| Sign In to gain access to subscriptions and/or personal tools. |
Spectroscopic Imaging of Mineral Maturation in Bovine Dentin
1 Hospital for Special Surgery, New York, NY, USA; Correspondence: * corresponding author, boskeya{at}hss.edu
Dentin is a useful model for the study of mineral maturation. Using Fourier Transform Infrared Imaging (FTIRI), we characterized distinct regions in developing dentin at 7-µm spatial resolution. Mineral-to-matrix ratio and crystallinity in bovine dentin from cervical and incisal parts of 3rd-trimester fetal compared with one-year-old incisor crowns showed that virtually all maturation stages in dentin could be spectroscopically isolated and analyzed. In the fetal incisors, mantle and circumpulpal dentin presented distinct patterns of mineral maturation. Gradients in both mineral properties examined were observed at the mineralization front and at the dentino-enamel junction.
Key Words: FTIR imaging • mineralization • dentinogenesis
Dentin presents an excellent model for the study of mineral maturation during biomineralization, since samples for experimental studies are abundant, and dentin, as distinct from bone, is not remodeled. Both predentin (the unmineralized precursor) and the mature mineralized tissue have been previously characterized as homogenates (Butler, 1984; Linde, 1984). Using spectroscopic, x-ray, and biomechanical microanalysis methods (Wentrup-Byrne et al., 1997; Tesch et al., 2001), investigators have shown dentin to present a wide variation of properties dependent on location. At the same time, details on the origin of this spatial variation and data showing its evolution with tissue maturation are lacking. Within mineralized dentin, there are two distinct dentin tissue compartments: The first formed is mantle dentin, which is adjacent to the enamel, and the remainder is circumpulpal dentin. These two compartments have distinct matrix composition and physical and biomechanical properties (Ten Cate, 1994). Patterns of biomineralization and mineral maturation should also be distinct in each. The existing anatomical variation within dentin (Pashley, 1989) mandates a comprehensive analysis of the components of both non-homogenized mantle and circumpulpal dentin (from the mineralization front to the dentino-enamel junction). In this study, we used Fourier Transform Infrared Imaging (FTIRI) to characterize the mineral in anatomically distinct regions of bovine dentin to provide insight into the developmental stages of mineral maturation.
Specimen Preparation Dentin samples from 6 developing unerupted incisors of 3rd-trimester calves and from 3 mature bovine incisors (year-old animals) were analyzed. Bovine jaws were obtained from a commercial source (Aries Scientific, Dallas, Tx, USA) and stored at -70°C prior to use, at which time teeth were extracted from the jaws. The IRB of the Hospital for Special Surgery has determined that materials obtained from an abbatoir are exempt from their review. The specimens were partially fixed in absolute methanol, dehydrated through a series of ethanol and acetone gradients, and embedded in polymethylmethacrylate (PMMA). Teeth were bisected longitudinally with a diamond wafer wheel saw, and non-decalcified 2-µm-thick sections of the incisor crowns were produced from one half by means of a Jung Polycut E microtome (Reichert-Jung, Heidelberg, Germany). The sections were mounted between 2 barium fluoride windows for FTIR imaging.
FTIR Imaging Analysis
Specimen Preparation Typical sections of bovine teeth used for FTIR analyses are shown in Figs. 1a  and 1b. FTIR images and spectroscopic information are shown from these specimens, but they are representative of all results obtained. The cervical (young tissue), mid-crown, and incisal (mature tissue) areas that were analyzed are indicated in Fig. 1. Only cervical and incisal areas were analyzed for the year-old incisor, for comparison of mineral and matrix properties and validation of the assumption that variation in dentin properties as a function of cervical/incisal location on the mature incisor is practically negligible.
Fig. 1c  shows selected superimposed spectra acquired from the fetal incisor section of Fig. 1a (continuous-line spectra), and from the cervical area of the mature year-old incisor (stippled spectrum). These spectra were extracted from images of the mantle dentin area ~ 50 µm from the DEJ at different distances from the cervix (as indicated). With tissue maturation, a continuous increase in the total area of  1,3 phosphate bands is obvious, suggesting a respective increase in relative mineral content, while, in parallel, there is a small decrease in the Amide I area. A distinct change in the 1,3 phosphate band contour is also visible.
FTIR Imaging
Fig. 4 shows the distribution of values for the mineral:matrix ratio and crystallinity along the predentin-enamel line, represented by the white arrows on the cervical field from the fetal bovine incisor (Fig. 4a), and proximal to the pulp cervical field of the year-old incisor (Fig. 4b), noted with an asterisk in Figs. 2a and 3a, respectively. The whole range of dentin (predentin through enamel) is covered in the distribution shown in Fig. 4a, whereas only the mineralization front and part of circumpulpal dentin is covered in Fig. 4b. In the mineralization front region for the fetal and year-old incisors, there is no sharp line demarcating the mineralized and non-mineralized matrix; rather, there is a zone of transition between the two (taking place within approximately 30 µm in young teeth). This transition is obvious in the mineral:matrix value distributions and partly evidenced in the crystallinity value distributions along the line analyzed. Adjacent to the DEJ in the fetal incisor, another transition occurs, from dentin mineral:matrix values to those characteristic of enamel. Enamel crystallinity is not shown, since the parameter cannot be calculated, because there is no "non-stoichiometric apatite" component, and crystallinity values for these pixels have been omitted. In mature teeth, the same distributions cannot be analyzed for the DEJ area, since enamel fractured away during sectioning.
