Jeffrey P. Gorski

Acidic phosphoproteins from bone matrix : a structural rationalization of their role in biomineralization

SummaryOsteopontin, bone sialoprotein, and bone acidic glycoprotein-75 are three acidic phosphoproteins that are isolated from the mineralized phase of bone matrix, are synthesized by osteoblastic cells, and are generally restricted in their distribution to calcified tissues. Although each is a distinct gene product, these proteins share aspartic/glutamic acid contents of 30–36% and each contains multiple phosphoryl and sialyl groups. These properties, plus a strict relationship of acidic macromolecules with cell-controlled mineralization throughout nature, suggest functions in calcium binding and nucleation of calcium hydroxyapatite crystal formation. However, direct proof for such roles is still largely indirect in nature. The purpose of this review is to present two speculative hypotheses regarding acidic phosphoprotein function. The goal was to use new sequence information along with database comparisons to develop a structural rationalization of how these proteins may function in calcium handling by bone. For example, our analysis has identified a conserved polyacidic stretch in all three phosphoproteins which we propose mediates metal binding. Also, conserved motifs were identified that are analogous with those for casein kinase II phosphorylation sites and whose number correlates well with that of phosphoryl groups/protein. A two-state conformational model of calcium binding by bone matrix acidic phosphoproteins is described which incorporates these findings.

Is all bone the same? Distinctive distributions and properties of non-collagenous matrix proteins in lamellar vs. woven bone imply the existence of different underlying osteogenic mechanisms.

The purpose of this review is to summarize recent functional and structural findings regarding non-collagenous matrix proteins in bone and teeth, to compare gene locations for bone and tooth matrix proteins with loci for hereditary skeletal diseases, and to present several provocative hypotheses which integrate this new information into a physiological context. Hypothesis I proposes that the molecular composition of rapidly deposited and mineralized woven bone, as well as the responsiveness of cells synthesizing woven bone to stimuli, is different from that for more slowly synthesized lamellar bone, implying the existence of distinctive osteogenic mechanisms. This review of recent research strongly supports this proposal. Briefly, the protein composition of woven bone matrix is enriched in acidic phosphoproteins BAG-75 and BSP, which are not expressed in lamellar bone, which is itself enriched in osteocalcin. De novo deposition and mineralization of woven bone occurs faster than in lamellar bone by means of a matrix-vesicle-assisted mechanism. Deposition of woven bone occurs at sites experiencing biomechanical strains higher than those experienced by lamellar bone. In addition, woven bone in metaphyseal regions is more susceptible to osteoclastic resorption after space flight, ovariectomy, and loss of weightbearing than is lamellar bone. Finally, osteoprogenitor cells responsive to parathyroid hormone reside in the metaphyseal region of long bones. Taken together, these findings suggest that Hypothesis I represents a useful paradigm for future studies. Specific functions mediated by most individual bone and tooth matrix proteins remain uncertain. A review of current literature suggests that the functionality of skeletal matrix proteins is expressed through specific binding sites composed of particular species-conserved structural motifs (Hypothesis 2). Examples include the previously recognized Asp-Ser-Ser motif of dentin phosphophoryns and the gamma-carboxyglutamic acid motif of matrix GLA protein and osteocalcin. A new polyacidic amino acid motif composed of consecutive Asp and Glu residues (n > 7) was defined in extracellular matrix components osteopontin, bone sialoprotein, and bone acidic glycoprotein-75 on the basis of strong functional analogies with similar polyacidic stretches in divalent metal storage proteins of the endoplasmic reticulum and sarcoplasmic reticulum. These structural motifs represent prime targets for future structure-function studies in vivo and in vitro.

Current Concepts of the Biology of Tooth Eruption

Tooth eruption is defined as the movement of a tooth from its site of development within the jaws to its position of function within the oral cavity. We present a critical review of evidence for the mechanisms and regulation of the intraosseous and supraosseous phases of eruption, with an emphasis upon the canine premolar model studied by the authors. Analyses at different stages of premolar eruption indicate that selective fragmentation of dental follicle protein DF-95 correlates with the presence of elevated levels of follicular collagenase and stromelysin, and with the onset of premolar movement. A dramatic decrease in these metalloproteinases followed initiation of movement. A biochemical and cell biological model for regulation of tooth eruption is proposed based upon these new and existing data.

Recombinant Expression and Characterization of Dentin Matrix Protein 1

Dentin matrix protein 1 (DMP1) is an extracellular matrix noncollagenous protein (NCP) initially isolated from dentin and now found to be present in calcified tissues like calvaria and long bone. The characteristic feature of DMP1 is that it contains a large number of acidic domains and has properties which implicate it as a key participant in regulating matrix mineralization. The level of DMP1 in the tissue is sparse and it is not easily isolated from dentin because it copurifies with other dentin NCPs. The exact function of DMP1 is not known and this is due to the inherent difficulty of obtaining enough protein from the mineralized tissues. In order to understand the physiologic role for DMP1 during the formation of mineralized tissues we have produced milligram quantities of recombinant DMP1 in E. coli. The objective of this work was: (1) to prepare unmodified apoprotein so that it could be used for studying the function of DMP1; and (2) to prepare polyclonal antibody against the recombinant DMP1 antigen. The DMP1 polyclonal antibody did not cross-react with other NCPs present in dentin or with bone acidic glycoprotein-75 (BAG-75) present in the bone matrix, confirming the specificity of this antibody and thus making it a valuable tool to determine the in vivo function of DMP1.

