In the 19th century, Wolff discussed adaptation of bone due to mechanical forces [1
]. The occlusal force, primarily used in grinding the hard diet fed to rats, is the most prominent force in the periodontium [33
]. Within this adaptation lies growth and function-related changes in bone and cementum, which will be illustrated through various imaging modalities.
Micro-XCT is a noninvasive technique that requires minimum specimen preparation. Specimens can be imaged under wet conditions, preserving different tissue structures and at microscopic resolution below 5 microns. Tomographies and virtual scans can be processed to evaluate mineral density, resorption volumes, displacement fields, and 3D spatial association of the root with the bony socket. The 3D tooth-bone association can provide insights to form-function relationships. Additionally, 2D images recorded with other techniques including light microscopy can be related to the 3D tomographies and 2D virtual sections, and potential artifacts due to specimen preparation can be identified. In this study, we used Micro-XCT for in situ imaging, to approximate interdigitation of disarticulated maxilla and mandible of a rat by imaging at 2x magnification (, left). Root morphology, and resorbed pits in bone and roots were measured. Structural analysis through the volume of a specimen was correlated with virtual serial sections with least interpolation. Additionally, X-ray attenuation indicative of mineral density variations at the bone-PDL interface was investigated.
Figure 1 Left top: Simulated occlusion of a left rat maxilla with the corresponding mandible imaged with Micro-XCT. Left bottom: sagittal sections simulate open (left) and closed (right) bite. Top right: occlusal surface and transversal section of a right maxilla. (more ...)
The Micro-XCT images presented in Figures – illustrate an accurate anatomy of the dentoalveolar complex. Accuracy is necessary to spatially correlate 2D measurements from other complementary studies by identifying landmarks, and anatomical planes within the dentoalveolar complex. Furthermore, 2D and 3D data will help identify deviations due to external perturbations such as, disease or load-mediated influences from baseline measurements.
Figure 3 3D reconstruction of the distal roots of 2nd molar viewed from different directions indicated by corresponding colors in (e). Accordingly, (a) shows the distal (yellow arrow), (b) the mesial (blue), (c) the buccal-distal (black), and (d) the lingual-mesial (more ...)
A rat hemimaxilla contains one incisor and three molars. We focused our investigation on the molars responsible for masticatory function. The crown of the 1st molar is the biggest, with the largest occlusal surface, the 2nd molar measures approximately 2/3
s, and the 3rd molar 3/5
s of the length of the 1st molar. In all specimens, a significant amount of occlusal wear was commonly observed. demonstrates enamel wear and exposed dentin on the occlusal surface. This could lead to varying contact between opposing teeth and is a potential cause for altered biomechanics and modeling-related adaptation in the bone-PDL-cementum complex throughout the lifespan of an organism.
The 1st molar is located mesially and the 3rd molar distally. Often times the challenge lies in identifying the anatomical directions of the specimen when only a part of it is imaged. Hence, certain predefined anatomical features are used to assign anatomical directions, and planes. Lingual and buccal sides of the hemimaxilla denote the tongue and cheek-sides, respectively. The bone around the molars is slightly curved with the center of the curve toward the lingual side. The occlusal surface of all molars contains 2 or 3 enclosed depressions, with the most mesially located in the lingual half. The roots on the lingual side are more uniform in appearance and more closely aligned. The 1st, 2nd, and 3rd molar have 5, 4, and 3 roots, respectively. The 3rd molar exhibits an additional, but smaller, 4th root. The majority of the roots are within 2 parallel planes, the lingual and buccal planes from the midsagittal section. The roots exhibit a slight distal inclination α
of 10°–15°. The angle of inclination is measured at the root apex, as the angle between root surface, and the normal to the occlusal surface, as shown in bottom right. According to the theory of Wolpoff [34
], inclinations promote distal drift of the molars. At the age of 4 months, the roots are not straight but exhibit mesial curvature along with increasing distal inclination towards the root apex (Figures –). The mesial root of the 1st and the distal root of the 3rd molar are exceptional cases, as they are centrally located with significant distal and mesial inclinations, respectively. Given such an anatomy, the rate of distal drift can also change due to a change in occlusal forces. Occlusal forces can generate distal force vectors because of the distal inclination of the roots [34
]. Interestingly, root inclination in humans and primates is mesial [8
] and a physiological mesial drift is reported that enables closing of gaps due to development and function [34
Based on the morphology of the tooth-bone complex in rats, it is conceivable that occlusal loads will compress the PDL in the distal root-tooth complex and simultaneously result in PDL tension within the mesial complex. Similar effects exist in orthodontics, where compression sites in the PDL promote resorption and tension sites promote formation, resulting in tooth migration along the dominant force vector [36
]. Cyclic compression and tension of the PDL during mastication could promote bone resorption and formation, respectively. Furthermore, it has been shown that different mechanical demands on the tooth-bone complex [4
], and compression and tension sites related to distal inclination of the roots (), affect microscale structure of alveolar bone and macroscale form of the bony socket. Consequently, bone morphologies in the distal and mesial root-tooth complex of the same root are inherently different as demonstrated in .