This study demonstrates the usefulness of both FTIR imaging and the fetal bovine tooth for characterizing changes in mineral properties of dentin during tooth development. Specifically, age- and site-dependent changes in relative mineral content, as well as relative crystal maturity, are demonstrated in the whole incisors. FTIRI as an analytical imaging method has been used to obtain qualitative and quantitative information on both mineral and matrix of bone (Marcott et al., 1999; Mendelsohn et al., 1999; Boskey et al., 2002b), calcifying cartilage cultures (Boskey et al., 2002a), and cartilage (Bhargava and Levin, 2001; Camacho et al., 2001), with a spatial resolution of ~ 7 µm. The two spectroscopic parameters, mineral:matrix ratio and crystallinity, were selected for dentin analysis because, as ratios, they are independent of sectioning artifacts, and they represent important properties. The mineral:matrix ratio is linearly related to the chemically measured ash weight (Pienkowski et al., 1997). Crystallinity reflects both the mineral crystallite size along the c-axis and the crystal perfection as determined by x-ray diffraction (Gadaleta et al., 1996). In the present study, incisors from year-old animals were selected as the mature controls, because, with age, attrition of the incisors introduces pathology into the dentinal tissues, making older samples impractical for study. A notable point from the results shown is that enamel in the 3rd-trimester incisors could be sectioned intact, facilitating study of the DEJ and developing enamel in the same sections.
The FTIRI data allowed for visualization of the two distinct dentin compartments, demonstrating an earlier initiation of mineralization and crystal growth in mantle dentin and a more prolonged crystal growth period in circumpulpal dentin, that finally reaches overall higher mineral:matrix ratios. It is also interesting to note that crystallinity continues to increase after complete mineralization (constant mineral:matrix ratio) in both mantle and circumpulpal dentin. While the mineral:matrix values are very similar between cervical and incisal regions in the year-old incisor (Fig. 3b Mantle dentin evolves as a separate entity from the rest of dentin. The amount of mantle dentin, whether hypo- or hypermineralized, as well as its relative content is controversial (Moss, 1974; Herr et al., 1986). The distribution of mineral content and crystallinity, with respect to that in the circumpulpal dentin, can be clearly defined by FTIRI. The distinct pattern in mineral maturation that appears in mantle dentin implies a separate mineralization mechanism in this area. Mantle dentin mineralization is believed to be initiated in matrix vesicles (Katchburian, 1973). The highly phosphorylated proteins, which are believed to regulate biomineralization in circumpulpal dentin (Butler, 1998), are less abundant in mantle dentin (Rahima et al., 1988). Minor circumpulpal dentin constituents, such as osteopontin and osteocalcin, are prominent in mantle dentin (McKee et al., 1996). Spatial variations in mineral properties are caused both by the presence of mantle and circumpulpal dentin and by variations in peritubular dentin density. This shows clearly in the distribution of values within circumpulpal dentin in the year-old incisor. Previous studies have also shown mineral variation as a function of location in mature teeth (Kinney et al., 1996, 2001). Tesch et al. (2001), using various methods, including FTIR microspectroscopy, also showed variations in the structural and mechanical properties of the mineral as a function of location in the mature human tooth. This variation is due in part to the decrease in dentinal tubule density and a respective decrease in peritubular dentin density in areas farther from the mineralization front (Pashley, 1989). Since the mineral concentration and, most likely, the nature of the organic matrix in peritubular dentin differ from those in intertubular dentin (Weiner et al., 1999), a spatial variation is anticipated. The linear gradients of mineral or matrix properties at the mineralization front and the DEJ shown in the present study also agree with observations made by other less-highly-resolved techniques. The extensive predentin-to-dentin transition was observed in the rat, both morphologically and by histochemical methods (Goldberg et al., 1998). The existence of a DEJ mineral gradient was hypothesized based on observations of biomechanical properties (White et al., 2000; Marshall et al., 2001)). A protein continuum is also hypothesized to exist at the DEJ (Bodier-Houllé et al., 2000). The fetal bovine model presented covers a maturation span for the tissue from very early stages to near-maturity. With tissue maturation, mineral proliferates through secondary nucleation, growth, and perfection of individual crystals and crystal agglomeration (Heywood et al., 1990), while the matrix is thought to be degraded after it is laid down in the predentin (Veis, 1993). In this way, each mineral:matrix value within dentin represents a single time point in the development of the particular tissue. A slight reduction in the relative matrix content while the relative mineral content is still increasing, as found in this study, is consistent with previous data from the analysis of dentin matrix in whole fetal incisors at different gestation stages (Lee et al., 1983). As the results from the year-old bovine incisor show, in the mature incisors, values for mineral and matrix content and mineral maturation from cervical-area dentin are at comparable levels with those of the incisal area, and the two areas are most likely eventually equivalent in properties. This validates the assumption that, in the 3rd trimester, differences in the tissue properties of fetal incisors between cervical and incisal areas, for comparable locations of tissue points from the pulp or the DEJ, are due to tissue age difference. Conversely, as images for the the mature tooth show, differences in spatial distribution do not necessarily represent temporal differences. FTIRI studies of dentin mineral and matrix maturation provide an opportunity for the analysis of the complex interactions between mineral and matrix of tissues during the biomineralization process. The model and technique evaluated here show the development of mantle and circumpulpal dentin as separate entities, the contribution of histologic variation in the properties of each, and the creation of histologic structures with characteristic properties, such as tissue interfaces in the mineralization front and DEJ area.
This study was funded by NIH grants K08 DE00467 and R01 DE04141. FTIR imaging was performed at the Hospital for Special Surgery IR Imaging Core, sponsored by NIH grant AR046121. This work was done in partial fulfillment of Dr. Verdelis’ requirements for a PhD in Oral Biology, Univ. of North Carolina in Chapel Hill. The authors are grateful to Dr. J.T. Wright for his suggestions and comments. Received for publication January 17, 2003. Revision received May 5, 2003. Accepted for publication June 3, 2003.
Journal of Dental Research, Vol. 82, No. 9,
697-702 (2003)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