Developmental changes in the extracellular matrix of the dental follicle during tooth eruption

Eruption of the third and fourth mandibular premolars in the dog begins at 15 weeks of age, is dependent upon the dental follicle, and is complete by 23 weeks. Our study covered the period from 12 to 20 weeks, and revealed several changes in extracellular matrix structure and organization of the follicle which correlate with specific physiological events in eruption. First, the average DNA content per follicle reached a maximum at 14 weeks. Two weeks later, follicle size had increased 1.3- to 2.4- times. Second, the collagen content of follicles increased 2.5-fold over the study period, with two-thirds of this increase over the last four weeks. Type I collagen was the major collagen at all stages of follicular development. The amount of proteoglycan rose 45% from 16 to 20 weeks of age. Third, the ultrastructure of the dental follicle prior to eruption (12 weeks) indicated a disorganized interstitial connective tissue matrix; during eruption, two size classes of fibrils were observed which clustered together in linearly aligned bundles. Fourth, gel electrophoretic analyses resolved more than twenty follicle proteins with the major species a Mr = 95k glycoprotein. Immunoblotting demonstrated only one minor component was derived from serum. Comparison of noncollagenous proteins from different aged follicles indicated that three small polypeptides (Mr = 20-25 k) were present primarily at 16 weeks, the same time at which root elongation begins. A different sequence of changes was exhibited by two other proteins of Mr = 13 and 15 k. These findings may serve as biochemical markers of stages of dental follicle development and facilitate a search for local control mechanisms.

Hydroxyapatite Induces Autolytic Degradation and Inactivation of Matrix Metalloproteinase-1 and -3†

In the course of studies to identify a protease capable of producing a long‐lived 50 kDa fragment of bone acidic glycoprotein‐75 (BAG‐75), it was observed that incubation of matrix metalloproteinase (MMP)‐3 (stromelysin 1) with preparations of BAG‐75 led to inactivation of proteolytic function, e.g., an inability to fragment125I‐labeled BAG‐75 added subsequently. MMP‐1 (interstitial collagenase) was also inactivated by exposure to BAG‐75 preparations. Investigation of the mechanism revealed that BAG‐75 preparations contained millimolar levels of inorganic phosphate which formed hydroxyapatite crystals under digestion conditions. Hydroxyapatite crystals alone and in BAG‐75–hydroxyapatite complexes induced the autolytic degradation of both active and precursor forms of MMP‐1 and MMP‐3. Autolytic degradation in the presence of hydroxyapatite was demonstrated by a loss in catalytic function assayed with peptide and/or protein substrates, and, by fragmentation into polypeptides of <10 kDa. The fate of MMP‐3 incubated with hydroxyapatite depends upon the time of incubation, the free calcium concentration, and the concentration of crystals. Specifically, hydroxyapatite‐induced autolysis requires a near physiological free calcium concentration of 0.5–1.0 mM. Autolysis was maximal in the presence of 150 μg/ml hydroxyapatite where MMP‐3 was only partially bound to crystals. However, autolysis also occurred at higher crystal concentrations where all input MMP‐3 was bound (>1000 μg/ml), suggesting that autolysis may be mediated by bound enzyme. The effect of hydroxyapatite appears to be specific for MMP‐1 and MMP‐3 since the catalytic activity of chymotrypsin, trypsin, papain, and thermolysin remained unchanged after exposure to hydroxyapatite. These results document for the first time a novel catalytic role for hydroxyapatite crystals in vitro and provide an initial biochemical characterization of the intermolecular, autolytic, calcium ion‐dependent, matrix metalloproteinase‐specific degradative mechanism.

The effect of removing the true dental follicle on premolar eruption in the dog

Eruption is a highly localized process during which the bone resorption and formation that occur on opposite sides of the tooth are dependent upon the surrounding soft tissues, the true dental follicle externally and the enamel organ internally. To examine the ability of the enamel organ to cause eruption the external layer (dental follicle) was removed just prior to and up to 4 weeks before eruption in 13 mandibular premolars in dogs and eruption followed clinically, radiographically and histologically. None of the teeth without dental follicles erupted but three teeth from which the follicle was separated then replaced did erupt. These data indicate that the enamel organ without the dental follicle cannot support tooth eruption and provide indirect evidence for the central role of the dental follicle, alone or in combination with the enamel organ, in eruption.

