Animals

Fifty-three 7-week-old female Wistar albino rats were used with a mean initial body weight of 173.7 ± 0.9 g (mean ± SEM; n = 53). The rats were randomly divided into one control group (CG) and two experimental groups, that is, a loading orthodontic force group (LG) and a removed orthodontic force group (RG). Furthermore, the LG was divided into 3d- (n = 6), 7d- (n = 6), 10d- (n = 7), and 14d-LG (n = 5) according to days of force application. The RG was divided into 7d- (n = 5), 14d- (n = 6), 21d- (n = 5), and 28d- RG according to days after force removal (n = 7) (Figure 1). Rats (7-week-old) without application of orthodontic force were used as the CG (n = 6). The response properties of PMRs are stable in 7- to 13-week-old rats.17 All animals were fed ad libitum with powder diet (Rodent Diet CE-2; Japan Clea Inc, Shizuoka, Japan) and had free access to drinking water. All procedures were performed under the guidelines of the Tokyo Medical and Dental University for Animal Research.

View Figure

• Download as Powerpoint
• Download Figure

Application of orthodontic force

The application of orthodontic force was based on a modified technique described elsewhere.18–20 The rats were anesthetized with ketamine hydrochloride (40 mg/kg, i.p.; Veterinary Ketalar50^®, Sankyo Co, Ltd, Tokyo, Japan) containing 20% xylazine hydrochloride (3.5 mg/kg, i.p.; Celactal^® 2% injections, Bayer-Japan Co, Ltd, Tokyo, Japan), after initial inhalant anesthesia with diethyl ether (Wako Pure Chemical Industries, Osaka, Japan).

One end of a closed coil spring (lumen: 0.9 mm) fabricated from superelastic titanium-nickel alloy wire (diameter: 0.093 mm; Furukawa Electric Co, Ltd, Tokyo, Japan) connected with a clamp was inserted into the cervical region of the right mandibular first molar (M1) and fixed with light curing composite resin (Clearfil^® Photo SC, Kuraray, Okayama, Japan). To confirm the fixation of the spring, the mandibular incisors were notched on the labial surface of the cervical region using a round steel bur (Meisinger ST1HP 008, Hager and Meing GmbH, Dusseldolf, Germany) with a dental drill, and the incisors were ligated with ligature wire (diameter: 0.25 mm; Tomy International, Tokyo, Japan) to the other end of the spring. The spring was further fixed with adhesive materials and composite resin. With this model, a mesially directed continuous force of approximately 24.5 mN¹³ was subjected to M1. After 14 days of application of orthodontic force, the springs of the RG were removed under inhalant anesthesia with diethyl ether.

Histological observation

Mechanical stimulation was limited to the distal apical one-third of the distal root of M1 (Figure 2a through c), where PMRs abundantly exist.2122 We observed the histology of the periodontal tissue of rats under the same conditions of the CG, 14d-LG, and 7d-RG. The rats were sacrificed under ether anesthesia, and the right half of the mandible of each rat was removed en bloc. The specimens were fixed with 4% paraformaldehyde in 0.1 M cacodylic acid buffer (pH 7.3) for three days (4°C). When the fixation was completed, the appliances of the 14d-LG were removed. The specimens were decalcified in 10% ethylenediamine-tetraacetic acid solution at 4°C for 35 days and embedded in paraffin by a conventional method. Sections (5 µm thick) sagittal to the long axis of the distal root of M1 were stained with hematoxylin and eosin and examined under light microscope (MICROPHOT-FXA, Nikon, Tokyo, Japan).

View Figure

• Download as Powerpoint
• Download Figure

In vitro jaw-nerve preparation and experimental setup

Under deep anesthesia using thiamylal sodium (60 mg/kg, i.p.; Isozol^®, Yoshitomi Pharmacy, Osaka, Japan), an in vitro jaw-nerve preparation was made as previously reported.1516 To apply direct mechanical stimulation to PMRs, three mandibular molars on the right side were extracted from the mandible. The chamber consisted of two pools (test and oil pools) separated by a thin plastic plate with a small hole drilled in its center. The inferior alveolar nerve trunk was passed through the hole and fixed in the oil pool by the cotton thread. The nerve trunk in the oil pool was slightly pulled to furnish enough tension for easy insertion of the recording electrode. The contents of the two pools were separated by filling the hole surrounding the nerve trunk with vaseline. The oil pool was filled with liquid paraffin to avoid dehydration of the nerve trunk. The mandible was placed on the rubber bed in the test pool and was fixed to the rubber bed with dental cement (GC Fuji I^®, GC Corporation, Tokyo, Japan). The test pool was perfused by a modified Krebs-Henseleit solution (107.4 mM NaCl, 3.4 mM KCl, 1.5 mM CaCl₂, 0.7 mM MgSO₄, 2.2 mM NaH₂PO₄, 26.2 mM NaHCO₃, 9.6 mM sodium gluconate, 5.5 mM glucose, 7.6 mM sucrose) saturated with a O₂–CO₂ (95:5%) gas mixture.

