P ost-translational M odifications of SIBLING P roteins and T heir R oles in O steogenesis and D entinogenesis

Abstract

The extracellular matrix (ECM) of bone and dentin contains several non-collagenous proteins. One category of non-collagenous protein is termed the SIBLING (Small Integrin-Binding LIgand, N-linked Glycoprotein) family, that includes osteopontin (OPN), bone sialoprotein (BSP), dentin matrix protein 1 (DMP1), dentin sialophosphoprotein (DSPP), and matrix extracellular phosphoglycoprotein (MEPE). These polyanionic SIBLING proteins are believed to play key biological roles in the mineralization of bone and dentin. Although the specific mechanisms involved in controlling bone and dentin formation are still unknown, it is clear that some functions of the SIBLING family members are dependent on the nature and extent of post-translational modifications (PTMs), such as phosphorylation, glycosylation, and proteolytic processing, since these PTMs would have significant effects on their structure. OPN and BSP are present in the ECM of bone and dentin as full-length forms, whereas amino acid sequencing indicates that DMP1 and DSPP exist as proteolytically processed fragments that result from scission of X-Asp bonds. We hypothesized that the processing of DMP1 and DSPP is catalyzed by the PHEX enzyme, since this protein, an endopeptidase that is predominantly expressed in bone and tooth, has a strong preference for cleavage at the NH₂-terminus of aspartyl residue. We envision that the proteolytic processing of DMP1 and DSPP may be an activation process that plays a significant, crucial role in osteogenesis and dentinogenesis, and that a failure in this processing would cause defective mineralization in bone and dentin, as observed in X-linked hypophosphatemic rickets.

Introduction

Bone and dentin are mineralized tissues that closely resemble each other in composition and mechanism of formation. Both are composed of organic matrices and mineral contents of plate-like apatite crystals. They are formed when osteoblasts and odontoblasts secrete type I collagen-rich unmineralized extracellular matrices (ECM), termed osteoid and predentin, respectively, which are subsequently mineralized when apatite crystals are deposited (Figs. 1a, 1b ). Formation of predentin and its transformation to dentin as mineralization takes place are highly controlled, orderly processes. The predentin width is rather uniform, indicating that the rates of synthesis of unmineralized matrix and its conversion to dentin at the predentin-dentin (PD) border are equal (Fig. 1b ). These considerations indicate that several unspecified, controlling mechanisms must be involved. The importance of the correct collagen structure is clearly seen in patients with osteogenesis imperfecta caused by mutations in type I collagen gene (Cole, 2002). In addition to type I collagen, the ECM of bone and dentin contains several non-collagenous proteins (NCPs). These NCPs are believed actively to promote and control mineralization of collagen fibers and crystal growth within osteoid and predentin, when these tissues are converted to bone and dentin. This belief is strongly supported by the fact that mutations in their genes (Xu et al., 1998; Xiao et al., 2001; Zhang et al., 2001; Feng et al., 2002a; Sreenath et al., 2003) are associated with dramatic phenotypic abnormalities in the mineralization process of bone and dentin. One category of the NCPs is termed the SIBLING (Small Integrin-Binding LIgand, N-linked Glycoprotein) family that includes osteopontin (OPN), bone sialoprotein (BSP), dentin matrix protein 1 (DMP1), dentin sialophosphoprotein (DSPP), and matrix extracellular phosphoglycoprotein (MEPE) (Fisher et al., 2001). Although the known sequences for SIBLING proteins have little similarity except in limited regions of the molecules, they share the following common features (Fisher et al., 2001). (1) SIBLING proteins are principally found in bone and dentin and are secreted into the ECM during the formation and mineralization of these tissues; the expression of BSP, OPN, DMP1, and DSPP is qualitatively similar but quantitatively different between the two tissues (Qin et al., 2001a). (2) They display an Arg-Gly-Asp (RGD) motif that mediates cell attachment/signaling via its reaction with cell-surface integrins. (3) Their gene organizations are similar, with the last one or two exons being the largest and containing the RGD tripeptide. (4) Their genes share common localization on human chromosome 4q21-23. (5) They share similarities in post-translational modifications (PTMs) such as phosphorylation and glycosylation. Some biological functions of these polyanionic SIBLING proteins are dependent on the nature and extent of the PTMs, including phosphorylation, glycosylation, proteolytic processing, sulphation, and transglutaminase cross-linking.

This brief review will provide information relative to the post-translational modifications of SIBLING proteins and the roles of these PTMs in osteogenesis and dentinogenesis. We are focusing mainly on OPN, BSP, DMP1, and DSPP, since little is known about the PTMs of MEPE, the newest member of this family that was discovered recently by cDNA cloning (Rowe et al., 2000). The PTMs of OPN and BSP have been recently discussed and summarized in several excellent reviews (Butler et al., 1996; Ganss et al., 1999; Sodek et al., 2000). For comparative purposes and for inclusion of newer data, we have included in this review a brief summary concerning these two proteins. Information relative to sulphation and transglutaminase cross-linking of OPN and BSP is excluded in this review, since no newer data have been reported since the publication of two former reviews (Ganss et al., 1999; Sodek et al., 2000). The naturally occurring, full-length forms of DMP1 and DSPP have not been isolated or characterized; these proteins appear as proteolytically processed fragments in the ECM of bone and dentin, probably cleaved by the same enzyme (Qin et al., 2003a). We have included a detailed discussion about the proteolytic processing of these two SIBLING members because the data are relatively new, and because we believe that this process may represent an important step that plays a significant role in osteogenesis and dentinogenesis.

For clarity, this review on the PTMs of OPN, BSP, DMP1, and DSPP has been organized in the following manner. A short general introduction (A) for each of them is given, followed by presentation of data relative to each type of PTMs, i.e., (B) phosphorylation, (C) glycosylation, (D) proteolytic processing. Finally, some ideas about potential future research (E) with regard to PTMs of the SIBLING proteins are included.

The Table is a summary of the PTMs for OPN, BSP, DMP1, and DSPP. The data presented in this review are based mainly on results obtained from rats when available. If the data from rats were unavailable, findings are from other species, such as cows and humans.

(A) General Introduction

(1) OPN

OPN was first described as a secreted 60-kDa transformation-specific phosphoprotein (Senger et al., 1979). Subsequently, it was independently discovered by several groups with different approaches and in different fields (Butler, 1989). The name ‘osteopontin’ was introduced to reflect the potential of this protein in bone to serve as a bridge between cells and hydroxyapatite through RGD and polyaspartic acid motifs in its primary amino acid sequence discovered by cDNA cloning (Oldberg et al., 1986). The secreted rat OPN contains 301 amino acids. Although OPN is present in bone in relatively large quantities, it is also expressed in a variety of tissues and cells (Butler et al., 1996; Sodek et al., 2000). The broad expression of OPN indicates a multiplicity of functions in diverse biological events (Butler et al., 1996; Giachelli and Steitz, 2000; O’Regan and Berman, 2000; Sodek et al., 2000; Denhardt et al., 2001). Some in vitro studies have shown that OPN is an effective inhibitor of apatite formation and growth (Boskey et al., 1993; Hunter et al., 1994; Goldberg and Hunter, 1995; Wada et al., 1999). Data from OPN null mice strengthen the conclusion that this protein may be a major inhibitory factor on mineralization (Boskey et al., 2002). Although there is a lack of a clear phenotype in OPN knockout mice, several studies have demonstrated that OPN deficiency affects the functions of osteoclasts (Rittling et al., 1998; Rittling and Denhardt, 1999; Ishijima et al., 2001; Suzuki et al., 2002; Chellaiah et al., 2003), consistent with our recent findings that OPN is involved in regulation of osteoclast activity (Razzouk et al., 2002).

(2) BSP

BSP was initially isolated from bovine cortical bone (Herring, 1972) and subsequently characterized by Fisher et al.(1983, 1987). The primary sequence of BSP was first deduced from a rat cDNA sequence (Oldberg et al., 1988). The secreted rat BSP contains 303 amino acids. Unlike OPN, the tissue distribution of BSP is restricted, being found almost exclusively in mineralized tissues including bone, dentin, mineralizing cartilage, and cementum (Oldberg et al., 1988; Bianco et al., 1991; Somerman et al., 1991; Chen et al., 1993). The quantity of BSP in bone is similar to that in dentin (Qin et al., 2001a). However, the biological functions of BSP in mineralized tissues are not known. Analysis of some data suggests that BSP acts as a nucleator of the initial apatite crystals; then, as this mineral grows on the collagen matrix, it acts as an inhibitor in directing the growth of the crystals (Hunter and Goldberg, 1994). BSP may also contribute to the activity of osteoclasts by interacting with cell-surface integrins (Ross et al., 1993; Razzouk et al., 2002).

