Healthy, non-smoking patients requiring extraction of the first premolars for orthodontic treatment were recruited from the Postgraduate Clinic, Faculty of Dentistry, Bangkokthonburi University, Thailand, after obtaining written informed consent. The study protocol (#9/2018) was approved by the Bangkokthonburi University Human Ethics Committee. The inclusion criteria were as follows: patients who had sound first premolars with complete root formation in their four quadrants without caries or restorations, and had good oral hygiene and a healthy periodontium. The exclusion criteria were as follows: patients with systemic infections or diseases who required medications, such as nonsteroidal inflammatory drugs, or those with severe crowding at the extraction sites.

Eighteen such patients undergoing orthodontic treatment were randomly divided into two groups for the measurement of PDL volume and determination of protein expression. For the measurement of PDL volume, 32 first premolars from eight patients (mean age, 21.0 ± 3.8 years; six females and two males) were selected. The sample size of at least eight teeth for each period of loading was determined using G*Power software ver. 3.1.9.2 (Franz Faul; University of Kiel, Kiel, Germany), with an effect size = 1.45 derived from the preliminary data, α = 0.05, and 1-β = 0.95. To determine protein expression, 40 first premolars from the ten remaining patients (mean age, 21.4 ± 4.0 years; eight females and two males) were selected. The sample size of at least nine teeth for each period of loading was determined with an effect size = 1.33 derived from the preliminary data, α = 0.05, and 1-β = 0.95.

The orthodontic leveling force was applied using a 0.016-inch continuous archwire, which is an improved superelastic nickel-titanium alloy wire (Sentalloy blue, Sentalloy^®; Tomy International Inc., Tokyo, Japan), inserted into a preadjusted edgewise bracket (Roth prescription; 0.018 × 0.025 inch, Tomy International Inc.) that was bonded on the first premolar and all remaining teeth. The bracket was attached using an indirect bonding technique to allow precise bracket positioning.¹⁷ For the first premolars, the brackets were passively placed to avoid either intrusive or extrusive movements. The wire possesses a shape-memory effect, which enables the generation of a light continuous force at 30–70 g in the oral cavity.¹⁸ The control unloaded premolars were extracted at each patient’s first visit, whereas the experimental loaded premolars were extracted after orthodontic loading for 1, 2, or 4 weeks. Simple exodontia was performed by one oral surgeon, with the patients under local anesthesia induced using 4% articaine and 1:100,000 epinephrine (Septanest SP^®; Septodont Co., Paris, France) administered via an infiltration technique (1.0 mL for the maxillary premolars) or an inferior alveolar nerve block together with lingual and long buccal nerve blocks (1.7 mL for the mandibular premolars). A conventional aspirating dental syringe and a disposable 30-gauge and 21-mm-long needle (Septoject XL^®; Septodont Co.) were used for the infiltration, and a disposable 27-gauge and 30-mm-long needle (Septoject XL^®; Septodont Co.) were used for the nerve blocks. Both experimental and control premolars were atraumatically extracted by gentle separation of the gingiva using a straight elevator no. EL3S (Hu-Friedy Mfg. Co., LLC, Chicago, IL, USA), followed by the use of a no. 151 forceps (Hu-Friedy Presidential^®; Hu-Friedy Mfg. Co., LLC) to avoid injury to the PDL tissue. The extracted premolars were then washed with normal saline to remove blood and tissue debris.

Toluidine blue staining and determination of the remaining PDL volume on the root surface were performed using the protocol of Nakdilok et al.¹⁹ Briefly, the extracted teeth were stained with 0.04% (w/v) toluidine blue (Sigma-Aldrich, St. Louis, MO, USA) and destained with phosphate buffered saline for 14 days. A stereomicroscope (Olympus SZX7; Olympus Corp., Tokyo, Japan) was used to observe all surfaces of the stained roots. A digitized image in a plane perpendicular to the tooth axis was then recorded using a charge-coupled device (Olympus E-330; Olympus Corp.) attached to the stereomicroscope. The ImageJ2 software (National Institutes of Health, Bethesda, MD, USA) was used to quantify the stained images, and the stained area in each image was determined as a percentage of the total area. The percentage of stained PDL was calculated using ImageJ2 by differentiating between the white and blue pixels, which represented the unstained and stained areas, respectively. The percentage of stained PDL of each tooth was then calculated.

