Material and methods

The Institutional Animal Care Committee of the University of Connecticut Health Center approved this study, which conformed to the ARRIVE guidelines. Data were obtained from 64 male CD1 mice (Charles River Laboratories, Wilmington, Mass; body weight, 24-30 g). The 12-week-old mice were randomly divided into 2 groups: control and experimental. The control group was further subdivided into 4 groups; each control group had 5 mice: (1) group 1 (baseline), no spring and no mechanical vibration; (2) group 2, no orthodontic spring, but 5 Hz vibration was applied to the maxillary first molars; (3) group 3, no orthodontic spring, but 10 Hz vibration was applied to the maxillary first molars; and (4) group 4, no orthodontic spring, but 20 Hz vibration was applied to the maxillary first molars. The experimental group was also subdivided into 4 groups, each with 11 mice: (1) group 5, orthodontic spring only but no vibration; (2) group 6, orthodontic spring and 5 Hz vibration applied to the maxillary first molars; (3) group 7, orthodontic spring and 10 Hz vibration applied to the maxillary first molars; (4) group 8, orthodontic spring and 20 Hz vibration applied to the maxillary first molars. The animals were allowed at least a week of acclimatization at our Health Center to compensate for their different origins.

The animals were placed under general anesthesia with xylazine (13 mg/kg) and ketamine (87 mg/kg). A custom mouth-prop was fabricated from 0.036-in stainless steel wire and placed between the maxillary and mandibular incisors to hold the mouth open. A 0.004-in stainless steel ligature wire was passed beneath the contact between the maxillary first and second molars and threaded to the maxillary first molar ( Fig 1 , A ). A low force/deflection rate nickel-titanium coil spring (Ultimate Wireforms, Bristol, Conn) delivering 10 g of force was attached to the 0.004-in stainless steel ligature around the first molar, and the other end of the spring was attached to the incisors with 0.004-in stainless steel wire. The force/deflection rate for the spring was determined to calibrate the amount of force produced by activation of the nickel-titanium coil spring.

A, Inserted nickel-titanium coil spring; B, application of LFMV on the maxillary right molar (the electromechanical actuator applies mechanical vibration at 5, 10, and 20 Hz at a compression force of 1 g); C, schematic of the tooth movement model with the spring attached to the maxillary first molar and maxillary incisors; D, intermolar distance ( M1-M2 ) at day 14 in the experimental groups; E – H, sagittal micro-CT sections used for measuring intermolar distance in the control groups ( E, baseline; F, no spring + 5 Hz vibration; G, no spring + 10 Hz vibration; H, no spring + 20 Hz vibration); I – L, sagittal micro-CT sections used for measuring intermolar distance in the experimental groups ( I, spring + no vibration; J, spring + 5 Hz vibration; K, spring + 10 Hz vibration; L, spring + 20 Hz vibration).

Additionally, grooves 0.5 mm from the gingival margin were prepared on the facial, mesial, and distal surfaces of the maxillary central incisors to prevent the ligatures from dislodging from the incisor because of their lingual curvature and eruption pattern. Self-etching primer (Transbond Plus; 3M Unitek, Monrovia, Calif) and light-cured adhesive resin cement (Transbond; 3M Unitek) were applied to the lingual surfaces of the maxillary first molars and incisors to secure the ligature wire. Moreover, to minimize the distal movement of the right incisor and to reinforce the anterior anchorage, the right and left incisors were joined together to act as a unit ( Fig 1 , A ). After appliance insertion, the mice were allowed to recover with an incandescent light for warmth; they were returned to their cages once full ambulation and self-cleansing returned. The appliances were checked every other day, and additional light-cured bonding material was added if necessary.

After anesthesia with ketamine (87 mg/kg) and xylazine (13 mg/kg), a custom mouth-prop fabricated from 0.017 × 0.025-in titanium-molybdenum alloy wire was placed between the maxillary and mandibular incisors to hold the mouth open. A feedback loop–controlled electromechanical actuator (model 3230; Bose Enduratec, Minnetonka, Minn) was used to apply unilateral LFMV to the occlusal surface of the maxillary right first molar along its long axis ( Fig 1 , B ). Loading protocols for each animal consisted of 15 minutes of LFMV at 1 cN of force with the electromechanical actuator at a frequency of 5, 10, or 20 Hz (cycles/second) depending on the mouse’s group. LFMV was applied to the maxillary right first molar at 3-day intervals (days 1, 4, 7, 10, and 13).