(3) DMP1

DMP1 was first discovered by cDNA cloning with the use of a rat odontoblast mRNA library (George et al., 1993). The full-length DMP1 secreted by the rat consists of 473 amino acids, containing an unusually large number of acidic domains, a property consistent with DMP1 being a possible participant in regulating matrix mineralization. The roles of DMP1 in mineralization are supported by observations that transgenic MC3T3-E1 cells overexpressing DMP1 show earlier onset of mineralization and produce mineralized nodules of significantly larger sizes when compared with the non-transgenic cells (Narayanan et al., 2001). More recently, Narayanan et al.(2003) showed that, in addition to its role in the regulation of mineral formation, DMP1 may act as a transcriptional component for activation of osteoblast-specific genes like osteocalcin. The conclusion that DMP1 plays an important role in mineralization of bone and dentin is further strengthened by the manifestations in DMP1 knockout mice. Although these DMP1-deficient mice appear normal at birth, delayed conversion of osteoid to bone and predentin to dentin occurs at three weeks, and this condition grows more severe by three months (Feng et al., 2002a). Since native DMP1 has not been isolated and characterized, properties resulting from its PTMs have not been tested. Recently, we obtained clear evidence suggesting that DMP1 is proteolytically processed into 37K and 57K fragments, present in the ECM of bone and dentin (Qin et al., 2003a), as discussed later. DMP1 was originally postulated to be dentin-specific, but later, its expression was observed in bone and brain (Hirst et al., 1997; MacDougall et al., 1998).

(4) DSPP

This precursor protein, predicted from the cloning of a cDNA (MacDougall et al., 1997; Gu et al., 2000; Ritchie et al., 2001; Yamakoshi et al., 2003), gives rise to two proteins, dentin sialoprotein (DSP) with coding sequences in the 5′ end and dentin phosphoprotein (DPP) representing the 3′ end of cDNA. The two proteins were independently identified as ECM components extracted from dentin when it is demineralized. DPP, discovered by Veis and his colleagues (Veis and Perry, 1967; Dickson et al., 1975; DiMuzio and Veis, 1978), is the most abundant NCP in the ECM of dentin. It contains large amounts of aspartic acid (Asp) and phosphoserine (Pse), largely present in repeating sequences of (Asp-Pse-Pse)_n and (Asp-Pse)_n. A sizeable data pool indicates that DPP is an important initiator and modulator of dentin apatite crystal formation (Linde et al., 1989; Boskey et al., 1990; George et al., 1996; Saito et al., 1997; Butler, 1998). DSP, discovered by Butler and his colleagues (Butler et al., 1981, 1992; Butler, 1987), is the second most abundant NCP in dentin ECM. The functions of DSP or a recently discovered isoform of DSP, HMW-DSP, that is present in dentin ECM (Qin et al., 2003b) are not known. The occurrence of one gene transcribing a single mRNA encoding DSP and DPP indicates that the translated product is proteolytically processed into DSP and DPP by unidentified proteinases. Analysis of recent data suggests that the proteolytic processing of rat DSPP to DSP and DPP may involve the same proteinase that cleaves DMP1 (Qin et al., 2003a; see below). The importance of DSPP in dentinogenesis is supported by experiments showing the association of mutations in the Dspp gene with dentinogenesis imperfecta in humans (Xiao et al., 2001; Zhang et al., 2001) and by defective mineralization of dentin in DSPP knockout mice (Sreenath et al., 2003). Originally, the expression of DSPP was thought to be tooth-specific, but recent studies have demonstrated its expression in bone and osteoblasts at a much lower level than in the tooth (Qin et al., 2002, 2003c).

(B) Phosphorylation

(1) OPN

Phosphorylation of OPN occurs mainly on serine residues and is principally catalyzed by casein kinases (Lasa et al., 1997). The amino acid sequence alignment from seven species indicates that OPN has a high level of serine conservation and that about half of the conserved serine residues are potential phosphorylation sites for casein kinases I and II (Butler et al., 1996). Rat bone OPN has an average of 12 phosphoserines and 1 phosphothreonine (Prince et al., 1987), with considerable heterogeneity (Neame and Butler, 1996), whereas 27 phosphorylated serines and one phosphorylated threonine have been identified in bovine milk OPN (Sorensen et al., 1995). Notably, different forms of OPN that vary in their phosphorylation have been reported in transformed osteogenic cells (Kasugai et al., 1991; Safran et al., 1998a). The level of phosphorylation of OPN was shown to be modulated by 1,25-dihydroxyvitamin D3 in a non-transcriptional mechanism (Safran et al., 1998a). The addition of this hormone to cultures of ROS17/2.8 cells lowers the level of OPN phosphorylation prior to transcriptional up-regulation by 1,25-dihydroxyvitamin D3.

Although data are available showing that phosphorylation of OPN is important for its function, the exact effect of phosphates on this protein is unknown. In vitro mineralization studies with a gelatin gel diffusion system used for the study of its effects on the formation and growth of hydroxyapatite have shown that OPN is a very effective inhibitor of these processes, but this activity was lost after removal of phosphate groups (Boskey et al., 1993). Likewise, the phosphorylation of OPN has been shown to be necessary for the inhibitory effect of this protein on smooth-muscle-cell calcification, since native (phosphorylated) OPN inhibits calcium deposition in these cells (Wada et al., 1999), whereas the dephosphorylated and the bacteria-derived recombinant OPN display no inhibitory effect (Jono et al., 2000). These results indicate that phosphorylation of OPN is essential for its inhibitory role on mineralization. Two studies have suggested that phosphorylation is necessary for OPN-stimulated osteoclast attachment (Ek-Rylander et al., 1994; Katayama et al., 1998). More recently, experiments performed with primary osteoclast cultures on bone slices have demonstrated that phosphates on OPN may be involved in regulating the resorptive activity of osteoclasts (Razzouk et al., 2002).

(2) BSP

Like OPN, phosphorylation in BSP occurs mostly at serine residues (Franzén and Heinegård, 1985). The cDNA sequence indicates that rat BSP has 10 serines and 8 threonine residues that are potential phosphorylation sites for casein kinases I and II (Oldberg et al., 1988). Based on consensus sequences in the primary structure of mammalian BSP, there are 5 potential sites for serine phosphorylation by protein kinase C, 9-11 sites for Ser/Thr phosphorylation by casein kinase II, and 1–2 sites for Tyr phosphorylation by tyrosine kinase (Ganss et al., 1999; Fig. 2). A recent study carried out on bovine BSP indicated that 5.8 phosphates are incorporated into the protein, predominantly by casein kinase II but not by casein kinase I (Salih, 2003). A detailed study using mass spectrometry combined with dephosphorylation indicated that human BSP contains 5 phosphates, 4 of which are located in the central region of its primary sequence (Zaia et al., 2001). The functions of phosphate groups of BSP are unknown; removal of the phosphates has no apparent influence on the ability of BSP to promote crystal formation, but they may affect crystal growth (Hunter and Goldberg, 1994).

(3) DMP1, 37K and 57K Fragments

The cDNA-deduced sequence indicates that full-length rat DMP1 has 65 Ser/Thr that are potential casein kinases I and II phosphorylation sites (George et al., 1993), with the NH₂-terminal (37K) portion containing 22 and the COOH-terminal (57K) region having 43. Phosphate analysis indicates that the 37K fragments have 12 phosphates/mol, while the 57K fragments have 41 (Qin et al., 2003a). Although the functions of phosphates on DMP1 are unknown, they may serve as sequestering groups for recruiting calcium ions, similar to the roles of phosphates on DPP (see below).

(4) DSPP, DSP, and DPP Fragments

Rat dentin DSP has 6.2 phosphates/mol, about half of the 13 Ser/Thr potential casein kinases I and II phosphorylation sites (Qin et al., 2003b). DSP also shows heterogeneity in phosphorylation (Qin et al., 2001b). HMW-DSP has 10.3 phosphates/mol (Qin et al., 2003b). The functions of phosphates relative to DSP are unknown.

DPP, the most abundant NCP in dentin, is highly phosphorylated: About 45% of the total sequence is serine residues, and most of them are phosphorylated (Richardson et al., 1978; Butler et al., 1983). Rat dentin DPP is present in three isoforms—HP, MP, and LP—with different levels of phosphorylation (Butler et al., 1983; Butler, 1998). HP, the most highly phosphorylated and also the most abundant form of DPP, has 209 phosphates/mol (Butler et al., 1983). DPP contains a large number of repeating sequences of (Asp-Pse-Pse) and (Asp-Pse). The predicted structures for these repeating sequences depict extended backbone structures, with long ridges of carboxylate and phosphate groups alternating on opposite sides of the stretched-out polypeptide chain (George et al., 1996). Assuming that such a structure is formed on a collagen matrix, one can visualize an orientation that structurally fits its purported function, with calcium phosphate crystals growing on this lattice (Butler, 1998). Several in vitro studies have consistently demonstrated that immobilized DPP initiates mineralization and interacts with growing apatite crystals, thereby affecting their growth (Linde et al., 1989; Boskey et al., 1990; Saito et al., 1997). The large number of phosphates is vital to the functions of DPP, since removal of phosphate groups results in the loss of its role in the nucleation of apatite onto collagen (Saito et al., 1997).