Figure 2 Transverse (a) and sagittal section (b) of the distal root of a 1st and the mesial root of a 2nd molar of a right maxilla exhibiting minimal interdental bone imaged with Micro-XCT; the white dotted lines indicate the position of the other section, respectively; (more ...)
Bone of the distal complex contains concave-rounded pits separated by narrow sharp ridges, resulting in a rough and pitted surface. This appearance is characteristic for distal bone and originates from the osteoclastic resorption activity also demonstrated in SEM micrographs () and histological images (i.e., Figures and ) of individual pits approximately 50 to 100μ
m in diameter [37
]. A reconstruction of the bony socket from the same Micro-XCT scan is illustrated in . The extent of the osteoclastic activity can be observed as several resorption channels cut through the volume of the bone ( left). Contrastingly, the bone from the mesial root-bone complex exhibits a smooth surface with convex-rounded bony protrusions into the PDL-space. The bony protrusions are separated by recesses or channels (). The channels can be related to blood vessel spaces with red blood cells as shown in the H&E stained sections (). The convex protrusions can be attributed to bone formation as demonstrated in the fluorochrome study ( right). Regardless, the 3D images of bone demonstrates continuity of blood vessels in the PDL to those in bone (). Several histology sections in this study also support this observation in particular, , top right.
Figure 5 (a–c) SEM image of bone from the distal root-bone complex; (b) note: resorption pits (white arrows), blood vessel space (white stars); (c) high magnification of resorption pit. (d–f) SEM image of a root exhibiting heavy root resorption; (more ...)
Figure 6 Histological sections stained with H&E: (a) shows the entire root and indicates the position of the mesial (b) and distal (c) root-bone complex at higher magnifications; rough pitted bone surface on the surface of the distal complex and the regular (more ...)
Figure 7 TRAP-positive cells located exclusively in the distal root-bone complex; insert shows multinucleated cells resorbing bone and secondary cementum, osteoclasts, and odontoclasts, respectively. De = dentin, RB = interradicular bone, PDL = periodontal ligament, (more ...)
Figure 4 Left panel: Attenuation profiles (arbitrary units) from left to right through interradicular bone; bone formation in the mesial root-bone complex (left side) to bone resorption in the distal complex (right side) of the same molar. Attenuation of the “old” (more ...)
The roots exhibit a significant structural anisotropy. The root surface, separated by a 100μ
m thick PDL from the resorbed bone in the distal root-bone complex also exhibits resorption pits (). The 3D images of the distal roots of a second molar in , viewed from different angles illustrates the distribution of resorption pits on the root. Resorption pits on the root are less frequent than on bone. Pits on the root are small, and can be identified predominantly in the coronal region in primary cementum. While the mesial surfaces of the coronal thirds/halves of the roots typically show a regular morphology, several pits were identified in the coronal-distal portion of the roots. The apically located secondary cementum shows an overall high roughness and minor pits on all sides. The rough appearance of the secondary cementum in the apical part of the root has already been reported for rats [38
], and was attributed to increased resorption and formation-related activities stimulated by occlusal loading.
Another remarkable anatomical feature is thinning of interdental bone, as shown in . The distal root of the 1st molar and the mesial root of the 2nd molar share the same PDL-space and are in physical contact. Interdental bone commonly reaches the cervix of the tooth, and is comparable to the interradicular bone between the distal and mesial roots of the 2nd molar. Previous studies reported partial thinning of interdental bone, and resulting root proximity [39
]. Figures and illustrate the apical parts of 2nd and the 3rd molars not separated by bone ( bottom right, ). In general, adequate interdental bone and PDL-space are maintained through combined resorption (distal) and apposition (mesial) related events, accomodating movement of the molars [8
]. Based on our observations and previous reports by others, we suggest different migration rates of the individual molars [40
Apposition and resorption-related events manifest into lower and higher X-ray attenuation profiles in the mesial and distal root-bone complex. Micro-XCT techniques measure X-ray attenuation differences that can be directly related to bone mineral density when calibrated with phantoms of known mineral density [11
]. Furthermore, mineral density differences can be exploited from virtual scans at a spatial resolution equivalent to the magnification at which the specimen was scanned. In this study highly attenuating regions in the specimen appear brighter and are related to higher mineral content. In the sagittal and transversal sections of the Micro-XCT images, we consistently observed darker areas representing lower X-ray attenuation close to the PDL interface in bone of the mesial root-bone complex ( left). The graphs demonstrate that attenuation of bone in the distal complex is generally higher and that the increase of attenuation from PDL to bone is steeper than in the mesial root-bone complex. Lower attenuation in bone is caused by lower degree of mineralization and/or crystallinity and can be related to the earlier stages of modeled bone associated with distal drift. This continuous apposition of bone in the mesial root-bone complex accompanied by resorption of bone in the distal complex, coupled with adaptations in primary and secondary cementum, is necessary to maintain a uniform functional PDL-space and accommodate the hard pellet diet in rats.
