Extracted permanent human teeth without caries or restorations were collected, and any soft tissue that remained attached was removed using a rotary instrument with coolant. All steps, including demineralization, washing were processed in a vacuum-ultrasonic device (VacuaSonic System; CosmoBioMedicare, Seoul, Korea). After sterilization of the teeth with sterilization reagent (peracetic acid ethanol solution) in a vacuum-ultrasonic device, the teeth were crushed into powder with particles measuring 0.8 to 1.0 mm in size on the experimental day, and demineralization using 0.6 N of hydrochloride was performed for 15 minutes under vacuum compression and ultrasonic vibration in a vacuum-ultrasonic device. The demineralized particulate tooth bone was washed with phosphate-buffered saline (PBS), sterilized with sterilization reagent, and washed again with PBS and distilled water.

Eight adult male New Zealand white rabbits weighing from 2.8 to 3.2 kg (mean, 3.0 kg) were included in the present study as experimental animals. This study was approved by the Animal Care and Use Committee at the Catholic University Medical Center of Daegu (DCIAFCR-160705-3-Y). Four rabbits were placed in each group, and all rabbits underwent the same surgical procedure under general anesthesia with a combination of 30 mg/kg of ketamine (Ketalar; Yuhan, Seoul, Korea) and 10 mg/kg of xylazine (Rompun; Bayer Korea, Seoul, Korea) administered intramuscularly. Before making an incision on the nasal dorsum, about 0.5 mL of lidocaine with 1:100,000 epinephrine was injected subcutaneously along the midline of the nasal dorsum. Skin and periosteal incisions were created with a 15-c blade at the middle of the nasal dorsum to expose the nasal bone and nasoincisional suture line. A rectangular replaceable bony window, measuring about 5 mm×10 mm, was created with a thin saw insert (S-Saw; Bukbu Dental, Daegu, Korea) connected to a piezoelectric device (Surgybone; Silfradent, Sofia, Italy). Two windows were established at both nasal bones, located about 20 mm anterior to the nasofrontal suture line and 5 mm from the middle suture line. The sinus mucosa was elevated with a blunt-ended curette to avoid membrane perforation and anteroventrally to accommodate the bone graft. In the control group, approximately 0.25 mL of anorganic bovine grafting material (Bio-Oss; Geistlich Pharma, Wolhusen, Switzerland) was grafted in the new compartment under the elevated sinus membrane, and the bony window was replaced after careful elevation of the sinus mucosa. In the experimental group, 0.25 mL of demineralized particulate human tooth bone was grafted in the new compartment under the elevated sinus membrane, and the bony window was repositioned over the bone graft.(Fig. 1) The flap was sutured with a 4-0 nylon suture (Blue nylon; Ailee, Busan, Korea). All animals were administered antibiotics intramuscularly, with 20 mg/kg of gentamycin (Donghwa, Seoul, Korea) given for three days postoperatively.

The rabbits were sacrificed at two and eight weeks under general anesthesia after all intramuscular injections of urethane (Junsei Chemical, Tokyo, Japan). The augmented sinus was segmented with a micro-saw from the cranium and fixed with neutral-buffered formalin for 24 hours, washed with 0.1 M of PBS, and demineralized with 10% formic acid for 10 days. The specimen was embedded in paraffin (Paraplast; Sigma-Aldrich, St. Louis, MO, USA) and sliced coronally into serial sections measuring 5 μm thick. Both augmented sinuses were included in the specimen for comparison. The specimens were stained with H&E and examined under light microscopy for newly formed bone and soft tissue changes in the new compartment of the maxillary sinus.

Serial sections in the control and experimental groups were stained immunohistochemically for markers of proliferating osteoprogenitor cells using BrdU (Sigma-Aldrich). To accomplish BrdU staining, operated rabbits were given 25 mg/kg of BrdU intraperitoneally. At 24 hours after administration of BrdU, the rabbits were sacrificed, and paraffin-embedded sections were created. These sections were incubated in 0.1% trypsin for 10 minutes at 37℃ and 1 N of HCl for 30 minutes at 56℃ to denature the DNA. Endogenous peroxidases were inhibited by preincubation in 0.3% H₂O₂ in PBS for 30 minutes, and nonspecific protein binding was blocked in PBS containing 10% normal horse serum (Vector Laboratories, Burlingame, CA, USA) for 30 minutes. The sections were incubated in monoclonal anti-BrdU (1:100; Sigma-Aldrich) for two hours at room temperature and washed three times with PBS. The secondary antibody (1:100), biotinylated anti-mouse immunoglobulin G (Vector Laboratories), was deposited for one hour at room temperature on the sections, which were washed three times with PBS. Next, avidin–biotin–peroxidase complex (Vector Laboratories) was placed on the sections for one hour and washed off by three rinses with PBS; this was followed by a peroxidase reaction using 0.05 M of Tris-HCl (pH 7.6) containing 0.01% H₂O₂ and 0.05% 3,3'-diaminobenzidine (Sigma-Aldrich). The sections were counterstained with hematoxylin and mounted. The slides were examined using the Axiophot photomicroscope (Carl Zeiss, Jena, Germany) and AxioCam MRc5 camera (Carl Zeiss). Each slide was evaluated for frequency of immunopositive cells or nuclei.