(C) Glycosylation

(1) OPN

Significant glycosylation occurs on OPN in most tissues (Sodek et al., 2000). The cDNA sequence indicates that rat OPN has one Asn-X-Ser/Thr motif that is a potential site for N-linked glycosylation and ~ 26 Ser-X-Glu sequences that are potential sites for O-linked glycosylation (Oldberg et al., 1986; Sodek et al., 2000). In rat bone OPN, a single N-linked and 5 or 6 O-linked oligosaccharides have been identified, which consist of about 16.6% of the protein weight (Prince et al. 1987). In bovine milk OPN, 3 O-linked oligosaccharides have been identified (Sorensen et al., 1995). Although OPN has a considerable amount of carbohydrates, their precise functions are unclear.

(2) BSP

The rat BSP sequence contains 3 Asn-X-Ser/Thr motifs that are putative sites for N-linked glycosylation (Oldberg et al., 1988), 2 of which are conserved in all of the mammalian species (Fisher et al., 1990; Ganss et al., 1999). It has been estimated that BSP synthesized by UMR 106-01 cells (a rat osteosarcoma cell line) has 3 N-linked and 21-24 O-linked oligosaccharides (Midura et al., 1990). A recent study with mass spectrometry combined with a series of deglycosylation procedures has indicated that human BSP contains 33.8% oligosaccharides, with 12.3% being N-linked and 21.5% O-linked (Zaia et al., 2001), while another study with similar approaches demonstrated that oligosaccharides made a mass contribution of 30% to BSP extracted from human bone (Wuttke et al., 2001). Although the precise roles of attached carbohydrates are unknown, the extensive glycosylation of BSP may affect cell attachment activity (Fujisawa et al., 1995), but does not seem to be important for nucleation and the formation of apatite (Hunter and Goldberg, 1994; Goldberg et al., 1996).

(3) DMP1, 37K, and 57K fragments

The cDNA-deduced sequence indicates that a full-length rat DMP1 has 3 putative Asn-X-Ser/Thr N-linked glycosylation sites; all are located in the 57K fragment portion, COOH-terminal to the RGD motif (George et al., 1993). However, information with regard to DMP1 glycosylation is unavailable. In the later, high [Cl-] region of the DEAE chromatogram of rat bone and dentin ECM extract, a series of protein bands with Mr of 150–200 kDa contains components that are immunoreactive to anti-57K antibodies (Qin et al., 2001a). The identity of these DMP1-related, high-molecular-weight components is unknown. They may be highly glycosylated forms of DMP1 fragments.

(4) DSPP, DSP, and DPP fragments

The cDNA-deduced sequence indicates that rat DSP has 6 Asn-X-Ser/Thr potential sites for N-linked glycosylation (Ritchie et al., 1994). DSP isolated from rat dentin has 29.6% carbohydrates that include sialic acids and N- and O-linked oligosaccharides (Butler et al., 1992; Qin et al., 2003b). The functions of carbohydrates on DSP are unclear. HMW-DSP is larger and more acidic than DSP, and deglycosylation analysis indicates that the slower migration rate of HMW-DSP on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) is due to its higher level of carbohydrate modifications (Qin et al., 2003b). The unusually slow migration rate and extreme acidity of HMW-DSP indicate that it may contain GAG chains and/or considerable amounts of sulphate, in addition to oligosaccharides. However, the exact nature of the carbohydrate compositions and their structures, as well as biological functions, is unknown. It has been demonstrated that the ~ 100-kDa DSP has little or no effect on in vitro mineralization (Boskey et al., 2000). Perhaps the HMW-DSP with extremely large amounts of carbohydrates may fulfill some special function different from that of the ~ 100-kDa form of DSP.

Although the cDNA-deduced sequence indicates that the DPP region of DSPP has some potential Asn-X-Ser/Thr N-linked glycosylation sites (MacDougall et al., 1997), the lack of detection of glucosamine and galactosamine in hydrolysates on the amino acid analyzer indicates that little or no carbohydrate is present on rat dentin DPP (Butler, unpublished observations).

(D) Proteolytic Processing

(1) OPN

Two fragments migrating at 20 kDa and 23 kDa on 15% SDS-PAGE have been reported for OPN isolated from porcine bone (Zhang et al., 1990). The NH₂-terminal sequence analysis revealed that these 20- and 23-kDa proteins were derived from the COOH-terminal end of OPN and appeared to be generated by trypsin-like cleavage at Arg-Ser and Lys-Ala bonds. The rapid generation of the 23-kDa fragment, and its presence in different bone tissues at different developmental stages suggested that the fragmentation of porcine OPN might be important in bone formation (Zhang et al., 1990). A 14-kDa protein from fetal rat calvarial cell cultures, although not reactive with anti-OPN antibodies, was thought to be a fragment of OPN (Nagata et al., 1991). This conclusion was primarily based on its susceptibility to thrombin cleavage (Nagata et al., 1991). However, since the initial description, no further information has been reported concerning the 14-kDa fragment. OPN isolated from rat long bone displays Mr of ~ 65 to ~ 75 kDa on 5–15% gradient SDS-PAGE (Prince et al., 1987; Qin et al., 2001a), while on 15% SDS-PAGE, it migrates at ~ 45 kDa (Prince et al., 1987). It is worth noting that rat bone OPN with different levels of phosphorylation displays variations in migration rate (Kasugai et al., 1991; Razzouk et al., 2002). No fragmentation has been noticed for rat bone OPN; with Western immunoblots, only the ~ 65- to ~ 75-kDa bands on 5–15% SDS-PAGE were immunoreactive to the highly specific anti-OPN antibodies, without any trace of lower-molecular-weight fragments (Qin et al., 2001a). Although it is unclear why fragmentation was observed in extracts of porcine but not rat bone, these variations could be due to species differences. It is worth noting that the 20- and 23-kDa fragments were much less abundant than the ‘full-length’ porcine OPN migrating at 67 kDa on 15% SDS-PAGE (Zhang et al., 1990).

(2) BSP

There has been no information about the proteolytic processing of BSP. The 23-kDa bovine bone sialoprotein originally described by Herring (1972) was later considered to be a fragment of BSP produced by proteolytic activity during isolation (Fisher et al., 1983).

(3) DMP1

Full-length cDNA coding for DMP1 mRNA has been cloned and sequenced in several mammalian and non-mammalian species, but the corresponding complete protein has not been isolated or characterized. Originally, a 61-kDa protein band from dentin extracts was recognized by immunoblotting with an antibody monospecific to recombinant rat DMP1; this 61-kDa protein was thought to be a full-length DMP1 (George et al., 1993). Later, the same group reported that the molecular weight of the bacteria-derived recombinant rat DMP1, determined by SDS-PAGE, was approximately 90–95 kDa, which is higher than that of the purported naturally occurring 61-kDa form of DMP1 (Srinivasan et al., 1999). We now believe that this 61-kDa band is a processed fragment(s) of DMP1, possibly equivalent to the 57K fragment (Qin et al., 2003a). In bone and dentin ECM extracts, there are two clusters of proteins (37K and 57K fragments) that elute in the same (D4b) fraction of the DEAE-Sephacel chromatogram as OPN and BSP (Qin et al., 2001a, 2003a). When the mixture from this fraction of bone was cleaved with trypsin, sequences originating from DMP1 (as well as OPN and BSP) were obtained (Safran et al., 1998b). With a combination of several chromatographic approaches, these rat bone DMP1 fragments were purified. Extensive sequencing and analysis of tryptic peptides derived from these highly pure materials (37K and 57K fragments, respectively) by Edman degradation and mass spectrometry have yielded some significant discoveries (Qin et al., 2003a). First, analysis of these data shows unequivocally that 37K fragments originate from the NH₂-terminal region, and that the 57K fragments are from the COOH-terminal region of the DMP1 amino acid sequence, strongly suggesting that DMP1 in bone is proteolytically processed (Fig. 2a).

Second, from 37K fragments, 2 COOH-termini, Phe¹⁷³ and Ser¹⁸⁰, were identified, and from 57K fragments, 2 NH₂-termini, Asp²¹⁸ and Asp²²², were detected. These findings, along with comparison with DMP1 sequence deduced from cDNA, indicate that this protein is proteolytically cleaved at 4 bonds, Phe¹⁷³-Asp¹⁷⁴, Ser¹⁸⁰-Asp¹⁸¹, Ser²¹⁷-Asp²¹⁸, and Gln²²¹-Asp²²² (Fig. 2b ). It should be noted that the sequences around the cleaved Phe¹⁷³-Asp¹⁷⁴ and Ser¹⁸⁰-Asp¹⁸¹ bonds are highly conserved, but those surrounding Ser²¹⁷-Asp²¹⁸ and Gln²²¹-Asp²²² are not conserved.

Third, the uniformity of cleavages at the NH₂-terminal peptide bonds of select aspartyl residues suggests that a single proteinase is involved (see below). Fourth, phosphate analysis indicates that the 57K fragments are more highly phosphorylated than the 37K. Additionally, dentin ECM also contains 37K and 57K fragments. The HPLC chromatographic profile of tryptic peptides derived from dentin 37K fragments was identical to that obtained from bone 37K fragments, while selective sequencing of tryptic peptides indicated that the NH₂-termini and COOH-termini of dentin 37K fragments were identical to those of bone (Qin et al., 2003a). Attempts to obtain pure dentin 57K fragments for sequence analysis have thus far been unsuccessful.