Complementing X-ray attenuation profiles, are results from fluorochrome labeling. Despite the cumbersome nature of the fluorochrome labeling technique, which includes injecting the animal periodically with different fluorescent dyes, followed by harvesting, specimen preparation, and imaging, the technique illustrates the dynamic nature of bone indicating potentially loaded areas in both tension and compression [41
]. Fluorochrome labeling is an effective method to study biomineralization-related events [42
]. Fluorochrome dyes form chelate complexes with exposed apatite within a mineralizing tissue. As a result, the fluorescent label demarcates active mineralizing fronts exhibiting mineralized tissue formation at the time of administration. By using alternating dyes and injecting at different timepoints, the stratified growth of bone can be temporally mapped [30
]. shows a section of interradicular bone of a 2nd molar from a 7-week-old specimen. The sequence of green-red-green lines in the mesial root-bone complex demonstrates bone formation. The space between two lines shows that the mineralization front moved approximately 10–30μ
m in the 3-4 days between two injections, which corresponds to a distal drift of approximately 20–30μ
m over the same period [43
]. Coronal bone in the distal complex also shows fluorochrome labeling. Furthermore, the surface appears to be regular and convex. This could indicate that bone apposition/repair occasionally also occurs in the distal root-bone complex. However, in the older 4-month-old specimen, significant bone formation was observed in the mesial complex.
Changes from resorption to apposition activity are the origin of cement lines [41
], shown as basophilic lines in H&E stained sections (). However, remaining distal bone illustrated a pitted surface, and no fluorochrome labeling, indicating resorption activity. Some of the pits exhibited a red lining on the surface, probably indicating local repair. Furthermore, the distal root in , shows regular deposition of predentin in the pulp chamber and minor repair/formation activity on secondary cementum.
Specimen preparation for SEM is more cumbersome than for Micro-XCT. In particular, dehydration, fracturing, and sputtering of the specimen along with imaging under high vacuum can induce several artifacts and affect structural integrity. High-energy electrons can result in disintegration of soft tissue, and the vacuum chamber limits in situ
experiments. However, its spatial resolution and magnification range are superior to the other techniques presented in this study. Hence, SEM measurements were conducted to study bone and root morphology, and in particular, resorption morphology at higher resolution. shows an SEM image of bone from a distal root-bone complex. At higher magnification, resorption pits and blood vessel openings can be identified (Figures -). In , a single pit of less than 50μ
m was observed. Figures – illustrate a case of excessive root resorption on the distal side of a 2nd molar, with a pattern of large and small pits and regular cementum surface. At a higher magnification, larger pits were subdivided into smaller pits with diameters of approximately 50μ
m, (). Inside the larger pits within primary cementum (), typical tubular structure of dentin was observed.
While Micro-XCT, SEM, and fluorochrome staining coupled with fluorescence microscopy identified adaptation of the calcified tissue, the distribution of the cells, proteins, and organic matrix could not be imaged sufficiently due to low-contrast for X-rays and high-energy electrons, respectively. Though highly attenuating stains like phosphotungstic acid, osmium tetraoxide, and gallocyanin-chromalum improve imaging of the PDL fibers and most likely cells [45
], the information that can be gathered from imaging following staining is still very limited compared to conventional histology and immunohistochemistry. Furthermore, the resolving power of the Micro-XCT is another limit. The latter would not pose a problem for SEM, but this method will introduce artifacts in the organic tissue due to specimen preparation, high vacuum, and higher-energy electrons as stated earlier. Hence chemically fixed histological sections were stained conventionally or using immunolabeling techniques. The preparation of the histological specimens is very time consuming, as it requires fixation, chemical processing, embedding, and sectioning. Moreover, it can introduce artifacts like delaminated interfaces and loss of structural integrity. The identification of stained organic matter is limited by the resolution of the analytical instruments: the optical or fluorescent microscopes. Furthermore, it should be noted that the sections prepared are 2D and can lead to misinterpretation of 3D structures, despite the interpolation between serial sections.