Stimulation and recording

For mechanical stimulation, eight intensities (2.61, 4.74, 7.35, 8.39, 9.41, 10.62, 13.79, and 15.84 mN) of calibrated plastic von Frey hairs (tip diameter: 0.2 mm) were manually applied (duration: 5–10 seconds) to the PMRs remaining in the stimulating site. Using the microneurographic method,23 single-unit activities were recorded from the nerve trunk with a tungsten microelectrode (Type: 25-10-1, FHC, Brunswick USA) inserted into the nerve trunk.

The signal was fed into a high-impedance, low-noise amplifier (DAM80-E, World Precision Instruments, Sarasota, Fla). The signal was digitized and displayed on the monitor with a data analyzing software (Spike2^® for Windows^® Version 4.02, Cambridge Electronic Design, Cambridge, UK), using a signal processing interface (CED 1401 plus, Cambridge Electronic Design, Cambridge, UK) and a DOS-V computer.

The units were divided into two types according to their response pattern: those that discharged continuously for more than five seconds while mechanical stimulation was applied were classified as a slowly adapting (SA) type, and the other units were classified as a rapidly adapting (RA) type.

Conduction velocity of the innervating fibers

To estimate the conduction velocity of the recorded single afferent fiber, electrical stimulation was applied to the tooth sockets of the distal root of M1 by a bipolar concentric tungsten electrode (tip diameter: 0.4 mm, tip distance: 1 mm; IMB-9004, Inter Medical, Nagoya, Japan) and the action potential was displayed on an oscilloscope (CS-6030, KENWOOD, Kanagawa, Japan). The conduction velocity was calculated from both the conduction time and the distance between the stimulating and recording electrodes and was corrected to the value at 37°C, using the Q₁₀ correlation reported by Paintal.24 Fibers with conduction velocity of less than 2.0 m/s were regarded as unmyelinated (C), those between 2.0 and 10.0 m/s as thin myelinated (Aδ), and those over 10.0 m/s as large myelinated (Aβ) fibers.25 In this study, only mechanoreceptors innervated by Aβ fibers were investigated further.

Statistical analysis

For statistical analysis, the data-analyzing software (StatView^® for Windows^®, Version 5.0, SAS Institute, Cary, NC, USA) was used. All data are expressed as mean ± SEM. The statistical differences between the CG and each experimental group were evaluated by the Mann-Whitney's U-test. A probability of less than .05 was considered significant.

Histological finding

In the CG, the periodontal collagen fibers ran wavy between cementum and alveolar bone. In the 14d-LG, the periodontal collagen fibers were irregularly arranged, and the thickness of the periodontal ligament tended to increase compared with the CG. However, in the 7d-RG, the periodontal collagen fibers showed almost the same arrangement, and the thickness of the periodontal ligament was almost equal to the CG. Excessive root resorption or inflammatory reactions were not detectable in any group (Figure 2).

Unit type

Responses to the mechanical stimulation were recorded from a total of 248 units. The units were classified into 244 RA and four SA units (Table 1). The RA units were classified into three subtypes: those that discharged at the beginning of continuous mechanical stimulation were classified as RA-on, those that discharged at the end as RA-off, and those that discharged at both the beginning and the end as RA-on/off type. The typical examples of each type are shown in Figure 3.

TABLE 1. Distribution of Subtypes of Units

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Mechanical threshold

Chronological changes were found in the threshold to mechanical stimulation during and after experimental orthodontic tooth movement. Table 2 shows the mean values of mechanical thresholds for each group. Figure 4 shows the chronological changes of mechanical thresholds. Groups 3d-, 7d-, 10d-, and 14d-LG had significantly lower threshold values compared with the CG. In contrast, there was no significant difference in the 7d-, 14d-, 21d-, and 28d-RG.

TABLE 2. Mean Values of the Mechanical Threshold and the Conduction Velocity^a

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Conduction velocity of innervating fibers

The conduction velocity of 248 mechanoreceptive units was distributed between 10.20 and 21.15 m/s. Table 2 shows the mean values of the conduction velocities for each group. Figure 5 shows the chronological changes in conduction velocity. Compared with the CG, significantly lower conduction velocities were found in the 7d-, 10d-, and 14d-LG and in the 7d- and 14d-RG. No significant difference was found in the 21d- and 28d-RG.