(4) DSPP

The two acidic proteins, DSP and DPP, are individual, distinctive proteins that are found in abundance in the ECM of dentin (Dickson et al., 1975; DiMuzio and Veis, 1978; Butler, 1987; Butler et al., 1992). It appeared that they had no relationship to each other, an idea that changed when results from cDNA cloning and sequencing of a mouse mRNA library showed that DSP and DPP are derived from the same precursor, DSPP (MacDougall et al., 1997). Subsequently, results from cDNA cloning and sequencing in human, rat, and porcine samples consistently demonstrated that the 5′ end of DSPP mRNA contains the sequences coding for DSP, followed by the DPP sequences on the 3′ end, in agreement with results obtained from cDNA cloning in the mouse (Gu et al., 2000; Ritchie et al., 2001; Yamakoshi et al., 2003). The occurrence of one gene transcribing a single mRNA encoding DSP and DPP indicates that the translated product, DSPP, is cleaved by a specific proteinase(s), giving rise to the individual proteins, DSP and DPP (Fig. 2c). This conclusion is supported by the fact that DSP and DPP, but not DSPP (the protein representing the entire sequence), are found in dentin ECM. The presence of HMW-DSP (Qin et al., 2003b) may be one of the reasons why there is more DPP than DSP in dentin ECM. To clarify proteolytic cleavage sites, we have isolated and characterized tryptic peptides from the COOH-terminal region of rat DSP (Qin et al., 2001b). Two COOH-termini were identified and characterized. The major peptide ended at Tyr⁴²¹ and the minor one at His⁴⁰⁶. Analysis of these data, along with comparison with the DSPP sequence deduced from cDNA and the established NH₂-terminus of DPP, shows that the proteolytic processing of rat DSPP to form DSP and DPP involves a proteinase(s) that cleaves Tyr⁴²¹-Asp⁴²² (giving rise to the COOH-terminus of DSP) and Gly⁴³⁰-Asp⁴³¹ (the NH₂-terminus of DPP) and, to lesser extent, His⁴⁰⁶-Ser⁴⁰⁷ (a second, minor COOH-terminus). The fact that the two major cleavages occur at conserved sequences at the NH₂-terminus of an aspartic acid residue (Fig. 2b) indicates that DSPP may be cleaved by the same proteinase involved in the proteolytic processing of DMP1 (possibly PHEX protein; see below).

In summary, DMP1 and DSPP are proteolytically processed, and they appear to be processed similarly. We hypothesize that full-length DMP1 and DSPP are inactive precursors and that the proteolytic processing is an activation step, releasing their functional fragments which play significant roles in the mineralization of bone and dentin. This view predicts that if these precursors are not properly processed, defective mineralization will result (see below).

(5) PHEX protein, a possible proteinase that cleaves DMP1 and DSPP

This protein, encoded by PHEX (phosphate-regulating gene with homologies to endopeptidases on X chromosome), is expressed predominantly in osteoblasts, osteocytes, and odontoblasts (Du et al., 1996; Guo and Quarles, 1997; Lipman et al., 1998; Ruchon et al., 1998, 2000; Thompson et al., 2002). It is a cell-membrane-associated glycoprotein with structural homologies to members of the M13 family of metalloendopeptidases (The HYP Consortium, 1995; Lipman et al., 1998; Boileau et al., 2001). The M13 endopeptidases are zinc-containing type II integral membrane glycoproteins with a relatively short cytoplasmic NH₂-terminal region, a single transmembrane domain, and a long extracytoplasmic domain, which contains the active site of the enzyme (Crine et al., 1997). Mutations of PHEX are associated with X-linked hypophosphatemic rickets (XLH) in humans, as well as with the murine homolog, Hyp (The HYP Consortium, 1995; Holm et al., 1997).

The PHEX protein is likely to be the enzyme responsible for the proteolytic processing of DMP1 and DSPP (Qin et al., 2003a). This conclusion is based on the following observations: (a) Several studies have shown that the PHEX protein has a strong preference for cleavage of peptide bonds at the NH₂-terminus of aspartyl residues (Lipman et al., 1998; Boileau et al., 2001; Sabbagh et al., 2003; Campos et al., 2003); as described above, DMP1 and DSPP are cleaved similarly (see Fig. 2b). (b) PHEX is expressed in osteoblasts, osteocytes, and odontoblasts (Du et al., 1996; Guo and Quarles, 1997; Lipman et al., 1998; Ruchon et al., 1998, 2000; Thompson et al., 2002), as are DMP1 (Butler et al., 2002; Feng et al., 2002b) and DSPP (Butler et al., 1992; D’Souza et al., 1992; Bègue-Kirn et al., 1998; Qin et al., 2002, 2003c). (c) The bony and dental defects of DMP1 knock-out mice at post-natal stage (after the age of 3 wks; Feng et al., 2002a) are similar to those of XLH and Hyp, while the dental defects in the DSPP null mice (Sreenath et al., 2003) show similarities to those of XLH and Hyp.

A more recent study (Steiglitz et al., 2004) described cleavage of recombinant full-length DMP1 by bone morphogenetic protein-1(BMP-1)/Tolloid-like proteinases that are widely expressed in mesenchymal tissues and that have broad substrate specificity. Two DMP1 fragments generated by the BMP-1/Tolloid-like proteinases are similar to those previously isolated from bone (Qin et al., 2003a). Based on the known roles of BMP-1/Tolloid-like proteinases in the biosynthetic conversion of precursor proteins into their mature functional forms, it is believed that proteolytic processing of DMP1 is an activation step, i.e., converting full-length DMP1 precursor into functional fragments. Although recombinant DMP1 was shown to be cleaved by BMP-1/Tolloid-like proteinases in vitro, it was stated that the possibility that PHEX is involved in DMP1 processing is not excluded (Steiglitz et al., 2003). It should be noted that the expression of BMP-1/Tolloid-like proteinases begins at a rather earlier stage of embryonic development, whereas DMP1 is expressed in a much later stage (Feng et al., 2002b) within the same time window of PHEX expression (Thompson et al., 2002). Additionally, BMP-1/Tolloid-like proteinases cleaved only one (Ser¹⁸⁰-Asp¹⁸¹) of the 4 cleavage sites that have been identified in rat bone DMP1 processing (Qin et al., 2003a). A combination of information (also see below) seems to suggest that PHEX protein is the major proteinase responsible for processing DMP1, functioning synergistically with BMP1/Tolloid-like proteinases in the proteolytic processing of DMP1 precursors.

(6) Relationship of XLH and Hyp to DMP1 and DSPP

XLH, the most common inherited form of rickets, is a dominant disorder characterized by defective calcification of bone and dentin, growth retardation, impaired renal tubular re-absorption of inorganic phosphate (P_i), and resistance to phosphorus and vitamin D therapy (Winters et al., 1958; Burnett et al., 1964; Rasmussen and Anast, 1983; The HYP Consortium, 1995; Grieff et al., 1997; Holm et al., 1997). As stated earlier, mutations of PHEX are associated with XLH in humans and the murine homolog, Hyp. PHEX encodes for a cell-surface metalloprotease. Since the physiologically relevant PHEX substrates have not been identified (Guo et al., 2001; Quarles, 2003), several different mechanisms have been considered to explain how inactivation of an enzyme results in both phosphaturia and defective mineralization. Currently, one widely discussed theory is that hypophosphatemia (phosphaturia) and defective mineralization are caused through different pathogenetic pathways (Bai et al., 2002; Liu et al., 2002; Quarles, 2003). Because the expression of PHEX is predominantly in bone and tooth tissues, and it is expressed by osteoblasts, osteocytes, and odontoblasts, locally produced substrates in the skeleton are thought to be involved in the defective mineralization of XLH (Xiao et al., 1998; Bai et al., 2002; Quarles, 2003). Since PHEX is also expressed less abundantly at extraskeletal sites, additional substrates in these non-mineralized tissues may be involved in the pathogenesis of hypophosphatemia (Quarles, 2003). In this review, we will discuss only possible mechanisms relative to the defective mineralization in XLH and Hyp.

Several lines of evidence support the view of intrinsic skeletal factor(s) responsible for impaired mineralization, independent of the hypophosphatemia condition: First, the bone defect in Hyp mice is more severe than can be explained by hypophosphatemia alone (Quarles, 2003). Second, transplantation of osteoblasts from Hyp mice into normal animals produces abnormal bone with impaired mineralization but with the absence of hypophosphatemia (Ecarot et al., 1992). Third, osteoblasts derived from the Hyp mouse have an impaired capacity to mineralize ECM in vitro; compared with cells from normal mice, osteoblast cultures from the Hyp mouse display diminished ⁴⁵Ca accumulation into the ECM and reduced formation of mineralization nodules (Xiao et al., 1998). Fourth, deletion of the renal sodium-dependent phosphate transporter in Npt-2 null mice displays both hypophosphatemia and abnormal skeletal development at 21 days of age, but with increasing age, there is a dramatic reversal and eventual overcompensation of the skeletal phenotype without ‘normalization’ of hypophosphatemia (Beck et al., 1998). The clearly age-dependent pattern of skeletal changes in Npt-2 null mice suggests that adaptive responses occur in osteoblasts, when PHEX is active, to minimize the effects of hypophosphatemia on bone (Quarles, 2003).

