Fabrication of GNPs+i-PRF
The protocol of GNPs preparation was described elsewhere [29]. Briefly, GNPs were fabricated by a two-step desolvation method using gelatin (from porcine skin, 300 Bloom), deionized water and acetone (Sigma–Aldrich, Sydney, Australia).
In addition, i-PRF was obtained by centrifugating the whole blood from beagle dogs at 700 rpm for 3 min using a duo-Centrifuge (Process for i-PRF, Nice, France) [30].
Finally, with the help of two injection syringes and a luer connector, GNPs with the concentration of 12, 15 and 20 w/v% were mixed with i-PRF to obtain the GNPs+i-PRF gel.
Characterization
Mechanical evaluation
The uniaxial compression test was firstly performed on the GNPs+i-PRF gels (12, 15 and 20 w/v% GNPs) with the size of 2.5 mm (height) x 5 mm (diameter) using a universal testing machine (MTS, USA) at a strain rate of 4 mm/s to analyze their mechanical properties respectively. An iron prop was then used to test on the mechanical property of 20 w/v% GNPs+i-PRF gel.
Hemostatic test
The whole blood clotting time (WBCT) was referred to the reported study but with a slight modification [31]. DBBM, DBBM+i-PRF, GNPs and GNPs-i-PRF (10 mg) were placed in a centrifuge tube and pre-warmed at 37 ℃ respectively. The blood (1 mL) extracted from beagle dogs was added and incubated for 3 min. The whole blood without any additional material was taken as the Blank group. The WBCT was correspondingly recorded. All groups were measured in triplicate.
Animals related procedures
Animals
Detailed protocol of this experiment was approved by the Animal Ethics Committee of Chongqing Medical University (CQHS-IRB-2018-07) and was carried out according to the ARRIVE guidelines [32]. Six adult male beagle dogs (mean age of 1 year old, mean weight of 10~15 kg each), specifically applied for experiment objective, were checked regarding intact dentition and healthy periodontium. The dogs were attended separately with sufficient food and water, while ambient temperature was controlled at 20-25°C with air humidity at 60%-70%. Accordingly, soft diet was adopted postoperatively on all animals.
Surgical protocol
All surgical procedures were administered under generalized anesthesia induced by ketamine/xylazine and kept in a sterile operating room. In addition, the dogs were performed under local anesthesia with a lidocaine injection (20 mg/kg; Huons, Sungnam, Korea). The mesial roots of left and right premolars and second molar (PM3, PM4 and M2) were hemisected bilaterally with fissure burs while cooling system was adopted simultaneously. Mesial roots were extracted atraumatically, during which all buccal bone walls were kept intact and granulation tissue was removed completely. The distal roots were retained and conducted with root canal therapy (Fig. 2 Aa). The extracted sites were then divided into the following experimental groups:
- Blank group (no grafting material applied)
- DBBM group
- DBBM+i-PRF group
- GNPs group
- GNPs+i-PRF group
The above five groups were allocated rotationally at the six extracted sockets in six animals so as to be evenly distributed; therein, the remnant six sites were prepared for pre-emergency in case of a failed extraction. Detailed allocation was listed in Table S1. All grafting materials were placed at the level of the bone crest (Fig. 2 Ab). An absorbable collagen membrane (Bio-Gide®, Geistlich Pharma AG, Wolhusen, Switzerland) was applied to cover the grafted area. Primary closure was performed using one horizontal mattress suture (Vicryl 5-0, Ethicon, MA, USA) and 2-3 single interrupted sutures per site (Fig. 2 Ac).
Post-surgical protocol
The dogs were arranged on a temperature-controlled pad after surgery and left to be observed for any abnormal signs until they could move freely. Post-operative analgesic (buprenorphine 0.05 mg/kg) was administered 2 times per day for 2 days, after which the level of comfort of animals was assessed and analgesia was provided as necessary. All the beagle dogs were fed with soft-pellet diet after surgery to prevent any trauma to the surgical area. Postoperative infection was controlled with penicillin (20 mg/kg/s.c./SID, Kurgan; Normon, Spain) for 3 days. An ultrasoft toothbrush with 0.12% chlorhexidine gluconate was used daily in the first two weeks to keep the oral environment sanitary and healthy.