The number of BrdU-positive nuclei or cells was counted. Ten randomly selected fields from each group were photographed, and the AxioVision SE64 (Carl Zeiss) program was used for analysis. The BrdU-labeled cells were counted in soft tissues, including fibrous tissue, vascular tissue, and bone marrow of the augmented sinus, and calculated per 1 mm² of soft tissue of the augmented sinus. Statistical analyses were performed with the IBM SPSS Statistics program for Windows (ver. 19.0; IBM, Armonk, NY, USA). In addition to standard descriptive statistical calculations (means and standard deviations), ANOVA was used to determine intra- and intergroup statistical differences. When ANOVA yielded significant results, indicating that the group was significantly different from the others, Tukey’s test was performed. Mean values are accompanied by 95% confidence intervals, and all data are expressed as mean±standard deviation. Statistical significance was noted at P<0.05.

Immunohistochemical studies were performed using mouse anti-PCNA monoclonal antibody (PC10; Santa Cruz Biotechnology, Dallas, TX, USA), mouse anti-collagen I monoclonal antibody (COL-1; Abcam, Cambridge, UK), mouse anti-osteopontin monoclonal antibody (53; Abcam), and mouse anti-osteocalcin monoclonal antibody (OCG3; Abcam). The paraffin-embedded sections were cleared, dehydrated, and washed with PBS. Endogenous peroxidases were inhibited by preincubation in 0.3% H₂O₂ in PBS for 30 minutes, and nonspecific protein binding was blocked in PBS containing 10% normal horse serum (Vector Laboratories) for 30 minutes. The sections were incubated in primary antibodies (1:50-1:200) for two hours at room temperature and washed three times with PBS. The secondary antibody (1:100), biotinylated anti-mouse immunoglobulin G, was placed on the sections for one hour at room temperature and washed off by three rinses with PBS. Then, avidin–biotin–peroxidase complex was placed on the sections for one hour and washed off three times with PBS. A peroxidase reaction using 0.05 M of Tris-HCl (pH 7.6) containing 0.01% H₂O₂ and 0.05% 3,3'-diaminobenzidine was provoked. The sections were counterstained with hematoxylin and mounted; slides were examined using the Axiophot photomicroscope and AxioCam MRc5 camera. All the experiment was performed from July 2016 to April 2017.

No signs of inflammation were observed in the control or experimental group by H&E staining under a light microscope; moreover, the Bio-Oss material and demineralized particulate human tooth bone were lightly stained and well-differentiated from surrounding tissue.(Fig. 2) In the control group, two weeks after surgery, newly formed bone was revealed on the surface of the Bio-Oss material.(Fig. 2. A) Eight weeks after surgery, the newly formed bone appeared thicker and contained many osteocytes. A significant amount of mature lamellar bone was observed inside the newly formed bone, and a portion of bone marrow containing adipose tissue was observed on the floor of the replaceable bone.(Fig. 2. B) In the experimental group, two weeks after surgery, newly formed bone was revealed on the surface of the demineralized particulate human tooth bone.(Fig. 2. A) Six weeks later, the thickness and the density of the new bone were increased. A significant amount of mature lamellar bone was revealed inside the newly formed bone, and a portion of bone marrow containing adipose tissue was observed on the floor of both the replaceable bone and nasal bone. The size and density of the demineralized particulate human tooth bone decreased relative to those at two weeks after surgery.(Fig. 2. B)

In the control group, two weeks after surgery, many BrdU-positive cells were observed on the surface of the Bio-Oss material and in the connective tissue.(Fig. 3. A) Also, some cells strongly positive for BrdU were observed among the osteoblasts on the surface of the newly formed bone and among the osteoprogenitor cells in the surrounding connective tissue. Eight weeks after surgery, a much weaker expression of cells positive for BrdU was observed.(Fig. 3. B) In the experimental group, two weeks after surgery, many BrdU-positive cells were observed on the surface of the demineralized particulate human tooth bone and the newly formed bone as well as in the connective tissue.(Fig. 3. C) Strong expression for BrdU was observed in the osteoblasts on the surface of the demineralized particulate human tooth bone. Some cells strongly positive for BrdU were observed among the osteoprogenitor cells in the surrounding connective tissue. Eight weeks after surgery, cells positive for BrdU showed moderate expression in osteoblasts on the surface of the demineralized particulate human tooth bone and the newly formed bone.(Fig. 3. D)