In summary, it is likely that protein substrates for the PHEX enzyme produced by osteoblasts, osteocytes, and odontoblasts are involved in the normal mineralization of bone and dentin. Analysis of amino acid sequencing data establishing cleavages at X-Asp bonds strongly suggests that DMP1 and DSPP are proteolytically processed by the PHEX protein, since this enzyme has a strong preference for these bonds. This endopeptidase, expressed predominantly in bones and teeth, is an integral membrane protein with a relatively large extracytoplasmic domain containing the catalytic site. This type of structure would imply that the proteolytic processing would occur during or after secretion. In this way, if the cleavage is an activation step, one would visualize activation in a controlled, timely manner, as part of a stepwise series of events. In the cases of DMP1 and DSPP, the conversion to 37K and 57K fragments and to DSP and DPP releases components that promote mineralization of osteoid and predentin (Fig. 3). Thus, we envision that mineralization of bone and dentin requires that DMP1 and DSPP be proteolytically processed by PHEX protein and that if these precursors are improperly processed, defective mineralization in XLH and Hyp results. This view regarding the proteolytic processing of DMP1 and DSPP by the PHEX protein resembles the activation of matrix metalloproteinase precursors (proMMPs) by membrane type MMPs (MT-MMPs). Through pericellular proteolysis, MT-MMPs process proMMPs, and the activated MMPs are secreted into the ECM, where they act on substrates remote from the producer cells (Seiki and Yana, 2003). For example, MT1-MMP has been identified as possessing the ability to activate proMMP-2 (Sato et al., 1994), the activated form of which is secreted into the ECM, where it plays an important role in the degradation of type IV collagen.

(E) Future Research

This review shows that PTMs of SIBLING family members may play vital roles in their functions. Despite excellent progress in this field, this review of the literature indicates that several areas relative to the PTMs of SIBLING proteins need further investigations. The roles potentially played by the carbohydrate moieties have not been studied. The nature of glycosylation bringing about heterogeneity of DSP and other SIBLING proteins warrants further studies. Phosphorylation of OPN clearly plays a role in the inhibition of mineralization and in osteoclast functions, but their exact roles have not been elucidated. This specific function of phosphorylation is accentuated by the realization that the seryl residues that are phosphorylated are highly conserved. The importance of post-translational proteolytic cleavage of DMP1 and DSPP must be clarified. The strong possibility that PHEX protein catalyzes this cleavage must be examined by a series of careful experiments. The hypothesis that this proteolytic cleavage of X-Asp bonds is an activation process needs further testing. Other proteins with X-Asp cleavage sites should be sought. One major objective in all the investigations will be to uncover the basic mechanistic steps in the mineralization process. The questions surrounding the unknown pathways that lead to the conversion of osteoid to bone (Fig. 1a) and predentin to dentin (Fig. 1b ) must be uncovered. These answers will lead to better methods for the diagnosis of malformations of these tissues as well as interventions that are scientifically based.

TABLE

PTMs ^a of OPN, BSP, DSPP ^b , and DMP1 ^c

	Phosphate (Pi/mol)	CHO Contents^d	N-Gly^e	O-Gly^f	SA^g	TGCL^h	Sulphation
^a Data presented here are from rats, except those marked with "i" (bovine).
^b DSPP is processed into DSP, HMW-DSP, and DPP.
^c DMPI is processed into 37K and 57K fragments.
^d Carbohydrate contents expressed as percent by weight if available.
^e N-linked glycosylation.
^f O-linked glycosylation.
^g Sialic acid contents expressed as percent by weight if available.
^h Transglutaminase cross-link; + = detected.
ⁱ Data from bovine sources are presented when those of rats are unavailable; ? = unknown.
^j Reference numbers are written as superscripts: ¹ Prince et al., 1987; ² Prince et al., 1991; ³ Kaartinen et al., 2002; ⁴ Nagata et al., 1989; ⁵ⁱ Salih, 2003; ⁶ⁱ Fisher et al., 1983; ⁷ Midura et al., 1990; ⁸ Qin et al., 2003b; ⁹ Butler et al., 1992; ¹⁰ Butler et al., 1983; ¹¹ Qin et al., 2003a.
OPN	13^j1	16.6%¹	+¹	+¹	7.3%¹	+^2,3	+⁴
BSP	5.85ⁱ	50%6ⁱ	+^5,7	+^5,7	12%⁶ⁱ	+³	+⁷
DSP	6⁸	29.5%⁹	+^8,9	+^8,9	9%⁹	?	?
HMW-DSP⁸	10⁸	+⁸	+⁸	+⁸	+⁸	?	?
DPP	HP: ~ 209¹⁰	?	?	?	?	?	?
37K¹¹	12¹¹	?	?	?	?	?	?
57K¹¹	41¹¹	?	?	?	?	?	?

Figure 1a.

H&E staining of a bucco-lingual section from the mandible of a three-week-old rat. Osteoblasts (ob) secrete unmineralized bone matrix, termed ‘osteoid’ (os, doubleheaded arrow), which is subsequently mineralized and becomes bone (mm, long arrow) when apatite crystals are deposited. Small arrows indicate osteocytes buried in the mineralized bone matrix. Bar = 40 μm.

Figure 1b.

H&E staining of a bucco-lingual section from a mandibular incisor of a five-week-old rat. Similar to osteoblasts, odontoblasts (od) secrete unmineralized predentin (pd, doubleheaded arrow), which is subsequently mineralized and becomes dentin (d, long arrow) when apatite crystals are deposited. The long columnar odontoblasts form processes that extend into dentin. The short arrows indicate dental tubules that house odontoblast processes. Bar = 40 μm.

Figure 2a.

Full-length DMP1 is an inactive precursor. It is proteolytically processed into 37K and 57K fragments, originating from the NH₂-terminal and COOH-terminal regions of the DMP1 amino acid sequence, respectively. The 57K fragments contain more phosphates (P) than the 37K fragments.

Figure 2b.

Amino acid sequences surrounding the cleavage sites of DMP1 and DSPP. Vertical arrows indicate cleavages sites. In DMP1, all 4 cleavages occur at the NH₂-terminus of the Asp residue (Qin et al., 2003a). Among the 3 cleavage sites in DSPP, 2 major ones are at the NH₂-terminus of the Asp residue, while the minor one, as indicated by a smaller arrow, is at the His-Ser bond (Qin et al., 2001b). Amino acid residues are numbered (written on the top of corresponding residues), starting from the NH₂-terminus of full-length DMP1 or DSPP, excluding signal peptide.

Figure 2c.

Full-length DSPP is an inactive precursor. It is proteolytically processed into DSP and DPP that are from the NH₂-terminal and COOH-terminal regions of DSPP, respectively. DSP is rich in carbohydrates (CHO) but contains fewer phosphates (P), whereas DPP is devoid of glycosylation but contains an unusually large number of phosphates. The phosphates on DPP are mainly present on one side of the protein backbone (George et al., 1996) and are postulated to sequester calcium ions (Ca⁺⁺).

Figure 3.

Hypothetical processing of DMP1 and DSPP by PHEX protein. DMP1 and DSPP are converted to functional fragments by PHEX enzyme on the surfaces of osteoblasts, osteocytes, and odontoblasts. PHEX protein, one of the M13 endopeptidases, has a short cytoplamic NH₂-terminal region, a single transmembrane domain, and a long extracytoplasmic domain that contains the active site of the enzyme (Crine et al., 1997). After DMP1 and DSPP are processed by cleavage of selected X-Asp bonds, their fragments—namely, 37K, 57K, DSP, and DPP—are transported to the ECM of bone and dentin, where they influence mineralization.

Footnotes

Acknowledgements

We are grateful to our colleague, Jan Brunn, for his excellent work relative to the purification of OPN, BSP, DSP, DPP, and DMP1 fragments from bone and dentin. Much of our data about the PTMs of SIBLING members has been obtained through analyses on the proteins he purified. We thank Richard Cook and Ralph Orkiszewski at the Protein Chemistry Core Laboratory, Baylor College of Medicine, for their continuous, wonderful collaboration in analyzing peptide samples of OPN, DSP, and DMP1 fragments. We are also thankful to Dr. Adele Boskey of the Hospital for Special Surgery, New York, for her continuous collaboration in testing the effects of SIBLING proteins on in vitro mineralization. We deeply appreciate consultations with Dr. Darryl Quarles at the Center for Bone and Mineral Disorders of Duke University Medical Center for giving us insights about PHEX. This work was supported by NIH Grant DE 05092 to William T. Butler.

