue, J., Margolis, H. C., Shimizu, M., DeHart, B. C., Hu, C. C., and Bartlett, J. D. (1998) Purication, characterization, and cloning of enamel matrix serine proteinase 1, J. Dent. Res. 77, 377386. Simmons, D., Gu, T. T., Krebsbach, P. H., Yamada, Y., and MacDougall, M. (1998) Identication and characterization of a cDNA for mouse ameloblastin, Connect. Tissue Res. 39, 312. Skobe, Z., Stern, D. N., and Prostak, K. S. (1995) The cell biology of amelogenesis, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 2357, CRC Press, Boca Raton, FL. Slavkin, H. C., and Diekwisch, T. G. (1997) Molecular strategies of tooth enamel formation are highly conserved during vertebrate evolution, Ciba Found. Symp. 205, 7380. Smales, F. C. (1975) Structural sub-units in prisms of immature rat enamel, Nature 258, 772774. Smith, C. E. (1979) Ameloblasts: Secretory and resorptive functions, J. Dent. Res. 58B, 703. Smith, C. E., Pompura, J. R., Borenstein, S., Fazel, A., and Nanci, A. (1989a) Degradation and loss of matrix proteins from developing enamel, Anat. Rec. 224, 292316. Smith, C. E., Borenstein, S., Fazel, A., and Nanci, A. (1989b) In vitro studies of the proteinases which degrade amelogenins in developing rat incisor enamel, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 286290, Florence, Yokohama. Smith, C. E., Chen, W. Y., Issid, M., and Fazel, A. (1995a) Enamel matrix protein turnover during amelogenesis: Basic biochemical properties of short-lived sulfated enamel proteins, Calcif. Tissue Int. 57, 133144. Smith, C. E., and Nanci, A. (1995b) Overview of morphological changes in enamel organ cells associated with major events in amelogenesis, Int. J. Dev. Biol. 39, 153161. Smith, C. E., Issid, M., Margolis, H. C., and Moreno, E. C. (1996a) Developmental changes in the pH of enamel uid and its effects on matrix-resident proteinases, Adv. Dent. Res. 10, 111. Smith, C. E., and Nanci, A. (1996b) Protein dynamics of amelogenesis, Anat. Rec. 245, 186207. Smith, C. E. (1998) Cellular and chemical events during enamel maturation, Crit. Rev. Oral Biol. Med. 9, 128161. Smith, M. M., and Coates, M. I. (1998) Evolutionary origins of the vertebrate dentition: Phylogenetic patterns and developmental evolution, Eur. J. Oral Sci. 106, 482500. Snead, M. L., Lau, E. C., Zeichner-David, M., Fincham, A. G., Woo, S. L., and Slavkin, H. C. (1985) DNA sequence for cloned cDNA for murine amelogenin reveal the amino acid sequence for enamel-specic protein, Biochem. Biophys. Res. Commun. 129, 812818.
298
FINCHAM, MORADIAN-OLDAK, AND SIMMER Uchida, T., Tanabe, T., Fukae, M., and Shimizu, M. (1991a) Immunocytochemical and immunochemical detection of a 32 kDa nonamelogenin and related proteins in porcine tooth germs, Arch. Histol. Cytol. 54, 527538. Uchida, T., Tanabe, T., Fukae, M., Shimizu, M., Yamada, M., Miake, K., and Kobayashi, S. (1991b) Immunochemical and immunohistochemical studies, using antisera against porcine 25 kDa amelogenin, 89 kDa enamelin and the 1317 kDa nonamelogenins, on immature enamel of the pig and rat, Histochemistry 96, 129138. Uchida, T., Fukae, M., Tanabe, T., Yamakoshi, Y., Satoda, T., Murakami, C., Takahashi, O., and Shimizu, M. (1995) Immunochemical and immunocytochemical study of a 15 kDa nonamelogenin and related proteins in the porcine immature enamel: Proposal of a new group of enamel proteins Sheath Proteins, Biomed. Res. 16, 131140. Uchida, T., Murakami, C., Dohi, N., Wakida, K., Satoda, T., and Takahashi, O. (1997) Synthesis, secretion, degradation and fate of ameloblastin during the matrix formation stage of the rat incisor as shown by immunocytochemistry and immunochemistry using region-specic antibodies, J. Histochem. Cytochem. 