In the control group, numbers of BrdU-labeled cells per 1 mm² of soft tissue of the augmented sinus at two and eight weeks totaled 178.26±27.96 and 49.52±7.03, respectively. One-way ANOVA and post-hoc comparisons showed that the number of BrdU-labeled cells at eight weeks was significantly less than that at two weeks, with P<0.05.(Fig. 4) In the experimental group, the numbers of BrdU-labeled cells per 1 mm² of soft tissue of the augmented sinus at two and eight weeks totaled 183.07±48.18 and 77.24±11.67, respectively, with the number of BrdU-labeled cells at eight weeks being significantly less than that at two weeks, with P<0.05.(Fig. 4) Notably, the number of BrdU-labeled cells at two weeks in the control group was not significantly different from that at two weeks in the experimental group, but the number of BrdU-labeled cells at eight weeks in the control group was significantly less than that at eight weeks in the experimental group, with P<0.05.(Fig. 4)

In the control group, two weeks after surgery, many PCNA-positive cells were observed on the surface of the Bio-Oss material and in the connective tissue.(Fig. 5. A) Eight weeks after surgery, the expression of PCNA was reduced and some PCNA-positive cells were observed on the surface of the newly formed bone.(Fig. 5. B) In the experimental group, two weeks after surgery, PCNA-positive cells were observed in high numbers on the surface of the demineralized particulate human tooth bone and newly formed bone (Fig. 5. C), and some PCNA-positive cells were observed in the soft tissue. Strong expression of PCNA was recorded among osteoblasts on the surface of the demineralized particulate human tooth bone. Eight weeks after surgery, the expression of PCNA was slightly decreased and PCNA-positive cells were observed on the surface of the demineralized particulate human tooth bone.(Fig. 5. D) Overall, the immunoreactivity of PCNA at eight weeks was slightly higher in the experimental group than in the control group.

In the control group, two weeks after surgery, weak expression of type I collagen was observed on the surface of the Bio-Oss material and in surrounding connective tissue.(Fig. 6. A) Eight weeks after surgery, strong expression of type I collagen was confirmed on the surface of the newly formed bone and Bio-Oss material and among osteoblasts.(Fig. 6. B) In the experimental group, two weeks after surgery, weak expression of type I collagen was observed on the surface of the demineralized particulate human tooth bone and in surrounding connective tissue.(Fig. 6. C) Finally, eight weeks after surgery, strong expression of type I collagen was observed on the surface of the newly formed bone and demineralized particulate human tooth bone and in fibroblast-rich dense connective tissue.(Fig. 6. D)

In the control group, two weeks after surgery, slight positive expression of osteopontin was observed on the surface of the Bio-Oss material and in surrounding connective tissue.(Fig. 7. A) Eight weeks after surgery, strong expression of osteopontin was present in surrounding connective tissue and on the surface of the newly formed bone and the Bio-Oss material.(Fig. 7. B) Meanwhile, in the experimental group, two weeks after surgery, weak expression of osteopontin was observed on the surface of the demineralized particulate human tooth bone and in surrounding connective tissue (Fig. 7. C), while, at eight weeks, strong expression of type I collagen was observed on the surface of the newly formed bone and demineralized particulate human tooth bone and in fibroblast-rich dense connective tissue.

In the control group, two weeks after surgery, weak expression of osteocalcin was observed on the surface of the Bio-Oss material and in surrounding connective tissue (Fig. 8. A), while, eight weeks after surgery, a modulating expression level of osteocalcin was present in surrounding connective tissue. Some osteocalcin-positive osteocytes were observed in the newly formed bone.(Fig. 8. B) In the experimental group, two weeks after surgery, slight positive expression of osteocalcin was observed on the surface of the demineralized particulate human tooth bone and in surrounding connective tissue (Fig. 8. C), and some degree of positive expression of osteocalcin was found in the matrix of the demineralized particulate human tooth bone. Eight weeks after surgery, a modulating expression of osteocalcin was observed on the surface of the newly formed bone and demineralized particulate human tooth bone and in fibroblast-rich dense connective tissue. Also, a strong expression level of osteocalcin was observed on the surface of newly formed bone and demineralized particulate human tooth bone and in fibroblast-rich dense connective tissue.(Fig. 8. D)