References

Bai X, Miao D, Panda D, Grady S, McKee MD, Goltzman D, et al. (2002). Partial rescue of the Hyp phenotype by osteoblast-targeted PHEX (phosphate-regulating gene with homologies to endopeptidases on the X chromosome) expression. Mol Endocrinol 16:2913–2925.

Beck L, Karaplis AC, Amizuka N, Hewson AS, Ozawa H, Tenenhouse HS (1998). Targeted inactivation of Npt2 in mice leads to severe renal phosphate wasting, hypercalciuria and skeletal abnormalities. Proc Natl Acad Sci USA 95:5372–5377.

Bègue-Kirn C, Ruch JV, Ridall AL, Butler WT (1998). Comparative analysis of mouse DSP and DPP expression in odontoblasts, preameloblasts, and experimentally induced odontoblast-like cells. Eur J Oral Sci 106(Suppl 1):254–259.

Bianco P, Fisher LW, Young MF, Termine JD, Robey PG (1991). Expression of bone sialoprotein (BSP) in developing human tissues. Calcif Tissue Int 49:421–426.

Boileau G, Tenenhouse HS, DesGroseillers L, Crine P (2001). Characterization of PHEX endopeptidase catalytic activity: identification of parathyroid-hormone-related peptide 107–139 as a substrate and osteocalcin, PPi and phosphate as inhibitors. Biochem J 355:707–713.

Boskey AL, Maresca M, Doty S, Sabsay B, Veis A (1990). Concentration-dependent effects of dentin phosphophoryn in the regulation of in vitro hydroxyapatite formation and growth. Bone Miner 11:55–65.

Boskey AL, Maresca M, Ullrich W, Doty SB, Butler WT, Prince CW (1993). Osteopontin-hydroxyapatite interactions in vitro: inhibition of hydroxyapatite formation and growth in a gelatin-gel. Bone Miner 22:147–159.

Boskey A, Spevak L, Tan M, Doty SB, Butler WT (2000). Dentin sialoprotein (DSP) has limited effects on in vitro apatite formation and growth. Calcif Tissue Int 67:472–478.

Boskey AL, Spevak L, Paschalis E, Doty SB, McKee MD (2002). Osteopontin deficiency increases mineral content and mineral crystallinity in mouse bone. Calcif Tissue Int 71:145–154.

10.

Burnett CH, Dent CE, Harper C, Warland BJ (1964). Vitamin D-resistant rickets: analysis of twenty-four pedigrees with hereditary and sporadic cases. Am J Med 36:222–232.

11.

Butler WT (1987). Dentin specific proteins. Methods Enzymol 145:290–303.

12.

Butler WT (1989). The nature and significance of osteopontin. Connect Tissue Res 23:123–136.

13.

Butler WT (1998). Dentin matrix proteins. Eur J Oral Sci 106(Suppl 1):204–210.

14.

Butler WT, Bhown M, Tomana M, Fretwell B, Schrohenloher RE (1981). Characterization of a unique dentin glycoprotein. In: The chemistry and biology of mineralized connective tissues. Veis A, editor. New York: Elsevier, pp. 399–402.

15.

Butler WT, Bhown M, DiMuzio MT, Cothran WC, Linde A (1983). Multiple forms of rat dentin phosphoproteins. Arch Biochem Biophys 225:178–186.

16.

Butler WT, Bhown M, Brunn JC, D’Souza RN, Farach-Carson MC, Happonen RP, et al. (1992). Isolation, characterization and immunolocalization of a 53-kDal dentin sialoprotein (DSP). Matrix 12:343–351.

17.

Butler WT, Ridall AL, McKee MD (1996). Osteopontin. In: Principles of bone biology. Bilezikian JP, Raisz LG, Rodan GA, editors. San Diego, CA: Academic Press, pp. 167–181.

18.

Butler WT, Brunn JC, Qin C, McKee MD (2002). Extracellular matrix proteins and the dynamics of dentin formation. Connect Tissue Res 43:301–307.

19.

Campos M, Couture C, Hirata IY, Juliano MA, Loisel TP, Crine P, et al. (2003). Human recombinant endopeptidase PHEX has a strict S1′ specificity for acidic residues and cleaves peptides derived from fibroblast growth factor-23 and matrix extracellular phosphoglycoprotein. Biochem J 373:271–279.

20.

Chellaiah MA, Kizer N, Biswas R, Alvarez U, Strauss-Schoenberger J, Rifas L, et al. (2003). Osteopontin deficiency produces osteoclast dysfunction due to reduced CD44 surface expression. Mol Biol Cell 14:173–189.

21.

Chen J, McCulloch CA, Sodek J (1993). Bone sialoprotein in developing porcine dental tissues: cellular expression and comparison of tissue localization with osteopontin and osteonectin. Arch Oral Biol 38:241–249.

22.

Cole WG (2002). Advances in osteogenesis imperfecta. Clin Orthop 40:6–16.

23.

Crine P, Dion N, Boileau G (1997). Endopeptidase-24.11. In: Cell-surface peptidases in health and disease. Kenny AJ, Boustead CM, editors. Oxford: BIOS Scientific Publishers Ltd., pp. 79–98.

24.

Denhardt DT, Noda M, O’Regan AW, Pavlin D, Berman JS (2001). Osteopontin as a means to cope with environmental insults: regulation of inflammation, tissue remodeling, and cell survival. J Clin Invest 107:1055–1061.

25.

D’Souza RN, Bronckers AL, Happonen RP, Doga DA, Farach-Carson MC, Butler WT (1992). Developmental expression of a 53 KD dentin sialoprotein in rat tooth organs. J Histochem Cytochem 40:359–366.

26.

Dickson IR, DiMuzio MT, Volpin D, Ananthanarayanan S, Veis A (1975). The extraction of phosphoproteins from bovine dentin. Calcif Tissue Res 19:51–61.

27.

DiMuzio MT, Veis A (1978). Phosphophoryns—major non-collageneous proteins of rat incisor dentin. Calcif Tissue Res 25:169–178.

28.

Du L, Desbarats M, Veil J, Glorieus FH, Cawthorn C, Ecarot B (1996). cDNA cloning of the murine Pex gene implicated in X-linked hypophosphatemia and evidence for expression in bone. Genomics 36:22–28.

29.

Ecarot B, Glorieux FH, Desbarats M, Travers R, Labelle L (1992). Defective bone formation by Hyp mouse bone cells transplanted into normal mice: evidence in favor of an intrinsic osteoblast defect. J Bone Mineral Res 7:215–220.

30.

Ek-Rylander B, Flores M, Wendel M, Heinegard D, Andersson G (1994). Dephosphorylation of osteopontin and bone sialoprotein by osteoclastic tartrate-resistant acid phosphatase. Modulation of osteoclast adhesion in vitro. J Biol Chem 269:14853–14856.

31.

Feng JQ, Ye L, Chen D, Huang H, Zhang J, Lu Y, et al. (2002a). Dmp1 deficient mice develop dwarfism, chondrodysplasia, and exhibit disorganized bone mineralization during post-natal development (abstract). J Bone Miner Res 17(Suppl 1):S127.

32.

Feng JQ, Zhang J, Dallas SL, Lu Y, Chen S, Tan X, et al. (2002b). Dentin matrix protein 1, a target molecule for Cbfa1 in bone, is a unique bone marker gene. J Bone Miner Res 17:1822–1831.

33.

Fisher LW, Whitson SW, Avioli LV, Termine JD (1983). Matrix sialoprotein of developing bone. J Biol Chem 258:12723–12727.

34.

Fisher LW, Hawkins GR, Tuross N, Termine JD (1987). Purification and partial characterization of small proteoglycans I and II, bone sialoproteins I and II, and osteonectin from the mineral compartment of developing human bone. J Biol Chem 262:9702–9708.

35.

Fisher LW, McBride OW, Termine JD, Young MF (1990). Human bone sialoprotein. Deduced protein sequence and chromosomal localization. J Biol Chem 265:2347–2351.

36.

Fisher LW, Torchia DA, Fohr B, Young MF, Fedarko NS (2001). Flexible structures of SIBLING proteins, bone sialoprotein, and osteopontin. Biochem Biophys Res Commun 280:460–465.

37.

Franzén A, Heinegård D (1985). Isolation and characterization of two sialoproteins present only in bone calcified matrix. Biochem J 232:715–724.

38.

Fujisawa R, Nodasaka Y, Kuboki Y (1995). Further characterization of interaction between bone sialoprotein (BSP) and collagen. Calcif Tissue Int 56:140–144.

39.

Ganss B, Kim RH, Sodek J (1999). Bone sialoprotein. Crit Rev Oral Biol Med 10:79–98.

40.

George A, Sabsay B, Simolan PAL, Veis A (1993). Characterization of a novel dentin matrix acidic phosphoprotein. J Biol Chem 268:12624–12630.

41.

George A, Bannon L, Sabsay B, Dillon JW, Malone J, Veis A, et al. (1996). The carboxyl-terminal domain of phosphophoryn contains unique extended triplet amino acid repeat sequences forming ordered carboxyl-phosphate interaction ridges that may be essential in the biomineralization process. J Biol Chem 271:32869–32873.

42.