Though H&E is a conventional stain, it is of value to the study, as it allows to distinguish basophilic structures that stain blue (nuclei), and eosinophilic structures that stain pink (intra and extracellular proteins), or red (red blood cells). This stain gives good structural contrast, and therefore the H&E stained sagittal section in can be compared to sagittal sections imaged with Micro-XCT (Figures and ). The bone surface in the mesial root-bone complex is regular and convexly rounded. The channels appearing prominently on the Micro-XCT images are also present and feature red blood cells. Thus, blood vessels in the PDL are continuous with those in bone. The mesial root surface is covered with a regular layer of cementum that broadens towards the apex. On the distal side, the layer of cementum is thin. Furthermore, the root exhibits a number of pits in dentin with a narrow pinch through cementum. Bone in the distal root-bone complex exhibits a strongly pitted surface, with basophilic lines (blue lines) around those pits. Other basophilic-rich regions include cementum resorption sites and cement lines in bone. The cement line is a remnant of the reversal from bone resorption to bone formation during a remodeling process. Hence, cement lines allow us to extrapolate beyond remodeling events. Cement lines can be found everywhere in bone. This shows that although resorption is the dominant process, remodeling is partly executed in the distal complex, as shown in the fluorochrome image ( right). The cement line close to the bone surface in the distal root-bone complex, shown in lower grey arrow) could be an example for such local remodeling.
Compared to H&E, TRAP staining is specific to mature osteoclasts and more suitable for investigating adaptation of mineralized tissues [24
]. It has been shown that the secretion of TRAP by osteoclasts correlates with their resorptive behavior [23
], and therefore serves as a selective marker for osteoclastic activity. In our study of the rat periodontium, TRAP-positive cells were almost exclusively observed in the PDL-space of the distal root-bone complex (), predominantly close to or in contact with the bone surface. TRAP-positive cells were also found on the surface of the roots, but consistently in lower numbers. These cells were multinucleated like osteoclasts (insert in ). TRAP-positive multinucleated cells resorbing cementum and dentin were also identified as odontoclasts [48
Specificity, as exhibited by the TRAP stain is also fundamental to immunohistochemistry. It utilizes antibodies to bind to specific antigens of interest. For our study, RANKL and OPN, proteins related to bone remodeling, were chosen. The primary antibody is targeted by a secondary antibody that is bound to a fluorophore. Fluorescence-based immunohistochemistry allowed us to identify the distribution of the desired protein in the tissue. RANKL expression is necessary for differentiation and survival of osteoclasts. An increased number of active osteoclasts is a prerequisite for ongoing resorption due to distal drift. Osteoclastogenesis begins with hematopoietic cells generating mononuclear cells that are stepwise differentiated into mature osteoclasts [41
]. Since the step in which mononuclear cells fuse into polykaryons, coincides with TRAP expression, RANKL [49
] is recognized to play a significant role. An orthodontic study in rats found that compressed PDL promotes expression of RANKL [25
]. In our study, shows that RANKL is upregulated in the compressed PDL of the distal root-bone complex, especially close to the bone-PDL attachment site, as compared to the attachment sites in the mesial complex. The large multinucleated cells in the resorption pits of bone and the root (osteoclasts and odontoclasts) also stained intensely for RANKL ( insert).
Figure 8 Histological sections immunostained with RANKL; the red RANKL stain is dominant in the PDL close to the bone surface of the distal root-bone complex (right image) compared to the mesial complex (left image); insert: note odontoclast on root and osteoclasts (more ...)
OPN supports bone remodeling and is produced by osteoblasts, osteoclasts, and a number of other cells. It belongs to the family of small integrin-binding ligand N-linked glycoproteins (SIBLING). The proposed function of OPN in biomineralization [50
] is threefold. It promotes cell adhesion of osteoclasts and osteoblasts. It regulates osteoclastic resorption and migration, and was shown to inhibit hydroxyapatite crystal growth by binding to its surface. In our study, most intense staining of OPN was found on the interface of bone and PDL in the distal complex (). Its presence is indicated by green lines on the bone surface and green-stained multinucleated cells attached or close to those lines ( right insert). The progression from bright to faint staining in the pits on the root surface on dentin, and on cementum, highlights resorption related events. The bulk cementum illustrated an undefined faint stain. Occasionally, brighter staining was found in cementum, specifically at the primary to secondary cementum transition in . This could be a sign of recent cementum repair as indicated by Jäger et al. [27
]. The progression from bright green to red lines in bone are seemingly related to cement lines, which regularly stain for OPN [28
]. The bone surface of the mesial complex neither exhibits bright green lines on the surface, nor osteoclasts complementing lack of TRAP staining in . However, close to the bone surface, faint lines were consistently observed ( left insert), and could indicate intermittent biomineralization.
Figure 9 Histological sections immunostained with OPN; the green stain of OPN dominates the distal root-bone complex (right image); note staining on resorption pits and multinucleated cells on bone (right insert), resorption pits on the root (white stars), remodeling (more ...)