Sacrifice and sample collection
At the endpoints of 2 (dog 1, 2, 3) and 8 (dog 4, 5, 6) weeks, three beagle dogs at each time-points were painlessly sacrificed under general anesthesia by overdose via intravenous injections of sodium pentobarbital and an overdose of potassium chloride through the carotid artery. The sample was subsequently collected.
Radiography evaluation
Micro-CT analysis
Micro-CT analysis was performed at 2 and 8 weeks postoperation. The micro-CT measurements of all the specimens were achieved applying the advisable equipment (vivaCT80, SCANCO Medical AG, Switzerland) after the proper placement of the tissue blocks. The exposure conditions were as follows: 240° rotation, 15.4-μm-thick aluminum filter, 70 kV, 112 μA. The reconstructed image data was saved to create 3D graphics and analyzed using the μCT80 (SCANCO Medical AG, Switzerland). The threshold, 220 ~ 400, was chosen to segment out newly formed bone and a color map indicating the density of bone mass was subsequently drawn.
The remaining distal roots of PM3, PM4 and M2 were used as retained tooth sites for comparing the linear alterations and healing processes with the extracted sites regarding Micro-CT calculation. Firstly, the buccolingual sections of all extracted groups were superimposed on those of the retained teeth. In this regard, for particular bone width measurement, two reference lines perpendicular to the long axis of the tooth were drawn to cut the socket into three equal intervals to observe changes in each part (Fig. 2 Ba). Among three parts (coronal, middle, and apical), the buccolingual bone widths of the retained and the extracted sites in the center of each part were measured (in millimetres). Accordingly, the width of the retained site was subtracted from that of the extracted site to establish the difference between them. For bone height analysis, the midbuccal and midlingual bone height of the retained site was subtracted from that of the extracted site (Fig. 2 Bb).
Histologic evaluation
Histologic processing
The harvested specimens were fixed in 4% paraformaldehyde (PFA) for 2 days. The specimens were subsequently demineralized in a 0.5 M ethylenediaminetetraacetic acid disodium salt solution (EDTA). After the process of demineralization, all specimens were dehydrated in a graded alcohol series. Afterwards, the specimens were serially sectioned parallel with the long axis of the extraction site to obtain 6-μm-thick sections, which were stained with hematoxylin-eosin(H&E) staining to evaluate early vascularization and later osteogenesis within the socket [33]; Aniline blue (AB) was used to further characterize new bone formation at the early time-point [34]. Moreover, the osteoclast activity at 2 weeks postoperation was observed using tartrate-resistant acid phosphatase (TRAP) staining [35]. The histologic analysis was performed with an incandescent light microscope (DP72; Olympus, Tokyo, Japan) and ImageJ 8.0 software (NIH, Bethesda, MD, USA). In this process, three consecutive sections were prepared the histomorphometric analysis.
Histomorphometric analysis
The histomorphometric calculation was carried out utilizing a bright-field microscopy (Olympus Research System Microscope BX51, Olympus, Tokyo, Japan) on the tissue sections stained with H&E, aniline blue and TRAP. Regarding the stained images, the following fraction of the most central area within the socket were observed and calculated respectively:
※ Vascularized area (%): area of red blood cells surrounded by a layer of epithelial cells is defined as a blood vessel using the H&E staining sections.
※ New osteoid formation area (%): area of blue-stained non-collagenous mineralized tissue area including osteocytes is taken as the new osteoid using the aniline blue sections.
※ Osteoclast area (%): area of neutral-red stained TRAP-positive cells was taken as the osteoclast using the TRAP staining sections.
Statistical analysis
Mean, median, standard deviation were calculated to describe each of the continuous variables respectively. Counts and percentages were applied to describe categorically scaled variables. SPSS Statistics version 20.0 (IBM Corp., Armonk, NY.) and Graphpad Prism Version 7 (Graphpad Software, USA) were adopted to test the normality of the data distribution using the one-sample Kolmogorov-Smirnov test and the significant differences between GNPs+i-PRF group and other groups were measured using one-way ANOVA followed by Student-Newman-Keuls correction for post hoc comparisons. A p < 0.05 was determined as significant.