Giachelli CM, Steitz S (2000). Osteopontin: a versatile regulator of inflammation and biomineralization. Matrix Biol 19:615–622.

43.

Goldberg HA, Hunter GK (1995). The inhibitory activity of osteopontin on hydroxyapatite formation in vitro. Ann NY Acad Sci 760:305–308.

44.

Goldberg HA, Warner KJ, Stillman MJ, Hunter GK (1996). Determination of the hydroxyapatite-nucleating region of bone sialoprotein. Connect Tissue Res 35:385–392.

45.

Grieff M, Mumm S, Waeltz P, Mazzarella R, Whyte MP, Thakker RV, et al. (1997). Expression and cloning of the human X-linked hypophosphatemia gene cDNA. Biochem Biophys Res Commun 231:635–639.

46.

Gu K, Chang S, Ritchie HH, Clarkson BH, Rutherford RB (2000). Molecular cloning of a human dentin sialophosphoprotein gene. Eur J Oral Sci 108:35–42.

47.

Guo R, Quarles LD (1997). Cloning and sequencing of human PEX from a bone cDNA library: evidence for its development stage-specific regulation in osteoblasts. J Bone Miner Res 12:1009–1017.

48.

Guo R, Liu S, Spurney RF, Quarles LD (2001). Analysis of recombinant PHEX: an endopeptidase in search of a substrate. Am J Physiol Endocrinol Metab 281:E837–E847.

49.

Herring GM (1972). The organic matrix of bone. In: The biochemistry and physiology of bone. Bourne GH, editor. New York: Academic Press, pp. 127–189.

50.

Hirst KL, Ibaraki-O’Connor K, Young MF, Dixon MJ (1997). Cloning and expression analysis of the bovine dentin matrix acidic phosphoprotein gene. J Dent Res 76:754–760.

51.

Holm IA, Huang X, Kunkel LM (1997). Mutational analysis of the PHEX gene in patients with X-linked hypophosphatemic rickets. Am J Hum Genet 60:790–797.

52.

Hunter GK, Goldberg HA (1994). Modulation of crystal formation by bone phosphoproteins: role of glutamic acid-rich sequences in the nucleation of hydroxyapatite by bone sialoprotein. Biochem J 302:175–179.

53.

Hunter GK, Kyle CL, Goldberg HA (1994). Modulation of crystal formation by bone phosphoproteins: structural specificity of the osteopontin-mediated inhibition of hydroxyapatite formation. Biochem J 300:723–728.

54.

HYP Consortium, The (1995). A gene (PHEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nat Genet 11:130–136.

55.

Ishijima M, Rittling SR, Yamashita T, Tsuji K, Kurosawa H, Nifuji A, et al. (2001). Enhancement of osteoclastic bone resorption and suppression of osteoblastic bone formation in response to reduced mechanical stress do not occur in the absence of osteopontin. J Exp Med 193:399–404.

56.

Jono S, Peinado C, Giachelli CM (2000). Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification. J Biol Chem 275:20197–20203.

57.

Kaartinen MT, El-Maadawy S, Rasanen NH, McKee MD (2002). Tissue transglutaminase and its substrates in bone. J Bone Miner Res 17:2161–2173.

58.

Kasugai S, Zhang Q, Overall CM, Wrana JL, Butler WT, Sodek J (1991). Differential regulation of the 55 and 44 kDa forms of secreted phosphoprotein 1 (SPP-1, osteopontin) in normal and transformed rat bone cells by osteotropic hormones, growth factors and a tumor promoter. Bone Miner 13:235–250.

59.

Katayama Y, House CM, Udagawa N, Kazama JJ, McFarland RJ, Martin TJ, et al. (1998). Casein kinase 2 phosphorylation of recombinant rat osteopontin enhances adhesion of osteoclasts but not osteoblasts. J Cell Physiol 176:179–187.

60.

Lasa M, Chang PL, Prince CW, Pinna LA (1997). Phosphorylation of osteopontin by Golgi apparatus casein kinase. Biochem Biophys Res Commun 240:602–605.

61.

Linde A, Lussi A, Crenshaw MA (1989). Mineral induction by immobilized polyanionic proteins. Calcif Tissue Int 44:286–295.

62.

Lipman ML, Panda D, Bennett HP, Henderson JE, Shane E, Shen Y, et al. (1998). Cloning of human PEX cDNA. Expression, subcellular localization, and endopeptidase activity. J Biol Chem 273:13729–13737.

63.

Liu S, Guo R, Tu Q, Quarles LD (2002). Overexpression of PHEX in osteoblasts fails to rescue the Hyp mouse phenotype. J Biol Chem 277:3686–3697.

64.

MacDougall M, Simmons D, Luan X, Nydegger J, Feng J, Gu TT (1997). Dentin phosphoprotein and dentin sialoprotein are cleavage products expressed from a single transcript coded by a gene on human chromosome 4. Dentin phosphoprotein DNA sequence determination. J Biol Chem 272:835–842.

65.

MacDougall M, Gu TT, Luan X, Simmons D, Chen J (1998). Identification of a novel isoform of mouse dentin matrix protein 1: spatial expression in mineralized tissue. J Bone Miner Res 13:422–431.

66.

Midura RJ, McQuillan DJ, Benham KJ, Fisher LW, Hascall VC (1990). A rat osteogenic cell line (UMR 106-01) synthesizes a highly sulfated form of bone sialoprotein. J Biol Chem 265:5285–5291.

67.

Nagata T, Todescan R, Goldberg HA, Zhang Q, Sodek J (1989). Sulphation of secreted phosphoprotein I (SPPI, osteopontin) is associated with mineralized tissue formation. Biochem Biophys Res Commun 165:234–240.

68.

Nagata T, Bellows CG, Kasugai S, Butler WT, Sodek J (1991). Biosynthesis of bone proteins [SPP-1 (secreted phosphoprotein-1, osteopontin), BSP (bone sialoprotein) and SPARC (osteonectin)] in association with mineralized-tissue formation by fetal-rat calvarial cells in culture. Biochem J 274:513–520.

69.

Narayanan K, Srinivas R, Ramachandran A, Hao J, Quinn B, George A (2001). Differentiation of embryonic mesenchymal cells to odontoblast-like cells by overexpression of dentin matrix protein 1. Proc Natl Acad Sci USA 98:4516–4521.

70.

Narayanan K, Ramachandran A, Hao J, He G, Park KW, Cho M, et al. (2003). Dual functional roles of dentin matrix protein 1. Implications in biomineralization and gene transcription by activation of intracellular Ca²⁺ store. J Biol Chem 278:17500–17508.

71.

Neame PJ, Butler WT (1996). Posttranslational modification in rat bone osteopontin. Connect Tissue Res 35:145–150.

72.

O’Regan A, Berman JS (2000). Osteopontin: a key cytokine in cell-mediated and granulomatous inflammation. Int J Exp Pathol 81:373–390.

73.

Oldberg A, Franzén A, Heinegård D (1986). Cloning and sequence analysis of rat bone sialoprotein (osteopontin) cDNA reveals an Arg-Gly-Asp cell-binding sequence. Proc Natl Acad Sci USA 83:8819–8823.

74.

Oldberg A, Franzén A, Heinegård D (1988). The primary structure of a cell-binding bone sialoprotein. J Biol Chem 263:19430–19432.

75.

Prince CW, Oosawa T, Butler WT, Tomana M, Bhown AS, Bhown M, et al. (1987). Isolation, characterization, and biosynthesis of a phosphorylated glycoprotein from rat bone. J Biol Chem 262:2900–2907.

76.

Prince CW, Dickie D, Krumdieck CL (1991). Osteopontin, a substrate for transglutaminase and factor XIII activity. Biochem Biophys Res Commun 177:1205–1210.

77.

Qin C, Brunn JC, Jones J, George A, Ramachandran A, Gorski JP, et al. (2001a). A comparative study of sialic acid-rich proteins in rat bone and dentin. Eur J Oral Sci 109:133–141.

78.

Qin C, Cook RG, Orkiszewski RS, Butler WT (2001b). Identification and characterization of the carboxyl-terminal region of rat dentin sialoprotein. J Biol Chem 276:904–909.

79.

Qin C, Brunn JC, Cadena E, Ridall A, Tsujigiwa H, Nagatsuka H, et al. (2002). The expression of dentin sialophosphoprotein gene in bone. J Dent Res 81:392–394.

80.

Qin C, Brunn JC, Cook RG, Orkiszewski RS, Malone JP, Veis A, et al. (2003a). Evidence for the proteolytic processing of dentin matrix protein 1: identification and characterization of processed fragments and cleavage sites. J Biol Chem 278:34700–34708.

81.

Qin C, Brunn JC, Baba O, Wygant JN, McIntyre BW, Butler WT (2003b). Dentin sialoprotein isoforms: detection and characterization of a high molecular weight dentin sialoprotein. Eur J Oral Sci 111:235–242.

82.

Qin C, Brunn JC, Cadena E, Ridall A, Butler WT (2003c). Dentin sialoprotein in bone and dentin sialophosphoprotein gene by osteoblasts. Connect Tissue Res 44(Suppl 1):179–183.