45, 13291340. Uchida, T., Murakami, C., Wakida, K., Dohi, N., Iwai, Y., Simmer, J. P., Fukae, M., Satoda, T., and Takahashi, O. (1998) Sheath proteins: Synthesis, secretion, degradation and fate in forming enamel, Eur. J. Oral Sci. 106, 308314. Uehara, K., and Miyoshi, S. (1987) Structure of the upper teeth of the lesh, Stephanolepis cirrhifer, Anat. Rec., 217, 1622. Wakida, K., Amizuka, N., Murakami, C., Satoda, T., Fukae, M., Simmer, J. P., Ozawa, H., and Uchida, T. (1999) Maturation ameloblasts of the porcine tooth germ do not express amelogenin, J. Histochem. Cytochem. 111, 297303. Warshawsky, H. (1985) Ultrastructural studies of amelogenesis, in Butler, W. T. (Ed.), Proceedings of the Second International Conference on the Chemistry and Biology of Mineralized Tissues, pp. 3345, Ebsco Media, Birmingham, AL. Watson, M. L. (1960) The extracellular nature of enamel in the rat, J. Biophys. Biochem. Cytol. 7, 489492. Weatherell, J. A., Weidmann, S. M., and Eyre, D. R. (1968) Histological appearance and chemical composition of enamel proteins from mature human molars, Caries Res. 2, 281293. Wen, H. B., Moradian-Oldak, J., Leung, W., Bringas, P. Jr., and Fincham, A. G. (1999) Microstructures of an amelogenin gel matrix, J. Struct. Biol. 126, 4251. Williamson, M. P. (1994) The structure and function of prolinerich regions in proteins, Biochem. J. 297, 249260. Wright, W. G., and Eastoe, J. E. (1989) Comparison of rodent and lagomorph enamel proteins, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 294298, Florence, Yokohama. Wright, J. T., Hall, K. I., and Yamauche, M. (1997) The enamel proteins in human amelogenesis imperfecta, Arch. Oral Biol. 42, 149159. Wurtz, T., Lundmark, C., Christersson, C., Bawden, J. W., Slaby, I., and Hammarstrom, L. (1996) Expression of amelogenin mRNA sequences during development of rat molars, J. Bone Miner. Res. 11, 125131. Wuthier, R. E., and Irving, J. T. (1964) Lipids in developing calf bone, J. Dent. Res. 43, 814815. Yamakoshi, Y. (1995) Carbohydrate moieties of porcine 32kDa enamelin, Calcif. Tissue Int. 56, 323330. Yamakoshi, Y., Tanabe, T., Fukae, M., and Shimizu, M. (1989) Amino acid sequence of porcine 25kDa Amelogenin, in Fearnhead, R. W. (Ed.), Tooth Enamel V, pp. 314321, Florence, Yokohama.
Snead, M. L., Luo, W., Lau, E. C., and Slavkin, H. C. (1988) Spatial- and temporal-restricted pattern for amelogenin gene expression during mouse molar tooth organogenesis, Development 104, 7785. Strawich, E., and Glimcher, M. J. (1989) Major enamelin protein in enamel of developing bovine teeth is albumin, Connect. Tissue Res. 22, 111121. Strawich, E., and Glimcher, M. J. (1990) Tooth enamelins identied mainly as serum proteins. Major enamelin is albumin, Eur. J. Biochem. 191, 4756. Strawich, E., Seyer, J., and Glimcher, M. J. (1993) Immunoidentication of two non-amelogenin proteins of developing bovine enamel isolated by affinity chromatography. Further proof that tooth enamelins are mainly serum proteins, Connect. Tissue Res. 29, 163169. Suga, S. (1970) Histochemical observations of proteolytic enzyme activity in the developing dental hard tissues of the rat, Arch. Oral Biol. 15, 555558. Suga, S., Wada, K., and Ogawa, M. (1978) Mineralization pattern and uoride distribution of the developing and matured enameloid of the shark, Jpn. J. Oral Biol. 20, 6781. Suga, S., Wada, K., and Ogawa, M. (1981) Fluoride concentration in teeth of tetraodontiform shes and its phylogenetic signicance, Jpn. J. Ichthyol 28, 304312. Takagi, T., Suzuki, M., Baba, T., Minegishi, K., and Sasaki, S. (1984) Complete amino acid sequence of amelogenin in developing bovine enamel, Biochem. Biophys. Res. Commun. 