The PDL tissue of the extracted premolars was scraped off from 3 mm below the cementoenamel junction to the root apex. Total protein was isolated by grinding the PDL tissue in a glass homogenizer using 350 μL of lysis buffer from a NucleoSpin^® RNA/Protein isolation kit (MACHEREY-NAGEL GmbH & Co. KG, Düren, Germany) and 1% (v/v) β-mercaptoethanol (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The lysate was transferred to a 1.5-mL microcentrifuge tube and stored at –80°C until simultaneous western blot analysis of every sample from the ten patients. The frozen lysates were thawed and their total protein was extracted using the NucleoSpin^® RNA/Protein isolation kit (MACHEREY-NAGEL GmbH & Co. KG) following the manufacturer’s instruction. In brief, the lysates were first centrifuged at 1,100 × g for 1 minute to remove tissue debris by using filter tubes, and the filtrated lysates were mixed with 70% ethanol at an equal volume. The mixtures were centrifuged using the NucleoSpin^® columns to collect their protein fraction from the flow-through solution. Protein in the solution was precipitated and dissolved in a protein solving buffer containing the reducing agent Tris (2-carboxyethyl) phosphine hydrochloride. The quantity of total protein in each sample was determined using ultraviolet absorbance at 280 nm in a NanoDrop^TM 2000 spectrophotometer (Thermo Scientific^TM, Rockford, IL, USA). A 20-μg quantity of total protein was resolved using 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis under an electrical current at 100 V/300 W for 135 minutes, and was transferred to nitrocellulose membranes under an electrical current at 100 mA/300 W for 12 hours. To block non-specific binding, the membranes were incubated for 1 hour in 5% (w/v) non-fat dry milk (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) in 0.1% (v/v) Tween-20/Tris-buffered saline. Subsequently, the membranes were incubated overnight at 4°C with the primary antibodies. On the following day, the membranes were washed and incubated with the secondary antibodies for 1 hour. The primary and secondary antibodies used in this study are summarized in Table 1. After washing, the membranes were reacted with LumiGLO Reserve^TM Chemiluminescent reagent (KPL, Gaithersburg, MD, USA). Chemiluminescent signals were captured using the ChemiDoc XRS gel documentation system (Bio-Rad Laboratories, Inc.). The intensities of RUNX2, ALP, RANKL, and OPG bands at their predicted size were measured using ImageJ2 and normalized by that of the beta actin band in each sample. Thereafter, the normalized intensities of these proteins were compared between the experimental and control groups. In addition, the relative ratios of normalized RANKL to OPG were determined and compared between the experimental and control groups.

Data were analyzed using IBM SPSS Statistics for Windows, version 19.0 (IBM Corp., Armonk, NY, USA). Comparisons between the percentages of overall stained PDL tissue at different time points, significant differences between protein expressions, and the RANKL/OPG ratios at different orthodontic loading durations were tested using the Friedman test, followed by the Wilcoxon signed-rank test. The differences were considered statistically significant if p-values were less than 0.05. Spearman’s method was used to analyze correlations between the percentages of stained PDL, protein expressions, or the RANKL/OPG ratios and increased loading durations.

Larger areas of stained PDL were observed on the root surface of the loaded teeth, particularly those loaded for 2 and 4 weeks, than on those of the control unloaded teeth (Figure 1). The median percentages of stained PDL were significantly greater in the teeth loaded for 2 and 4 weeks than in the control unloaded teeth and in the teeth loaded for 1 week (p < 0.05; Figure 2). In addition, a strong positive correlation was found between the percentage of stained PDL and increased preloading durations (r = 0.608, p < 0.001).

The expressions of RUNX2 and ALP were more enhanced in the premolars loaded for 1, 2, and 4 weeks than in the control unloaded teeth, whereas those of RANKL and OPG varied among the unloaded and loaded teeth over different preloading durations (Figure 3). The expression of beta actin was approximately equal among the different samples within each patient. Densitometry revealed that the normalized median expressions of RUNX2 and ALP were significantly greater in the premolars loaded for 2 and 4 weeks than in the control unloaded premolars (Figure 4A and 4B). Moreover, moderately positive correlations were found between the normalized expression of RUNX2 or ALP and increased preloading durations (r = 0.384, p = 0.015 and r = 0.403, p = 0.010, respectively). By contrast, no difference was observed in the normalized median expression of RANKL or OPG between the loaded and unloaded premolars (data not shown). However, the median RANKL/OPG ratios were significantly higher in the premolars loaded for 1 and 4 weeks than in the unloaded premolars (p < 0.05; Figure 4C). Furthermore, the RANKL/OPG ratio positively correlated with increased preloading durations (r = 0.329, p = 0.038).