83.

Quarles LD (2003). FGF23, PHEX, and MEPE regulation of phosphate homeostasis and skeletal mineralization. Am J Physiol Endocrinol Metab 285:E1–E9.

84.

Rasmussen H, Anast C (1983). Familial hypophosphatemic rickets and vitamin D-dependent rickets. In: The metabolic basis of inherited disease. Stanbury JB, Wyngaarden JB, Fredrickson DS, Goldstein JL, Brown MS, editors. New York: McGraw-Hill, pp. 1743–1773.

85.

Razzouk S, Brunn JC, Qin C, Tye CB, Goldberg HA, Butler WT (2002). Osteopontin posttranslational modifications, possibly phosphorylation, are required for in vitro bone resorption but not osteoclast adhesion. Bone 30:40–47.

86.

Richardson W, Munksgaard EC, Butler WT (1978). Rat incisor phosphoprotein: the nature of the phosphate and quantitation of the phosphoserine. J Biol Chem 253:8042–8046.

87.

Ritchie HH, Hou H, Veis A, Butler WT (1994). Cloning and sequence determination of rat dentin sialoprotein, a novel dentin protein. J Biol Chem 269:3698–3702.

88.

Ritchie HH, Wang LH, Knudtson K (2001). A novel rat 523 amino acid phosphophoryn: nucleotide sequence and genomic organization. Biochim Biophys Acta 1520:212–222.

89.

Rittling SR, Denhardt DT (1999). Osteopontin function in pathology: lessons from osteopontin-deficient mice. Exp Nephrol 7:103–113.

90.

Rittling SR, Matsumoto HN, McKee MD, Nanci A, An XR, Novick KE, et al. (1998). Mice lacking osteopontin show normal development and bone structure but display altered osteoclast formation in vitro. J Bone Miner Res 13:1101–1111.

91.

Ross FP, Chappel J, Alvarez JI, Sander D, Butler WT, Farach-Carson MC, et al. (1993). Interactions between the bone matrix proteins osteopontin and bone sialoprotein and the osteoclast integrin alpha v beta 3 potentiate bone resorption. J Biol Chem 268:9901–9907.

92.

Rowe PS, de Zoysa PA, Dong R, Wang HR, White KE, Econs MJ, et al. (2000). MEPE, a new gene expressed in bone marrow and tumors causing osteomalacia. Genomics 67:54–68.

93.

Ruchon AF, Marcinkiewicz M, Siegfried G, Tenenhouse HS, DesGroseillers L, Crine P, et al. (1998). Pex mRNA is localized in developing mouse osteoblasts and odontoblasts. J Histochem Cytochem 46:459–468.

94.

Ruchon AF, Tenenhouse HS, Marcinkiewicz M, Siegfried G, Aubin JE, DesGroseillers L, et al. (2000). Developmental expression and tissue distribution of PHEX protein: effect of the Hyp mutation and relationship to bone markers. J Bone Miner Res 15:1440–1450.

95.

Sabbagh Y, Boileau G, Campos M, Carmona AK, Tenenhouse HS (2003). Structure and function of disease-causing missense mutations in the PHEX gene. J Clin Endocrinol Metab 88:2213–2222.

96.

Safran JB, Butler WT, Farach-Carson MC (1998a). Modulation of osteopontin post-translational state by 1,25-(OH)₂-vitamin D3. Dependence on Ca²⁺ influx. J Biol Chem 273:29935–29941.

97.

Safran JB, Butler WT, Farach-Carson MC (1998b). Posttranslational regulation of osteopontin (OPN) by 1,25-dihydroxy vitamin D3. In: Proceedings of the Sixth International Conference on Chemistry and Biology of Mineralized Tissue, Vittel, France. Goldberg M, Boskey A, Robinson C, editors. Rosemont, IL: American Academy of Orthopaedic Surgeons, pp. 113–116.

98.

Saito T, Arsenault AL, Yamauchi M, Kuboki Y, Crenshaw MA (1997). Mineral induction by immobilized phosphoproteins. Bone 21:305–311.

99.

Salih E (2003). In vivo and in vitro phosphorylation regions of bovine bone sialoprotein. Connect Tissue Res 44:223–229.

100.

Sato H, Takino T, Okada Y, Cao J, Shinagawa A, Yamamoto E, et al. (1994). A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370:61–65.

101.

Seiki M, Yana I (2003). Roles of pericellular proteolysis by membrane type-1 matrix metalloproteinase in cancer invasion and angiogenesis. Cancer Sci 94:569–574.

102.

Senger DR, Wirth DF, Hynes RO (1979). Transformed mammalian cells secrete specific proteins and phosphoproteins. Cell 16:885–893.

103.

Sodek J, Ganss B, McKee MD (2000). Osteopontin. Crit Rev Oral Biol Med 11:279–303.

104.

Somerman MJ, Sauk JJ, Foster RA, Norris K, Dickerson K, Argraves WS (1991). Cell attachment activity of cementum: bone sialoprotein II identified in cementum. J Periodontal Res 26:10–16.

105.

Sorensen ES, Hojrup P, Petersen TE (1995). Posttranslational modifications of bovine osteopontin: identification of twenty-eight phosphorylation and three O-glycosylation sites. Protein Sci 4:2040–2049.

106.

Sreenath T, Thyagarajan T, Hall B, Longenecker G, D’Souza R, Hong S, et al. (2003). Dentin sialophosphoprotein knockout mouse teeth display widened predentin zone and develop defective dentin mineralization similar to human dentinogenesis imperfecta-III. J Biol Chem 278:24874–24880.

107.

Srinivasan R, Chen B, Gorski JP, George A (1999). Recombinant expression and characterization of dentin matrix protein 1. Connect Tissue Res 40:251–258.

108.

Steiglitz BM, Ayala M, Narayanan K, George A, Greenspan DS (2004). Bone morphogenetic protein-1/tolloid-like proteinases process dentin matrix protein-1. J Biol Chem 279:980–986.

109.

Suzuki K, Zhu B, Rittling SR, Denhardt DT, Goldberg HA, McCulloch CA, et al. (2002). Colocalization of intracellular osteopontin with CD44 is associated with migration, cell fusion, and resorption in osteoclasts. J Bone Miner Res 17:1486–1497.

110.

Thompson DL, Sabbagh Y, Tenenhouse HS, Roche PC, Drezner MK, Salisbury JL, et al. (2002). Ontogeny of PHEX/PHEX protein expression in mouse embryo and subcellular localization in osteoblasts. J Bone Miner Res 17:311–320.

111.

Veis A, Perry A (1967). The phosphorylation of the dentin matrix. Biochemistry 6:2409–2416.

112.

Wada T, McKee MD, Steitz S, Giachelli CM (1999). Calcification of vascular smooth muscle cell cultures: inhibition by osteopontin. Circ Res 84:166–178.

113.

Winters RW, Graham JB, Williams TF, McFalls VW, Burnett CH (1958). A genetic study of familial hypophosphatemia and vitamin D resistant rickets with a review of the literature. Medicine 37:97–142.

114.

Wuttke M, Muller S, Nitsche DP, Paulsson M, Hanisch FG, Maurer P (2001). Structural characterization of human recombinant and bone-derived bone sialoprotein. Functional implications for cell attachment and hydroxyapatite binding. J Biol Chem 276:36839–36848.

115.

Xiao S, Yu C, Chou X, Yuan W, Wang Y, Bu L, et al. (2001). Dentinogenesis imperfecta 1 with or without progressive hearing loss is associated with distinct mutations in DSPP. Nat Genet 27:201–204.

116.

Xiao ZS, Crenshaw M, Guo R, Nesbitt T, Drezner MK, Quarles LD (1998). Intrinsic mineralization defect in Hyp mouse osteoblasts. Am J Physiol Endocrinol Metab 275:E700–E708.

117.

Xu T, Bianco P, Fisher LW, Longenecker G, Smith E, Goldstein S, et al. (1998). Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat Genet 20:78–82.

118.

Yamakoshi Y, Hu JC, Liu S, Zhang C, Oida S, Fukae M, et al. (2003). Characterization of porcine dentin sialoprotein (DSP) and dentin sialophosphoprotein (DSPP) cDNA clones. Eur J Oral Sci 111:60–67.

119.

Zaia J, Boynton R, Heinegard D, Barry F (2001). Posttranslational modifications to human bone sialoprotein determined by mass spectrometry. Biochemistry 40:12983–12991.

120.

Zhang Q, Domenicucci C, Goldberg HA, Wrana JL, Sodek J (1990). Characterization of fetal porcine bone sialoproteins, secreted phosphoprotein I (SPPI, osteopontin), bone sialoprotein, and a 23-kDa glycoprotein. Demonstration that the 23-kDa glycoprotein is derived from the carboxyl terminus of SPPI. J Biol Chem 265:7583–7589.

121.

Zhang X, Zhao J, Li C, Gao S, Qiu C, Liu P, et al. (2001). DSPP mutation in dentinogenesis imperfecta Shields type II. Nat Genet 27:151–152.