121, 592597. Tan, J., Leung, W., Moradian-Oldak, J., Zeichner-David, M., and Fincham, A. G. (1998) Quantitative analysis of amelogenin solubility, J. Dent. Res. 77, 13881396. Tanabe, T., Aoba, T., Moreno, E. C., Fukae, M., and Shimuzu, M. (1990) Properties of phosphorylated 32 kd nonamelogenin proteins isolated from porcine secretory enamel, Calcif. Tissue Int. 46, 205215. Tanabe, T., Fukae, M., Uchida, T., and Shimizu, M. (1992) The localization and characterization of proteinases for the initial cleavage of porcine amelogenin, Calcif. Tissue Int. 51, 213217. Tanabe, T., Fukae, M., and Shimizu, M. (1994) Degradation of enamelins by proteinases found in porcine secretory enamel in vitro, Arch. Oral Biol. 39, 277281. Termine, J. D., and Torchia, D. A. (1980) 13C-1H magnetic doubleresonance study of fetal enamel matrix proteins, Biopolymers 19, 741750. Termine, J. D., Belcourt, A. B., Christner, P. J., Conn, K. M., and Nylen, M. U. (1980) Properties of dissociatively extracted fetal tooth matrix proteins. Principal molecular species in developing bovine enamel, J. Biol. Chem. 20, 97609768. Thesleff, I. (1995) Differentiation of ameloblasts and its regulation by epithelialmesenchymal interactions, in Robinson, C., Kirkham, J., and Shore, R. C. (Eds.), Dental Enamel, Formation to Destruction, pp. 122, CRC Press, Boca Raton, FL. Thesleff, I., and Berg, T. (1997) Tooth morphogenesis and the differentiation of ameloblasts, Ciba Found. Symp. 205, 312. Toyosawa, S., OhUigin, C., Figueroa, F., Tichy, H., and Klein, J. (1998) Identication and characterization of amelogenin genes in monotremes, reptiles, and amphibians, Proc. Natl. Acad. Sci. USA 95, 1305613061. Travis, D. F., and Glimcher, M. J. (1964) The structure and organization of, and the relationships between the organic matrix and the inorganic crystals of embryonic bovine enamel, J. Cell. Biol. 23, 477497.
DENTAL ENAMEL MATRIX BIOMINERALIZATION Yamakoshi, Y., Pinheiro, F. H., Tanabe, T., Fukae, M., Shimizu, M. (1998) Sites of asparagine-linked oligosaccharides in porcine 32kDa enamelin, Connect. Tissue Res. 39, 3946. Yuan, Z. A., Collier, P. M., Rosenbloom, J., and Gibson, C. W. (1996) Analysis of amelogenin mRNA during bovine tooth development, Arch. Oral Biol. 41, 205213. Zalut, C., Henzel, W. J., and Harris, H. W., Jr. (1980) Micro quantitative Edman manual sequencing, J. Biochem. Biophys. Meth. 3, 1130. Zeichner-David, M., Vides, J., MacDougall, M., Fincham, A. G., Snead, M. L., Bessem, C., and Slavkin, H. C. (1988) Biosynthe-
299
sis and characterization of rabbit tooth enamel extracellularmatrix proteins, Biochem. J. 251, 631641. Zeichner-David, M., Vo, H., Tan, H., Diekwisch, T., Berman, B., Thiemann, F., Alcocer, D., Hsu, P., Wang, T., Eyna, J., Caton, J., Slavkin, H. C., and MacDougall, M. (1997) Timing and expression of enamel gene products during mouse tooth development, Int. J. Dev. Biol. 41, 2738. Zheng, S., Tu, A. T., Renugopalakrishnan, V., Strawich, E., and Glimcher, M. J. (1987) A mixed beta-turn and beta-sheet structure for bovine tooth enamel amelogenin: Raman spectroscopic evidence, Biopolymers 26, 18091813.

Desarrollo Matriz Esmalte

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Desarrollo Matriz Esmalte

Hochgeladen von

Copyright:

Verfügbare Formate

Journal of Structural Biology 126, 270299 (1999) Article ID jsbi.1999.4130, available online at http://www.idealibrary.

The Structural Biology of the Developing Dental Enamel Matrix

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

3.2. The Amelogenins

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

4.4. Amelogenin Mineral Interactions

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

FINCHAM, MORADIAN-OLDAK, AND SIMMER

DENTAL ENAMEL MATRIX BIOMINERALIZATION

Das könnte Ihnen auch gefallen