After the 14 days of the experimental period, the animals were killed by inhalation of carbon dioxide followed by cervical dislocation.

The maxilla was hemisected, and the attached soft tissues and muscles were removed. Subsequently, the hemisected maxilla was placed in 10% formalin for 5 days at 4°C. The samples were then decalcified in 14% ethylenediaminetetraacetic acid for 3 weeks and then processed for standard paraffin embedding. Serial sagittal sections (5 μm) were obtained from the mesial and distal roots of the maxillary right first molar.

All the animals in the different control (5 in each group) and experimental (11 in each group) groups were used to measure tooth movements. Three-dimensional image arrays of the hemisected right maxillae were collected using micro-CT. OTM was defined and measured as the distance between the maxillary first and second molars at the most mesial point of the second molar crown and the most distal point of the first molar crown. The distance at day 0 was 0 mm in all groups (ie, the convex crown surfaces were touching). The OTM measurements were done in the sagittal plane, locating the image plane that showed the most root structure. The original 2-dimensional image was then magnified 10 times for more precise line drawings, which were made at the closest proximity of the 2 convex molar crown surfaces. The OTM was measured in the 3 sagittal sections of each animal in all groups.

Micro-CT images were used for quantitative analyses of bone changes occurring in the region of the maxillary first molar. Changes in the alveolar bone were studied by analyzing the furcation area of the maxillary first molar. The region of interest for the alveolar bone analysis was defined vertically as the most occlusal point of the furcation to the apex of the maxillary first molar roots; transversely, it was defined as the space between the buccal and lingual cortical bone; sagittally, it included 50 sections (10 μm) beginning at the mesial root and continuing distally ( Fig 2 , A ). The parameters studied and measured using the established algorithms were bone volume fraction (BVF) and tissue density.

Micro-CT data showing BVF and tissue density. Each value in each graph represents the mean ± the standard deviation: A, region of interest where the bone parameters (BVF and tissue density) were measured; B, BVF values in the control groups (significantly [*] higher BVF in the no spring + 5 Hz vibration group compared with the no spring + 20 Hz vibration group); C, BVF values in the experimental groups; BVF decreases with tooth movement; baseline had a significantly higher BVF compared with the spring + no vibration group, the spring + 5 Hz vibration group, the spring + 10 Hz vibration group, and the spring + 20 Hz vibration group; moreover, the spring + no vibration group had significantly less BVF than did the spring + 5 Hz vibration (*) and the spring + 10 Hz vibration (#) groups. D, Tissue density in the control groups; note the trend for an increase in tissue density with vibration; the spring + 5 Hz group had significantly higher bone density compared with baseline (*); a,b,c,d signifies P < 0.5. E, tissue density in the experimental groups.

Histology, immunohistochemistry, tartrate-resistant acid phosphatase (TRAP), and picrosirius red staining histology were performed on each section. After blocking for 10 minutes at room temperature with 1x universal blocking reagent (HK085-5k; BioGenex, Fremont, Calif), deparaffinized histologic sections were incubated with goat polyclonal anti-SOST antibody (AF1589; R&D Systems, Minneapolis, Minn) overnight at +40°C at a concentration of 5 μg per milliliter. Subsequently, the sections were washed with phosphate-buffered saline and incubated with Alexa Fluor 594 donkey anti-goat IgG (A-11058; Life Technologies, Grand Island, NY) at a concentration of 1:300 for 1 hour at room temperature and mounted with a suspension of 50% glycerol in phosphate-buffered saline containing the nuclear stain (Hoechst H3570; Life Technologies) at a concentration of 1:5000.

TRAP staining was performed using a leukocyte acid phosphatase (TRAP) kit (386-1 KT; Sigma-Aldrich, St Louis, Mo) according to the manufacturer’s instructions. TRAP positive, multinucleated cells were counted on the alveolar bone surfaces on the mesial sides of the distobuccal roots. The area for quantification included a square with 1 side extending from the apex to the bifurcation, and the other side extending 200 μm from the PDL border inside the alveolar bone. The osteoclast numbers were counted in 6 sections from 4 mice in each group, and the values were then averaged for each animal to run a statistical test.

Deparaffinized sections were also stained with 1% picrosirius red for 1 hour, washed with acidified water (0.5% acetic acid water), and dehydrated with serial ethanol washes before mounting. The sections were examined with a fluorescent microscope (Carl Zeiss, Thornwood, NY).