Conformational Analyses on Soluble and Surface Bound Osteopontin

Immunohistology of calvarial sections revealed that staining with monoclonal anti-osteopontin antibodies (clone MPIIIB10) is minimal unless sections are first treated with EDTA. In contrast, following treatment of sections with EDTA, strong staining of mineralizing osteoid areas and osteoblast-like cells was noted (Fig. 1B). Immunostaining for osteopontin appeared to be specific in that controls which substituted rabbit IgG or normal mouse ascites fluid for monoclonal antibody, or which omitted monoclonal antibody uniformly gave background results (Fig. 1C). In an effort to circumvent problems of antibody accessibility we examined the immunoreactivity of OP when adsorbed to plastic and hydroxyapatite surfaces. Although OP bound to plastic surfaces is reactive with MPIIIB10 antibodies, OP adsorbed to hydroxyapatite crystal surfaces is not recognized by these antibodies as assessed by two detection methods. These results demonstrate that most or all of OP bound to hydroxyapatite exhibits a different conformation than when bound to plastic surfaces. On the basis of immunohistologic results with calvarial sections, we suggest that the conformation of native OP in bone and of isolated OP adsorbed to hydroxyapatite may be similar. Finally, solution circular dichroism and Fourier-transformed infrared spectroscopic studies indicate that the conformation of bone OP is dependent upon its concentration, and, secondarily to the presence or absence of calcium ion. With both spectroscopic methods, addition of calcium appeared to increase the extent of disordered structure. We suggest that these findings support our hypothesis that bone matrix proteins exhibit a different conformation when adsorbed on hydroxyapatite crystal surfaces. Assumption of a more organized secondary structure in concentrated OP solutions (i.e., 15 mg/ml) is consistent with these results in that local concentrations of OP within a semisolid matrix may approach or exceed levels used here.

Radiotherapy effect on nano-mechanical properties and chemical composition of enamel and dentine.

OBJECTIVE To understand radiotherapy-induced dental lesions characterized by enamel loss or delamination near the dentine-enamel junction (DEJ), this study evaluated enamel and dentine nano-mechanical properties and chemical composition before and after simulated oral cancer radiotherapy. DESIGN Sections from seven non-carious third molars were exposed to 2 Gy fractions, 5 days/week for 7 weeks for a total of 70 Gy. Nanoindentation was used to evaluate Youngs modulus, while Raman microspectroscopy was used to measure protein/mineral ratios, carbonate/phosphate ratios, and phosphate peak width. All measures were completed prior to and following radiation at the same four buccal and lingual sites 500 and 30 μm from the DEJ in enamel and dentine (E-500, E-30, D-30 and D-500). RESULTS The elastic modulus of enamel and dentine was significantly increased (P ≤ 0.05) following radiation. Based on Raman spectroscopic analysis, there was a significant decrease in the protein to mineral ratio (2931/430 cm(-1)) following radiation at all sites tested except at D-500, while the carbonate to phosphate ratio (1070/960 cm(-1)) increased at E-30 and decreased at D-500. Finally, phosphate peak width as measured by FWHM at 960 cm(-1) significantly decreased at both D-30 and D-500 following radiation. CONCLUSIONS Simulated radiotherapy produced an increase in the stiffness of enamel and dentine near the DEJ. Increased stiffness is speculated to be the result of the radiation-induced decrease in the protein content, with the percent reduction much greater in the enamel sites. Such changes in mechanical properties and chemical composition could potentially contribute to DEJ biomechanical failure leading to enamel delamination that occurs post-radiotherapy. However, other analyses are required for a better understanding of radiotherapy-induced effects on tooth structure to improve preventive and restorative treatments for oral cancer patients.

Type IV Collagen is a Novel DEJ Biomarker that is Reduced by Radiotherapy

The dental basement membrane (BM) is composed of collagen types IV, VI, VII, and XVII, fibronectin, and laminin and plays an inductive role in epithelial-mesenchymal interactions during tooth development. The BM is degraded and removed during later-stage tooth morphogenesis; however, its original position defines the location of the dentin-enamel junction (DEJ) in mature teeth. We recently demonstrated that type VII collagen is a novel component of the inner enamel organic matrix layer contiguous with the DEJ. Since it is frequently co-expressed with and forms functional complexes with type VII collagen, we hypothesized that type IV collagen should also be localized to the DEJ in mature human teeth. To identify collagen IV, we first evaluated defect-free erupted teeth from various donors. To investigate a possible stabilizing role, we also evaluated extracted teeth exposed to high-dose radiotherapy – teeth that manifest post-radiotherapy DEJ instability. We now show that type IV collagen is a component within the morphological DEJ of posterior and anterior teeth from individuals aged 18 to 80 yr. Confocal microscopy revealed that immunostained type IV collagen was restricted to the 5- to 10-µm-wide optical DEJ, while collagenase treatment or previous in vivo tooth-level exposure to > 60 Gray irradiation severely reduced immunoreactivity. This assignment was confirmed by Western blotting with whole-tooth crown and enamel extracts. Without reduction, type IV collagen contained macromolecular α-chains of 225 and 250 kDa. Compositionally, our results identify type IV collagen as the first macromolecular biomarker of the morphological DEJ of mature teeth. Given its network structure and propensity to stabilize the dermal-epidermal junction, we propose that a collagen-IV-enriched DEJ may, in part, explain its well-known fracture toughness, crack propagation resistance, and stability. In contrast, loss of type IV collagen may represent a biochemical rationale for the DEJ instability observed following oral cancer radiotherapy.