2.1

Reduced chewing loads in rats, an animal model [ , ]

All experimental protocols were compliant and followed the guidelines of the Institutional Animal Care and Use Committee (IACUC). Male Sprague Dawley rats (N = 60; Charles River Laboratories, Inc., Willmington, MA, USA) at four weeks of age were divided into two groups and were fed one of two nutritionally equivalent foods: hard pellet food (Hard Food = HF; N = 30) or soft powdered chow (Soft Food = SF; N = 30) (PicoLab 5058, LabDiet, Deans Animal Feeds, Redwood City, CA, USA); foods differed only in hardness (i.e., the softer food was a powdered version of the harder food) [ , , ]. Hemimandibles were dissected and the right hemimandible was saved for in situ loading, micro X-ray computed tomography, and nanoindentation, while the left hemimandible was processed for histology. Adaptations as related to biochemical and material property changes as an effect of reduced chewing loads were recorded at 8, 12, 16, 20, and 24 weeks.

2.2

Biomechanical testing in situ and micro X-ray tomography (micro XCT) [ , , ]

Preparation of hemimandibles for biomechanical testing under wet conditions constituted loading the DAJ in situ by using validated loading schemes while visualizing using a micro X-ray computed tomography system [ , , ]. Prepared specimens were loaded under wet conditions using a load compression cell (MT200CT, Deben UK Ltd, Suffolk, UK) custom-fitted to a micro-CT (Micro XCT-200, Carl Zeiss X-ray Microscopy, Pleasanton, CA, USA). Each specimen was loaded to each permutation of a peak load (5 N, 6 N, 8 N, 10 N, and 15 N) and a displacement rate (0.2, 0.5, 1.0, 1.5, 2.0 mm/min) [ ]. Two separate specimens (one 8-week HF and one 8-week SF) were prepared for visualization of tooth position relative to the socket surface, and consequently DAJ biomechanics in situ (Supplemental Figure S1) and digital volume correlation (DVC) was subsequently performed to generate maximum and minimum principal strains equivalent to tensile and compressive strain profiles within interradicular alveolar bone [ , , ].

2.3

Evaluation of functional space of a loaded DAJ

In this study, PDL-space under load is defined as “functional space”, and at no load is defined as “PDL-space”. Tomograms were analyzed using AVIZO (Avizo 2019.4, Thermo Fisher Scientific, Waltham, MA, USA). Segmentations of tooth and alveolar bone were calculated based on intensity cues resulting from X-ray attenuation and segmented data were used to generate surface meshes. The segmented data were checked manually to confirm accurate extraction of morphological features. The net PDL-space was calculated by subtracting PDL-space under load and from PDL-space under unloaded conditions. The resulting net PDL-space was mapped on the tooth surface.

2.4

Mapping tartrate-resistant acid phosphatase positive (TRAP)(+) regions within a complex [ , ]

Left hemimandibles were fixed, decalcified, and embedded in paraffin as previously described [ , ]. TRAP staining for osteoclastic activity was performed and tissue sections were counterstained using hematoxylin. Individual multinucleated cells in the distal periodontal complex and within the interradicular alveolar bone displaying a TRAP(+) response were counted. Statistically significant differences between HF and SF groups were calculated for each age group using Student’s t-test.

2.5

Nanoindentation and bone volume fraction (BVF) of interradicular bone [ ]

Hemimandibles (N = 3 for each age group from respective SF or HF groups) were prepared for nanoindentation by polishing using silicon carbide grit and diamond slurries of various grades (Buehler Ltd., IL, USA) [ ]. The polished specimens were immobilized for nanoindentation using epoxy and care was taken to avoid its infiltration into the specimen. Nanoindentation (Nanoscan 4D, Nanounity, Moscow, Russia) on polished surfaces was performed under wet conditions using a Berkovich tip. On average, 20–25 indents were made using 2000 μN load and each indent was placed 16 μm apart on the IB and specifically in regions that were in direct association with the second molar. Reduced elastic modulus (Er) [ ] was calculated and elastic modulus by age and by food hardness was skewed, so a nonparametric test of age × food hardness, overall food hardness, overall age, and age-specific food hardness effects was performed with stepdown Šidák tests to correct for multiple comparisons. Statistical analyses were performed with SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).

Bone volume fraction (BVF) was calculated using subvolumes within three-dimensional volumes reconstructed from X-ray tomograms [ , ]. Statistically significant differences between harder and softer foods were calculated for each age group using the Student’s t-test.

2.6

Image processing and digital volume correlation to map maximum and minimum principal strains in interradicular alveolar bone [ , ]

Strain mapping of an interradicular alveolar bone under load was performed on a single specimen from each group (8 weeks only). Piecewise DVC was performed to compare deformations between 0−7 N, 7−15 N, and 15−20 N load configurations (DaVis software, LaVision Inc., Ypsilanti, MI, USA). The specimens were loaded at rate of 2 mm/min. Maximum and minimum principal strains equivalent to tensile and compressive strains within interradicular alveolar bone [ , , ] were evaluated and mapped.

Prior to digitally correlating the volumes of interradicular bones at no load and under load, all X-ray tomograms were filtered, registered, and masked. After reconstruction of the volumes, 3D images were further processed to optimize the calculation of strains by filtering, rigid body registration, cropping, and masking. Firstly, X-ray tomograms were filtered using an anisotropic diffusion algorithm within Avizo (Avizo 8.1.0, FEI, Germany) to remove ‘salt and pepper’ noise while preserving features. Secondly, scans acquired under loaded and unloaded conditions were used to track rigid body movement. Cropping was performed to focus on regions of interest and remove extraneous regions that would add noise (such as the beam hardening of the anvil). Finally, the tooth was removed digitally by masking away the regions of tooth structure from the image space using an intensity segmentation method. In DVC software, correlation between unloaded and loaded scans was performed by setting the interrogation window size at 64 pixels with a 50% overlap and a 25% minimum fraction of valid pixel (mvfp) [ ]. The noise floor for DVC and calculation to obtain maximum and minimum normal strains are shown in Supplemental Figures S2 and S3(a), respectively.

2.7

Statistical modeling and overall significance

Nonlinear quadratic mixed effects regression models with normal random animal × replicate effects were used for each of 125 combinations of age × force × displacement rate to model the relationship of displacement on load (load-displacement curve) between foods of two different hardness values (HF, SF) with a 3-degree-of-freedom (3 d.f.) test of the food hardness difference parameters (intercept, linear slope, and quadratic slope). These models were also used to predict the difference between food hardness in displacement to yield 5 N, 10 N, and 15 N loads as

Stay updated, free dental videos. Join our Telegram channel

Join