View Figure

• Download as Powerpoint
• Download Figure

Histological finding

The periodontal ligament incessantly receives the occlusal forces, and active tissue remodeling constantly occurs.26 The periodontal ligament was abundantly supplied by two kinds of receptors, Ruffini-like endings and free nerve endings,21 which have their structures changed in response to external stimuli like orthodontic force or occlusal trauma.2728 PMRs also have a high potential for neural plasticity.2728 The active remodeling of the periodontal ligament and the structural alteration of PMRs are positively related to each other.2729 Our histological findings suggest that structural alterations of PMRs took place. Moreover, because typical figurations to traumatic tissue,3031 excessive infiltration of inflammatory cells, or wide range of root resorption were not observed, occlusal trauma has no influence under these conditions. Thus, the orthodontic force was applied without triggering degenerative alterations.

Unit type

Our result showed that the RA type was the most frequently found unit type (Table 1); this result agrees with previous reports under conditions of normal occlusion.16 In contrast, the previous study in which PMRs were stimulated indirectly has reported on the different proportion of RA and SA types.3233 Cash and Linden34 hypothesized that the adaptation property depended on the stimulating method. Receptors being stimulated over a wide area such as a tooth would respond as SA; on the contrary, those being stimulated directly with the tip of fine von Frey hairs would respond as RA.34 Thus, the discrepancy between our results and those of the authors regarding unit type proportion may be related to different methodology.

Mechanical threshold and conduction velocity

As shown in Table 2 and Figure 4, the thresholds in the LG were significantly lower compared with the CG, as previously reported.13 In contrast, the thresholds in the RG showed no significant difference compared with the CG. One possible explanation is that the structural alteration of PMRs by orthodontic force could have been responsible for the changes in the mechanical thresholds. Histological study using growth-associated protein-43 (GAP-43), which is involved in neural development and regeneration, has shown that the Schwann cells, which are thought to perceive distortion of periodontal collagen fibers,35 exhibited GAP-43-like immunoreactivity during experimental orthodontic tooth movement.27 Therefore, the chronological changes of the mechanical threshold seem to be a functional alteration in accordance with the structural alteration of PMRs by the orthodontic force.

Along with mechanical thresholds, chronological changes were also found in the conduction velocity of the innervating fibers (see Figure 5; and Table 2). During experimental orthodontic tooth movement not only Schwann cells but also axon terminals show GAP-43–like immunoreactivity.27 Thus, the afferent fibers innervating the PMRs also suffered changes after the morphological or functional alterations of PMRs resulting in lower conduction velocities.

This study revealed that different recovery periods were required for PMRs (mechanical threshold) and their innervating fibers (conduction velocity). These discrepancies may be due to the time lag between the morphological or functional alterations of PMRs and the characteristic alteration of afferent fibers. This idea can also be backed up by other results showing various changes in the localization pattern of GAP-43-like immunoreactivity between Schwann cells and axon terminals during experimental orthodontic tooth movement.27

After seven days of force removal, the periodontal collagen fiber arrangement became similar to the CG, and mechanical threshold also showed no significant difference to the CG. This result suggests that the adaptation of PMRs, due to their high potential for neural plasticity,2728 took place early within seven days along with the alteration of the periodontal environment. Because the recovery of the conduction velocity took a longer time, the adaptation of the afferent nerve fibers may require a longer period than that of PMRs. However, afferent nerve fibers also adapted to the new periodontal environment.

Functional adaptation after removal of orthodontic force

The present results showed that the response properties of PMRs adapted to the newly acquired tooth position. Thus, the alteration observed during experimental orthodontic tooth movement is considered recoverable. However, there still is a controversy between the present result and other reports showing an alteration of morphological or physiological characteristic of PMRs during experimental orthodontic tooth movement, without any recovery after a long period.1436 This discrepancy may be due to different experimental conditions, where the opposed tooth of observation site were extracted and relatively strong magnitude of orthodontic force was applied (2000 mN).37 The mechanical stimuli from the occlusal force is considered an important factor to maintain distribution and morphology of periodontal Ruffini-like endings,17272938 and occlusal hypofunction may affect the response properties of PMRs.39 Furthermore, the stronger the magnitude of orthodontic force, the larger the site of pathological degeneration of periodontal tissue.2040 The root surface of M1 has an area that is about 20 times smaller than that of a human first molar.37 The orthodontic force of 24.5 mN, used in this study, was almost equal to the clinical low orthodontic force of 490 mN. Unlike the other studies, we used a low orthodontic force of 24.5 mN together with the occlusal force. Therefore, a faster recovery of the response properties of PMRs results.