Phenotypic characterization

BMCs were submitted to osteogenic conditions to determine their potential to promote mineral nodules formation in vitro and to express osteoblastic cell markers . In vitro mineral nodule formation was assessed using the von Kossa assay. Gene expression analyses evaluated the genes: alkaline phosphatase (ALP), bone sialoprotein (BSP) and type I collagen (COL I) ( Table 1 – Appendix A ). Gyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a reference gene. Amplified samples were visualized on 2.0% agarose gels stained with ethidium bromide and photographed under UV light.

To examine cell morphology, spreading and adhesion to the scaffolds (BD 3D Scaffold Composite, BD Biosciences, San Jose, CA, USA), scanning electron microscopy (SEM) analysis was performed. Cells were seeded into the scaffold in triplicate, incubated for 3 days in standard media and examined under SEM for descriptive analysis .

To evaluate the cells’ capacity to maintain the osteogenic phenotype after seeding in the scaffolds, a mineral nodule formation assay was performed on cell seeded scaffolds, which were kept under osteogenic conditions for 14 days. Cell seeded scaffolds were incubated with standard media, during the same period, as the control. The scaffolds were embedded in paraffin, sections were stained with alizarin red solution, and histological descriptive analysis was performed.

Surgical procedure

After surgical anaesthesia, the third and fourth mandibular premolars (P3, P4) were bilaterally extracted. Three dehiscence type defects were created in each mandible , after 3 months. Each mandible received three machined surface screw-shaped commercial pure titanium implants of 4 × 8.5 mm (Biomet – 3i™ do Brasil LTDA, São Paulo, SP, Brazil) ( Fig. 1 A). One implant was placed at one side and two were placed at the other side; the side was assigned randomly.

(A) Non-treated defect (control group). (B) Cell-treated defect (BMC group). (C) Defect treated with BMCs associated with barrier (BMCs + GBR group).

All animals received the three treatments proposed; one treatment in each defect. Randomization was performed according to a computer-generated code. Each defect was allocated to one of the following groups: BMCs, bone marrow cells loaded onto scaffold (2 × 10 ⁷ cells/scaffold); BMCs + GBR, bone marrow cells loaded onto scaffold (2 × 10 ⁷ cells/scaffold) associated with application of a titanium reinforced expanded polytetrafluoroethylene membrane (Gore-Tex, TR4Y, Flagstaff, AZ, USA); and control, no treatment ( Fig. 1 B–D). Seeded cells were cultured in standard medium for 24 h. The medium was then changed to the osteogenic inducing type and the cells were incubated for 3 days. After the treatments, the flaps were repositioned and sutured. Following surgery, flunixin meglumine (Banamines, Schering-Plough Veterinary, Rio de Janeiro, RJ, Brazil) was administered for 3 consecutive days. Penicillin/erythromycin (Pentabiotic, Wyeth-Whitehall, São Paulo, SP, Brazil) was administrated on the first and fourth days. Postoperative plaque control was performed during the whole experimental phase.

Histomorphometric analysis

After 3 months, the animals were killed and the oral tissues were fixed by perfusion with 10% buffered formalin. One undecalcified buccolingual section was prepared for each implant . The sections were stained with 1% toluidine blue. Histometric analysis was performed using light microscopy and a PC-based image analysis system to record the following parameters : bone-to-implant contact (BIC), percentage of BIC along the threads of the implant surface within the defect area; bone fill (BF), percentage of mineralized bone formed within the threads of the implant located into the defect region; new bone area (BA), total area (μm ² ) of new bone formation out of the threads of the implant into the defect region; new bone height (BH), linear vertical distance (mm) from the bottom of the defect to the most coronal level of the newly formed bone; new bone weight (BW), linear horizontal distance (mm) from the thread of the implant surface to the most external level of the newly formed bone at bottom of the bone defect.

All measurements were carried out by the same calibrated masked examiner, after intra-examiner calibration by evaluating 7 non-study photomicrographs presenting peri-implant dehiscence defects. The examiner measured each parameter of all photomicrographs twice within 24 h. The intra class correlation showed 97% reproducibility for BF, 92% for BA, 95% for BIC and BW, and 98% for BH.

Statistical analysis

A randomized block design was used in the study, as each dog received all three treatments. The data were statistically analysed using the two-way ANOVA ( α = 5%) to test the hypothesis that there were no intergroup differences for the assessed parameters. When statistical difference was found, analysis of the difference was determined using Tukey’s method. Values of P < 0.05 were considered statistically significant.

Phenotypic characterization

BMCs were submitted to osteogenic conditions to determine their potential to promote mineral nodules formation in vitro and to express osteoblastic cell markers . In vitro mineral nodule formation was assessed using the von Kossa assay. Gene expression analyses evaluated the genes: alkaline phosphatase (ALP), bone sialoprotein (BSP) and type I collagen (COL I) ( Table 1 – Appendix A ). Gyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a reference gene. Amplified samples were visualized on 2.0% agarose gels stained with ethidium bromide and photographed under UV light.

To examine cell morphology, spreading and adhesion to the scaffolds (BD 3D Scaffold Composite, BD Biosciences, San Jose, CA, USA), scanning electron microscopy (SEM) analysis was performed. Cells were seeded into the scaffold in triplicate, incubated for 3 days in standard media and examined under SEM for descriptive analysis .

To evaluate the cells’ capacity to maintain the osteogenic phenotype after seeding in the scaffolds, a mineral nodule formation assay was performed on cell seeded scaffolds, which were kept under osteogenic conditions for 14 days. Cell seeded scaffolds were incubated with standard media, during the same period, as the control. The scaffolds were embedded in paraffin, sections were stained with alizarin red solution, and histological descriptive analysis was performed.

Surgical procedure

After surgical anaesthesia, the third and fourth mandibular premolars (P3, P4) were bilaterally extracted. Three dehiscence type defects were created in each mandible , after 3 months. Each mandible received three machined surface screw-shaped commercial pure titanium implants of 4 × 8.5 mm (Biomet – 3i™ do Brasil LTDA, São Paulo, SP, Brazil) ( Fig. 1 A). One implant was placed at one side and two were placed at the other side; the side was assigned randomly.

(A) Non-treated defect (control group). (B) Cell-treated defect (BMC group). (C) Defect treated with BMCs associated with barrier (BMCs + GBR group).

All animals received the three treatments proposed; one treatment in each defect. Randomization was performed according to a computer-generated code. Each defect was allocated to one of the following groups: BMCs, bone marrow cells loaded onto scaffold (2 × 10 ⁷ cells/scaffold); BMCs + GBR, bone marrow cells loaded onto scaffold (2 × 10 ⁷ cells/scaffold) associated with application of a titanium reinforced expanded polytetrafluoroethylene membrane (Gore-Tex, TR4Y, Flagstaff, AZ, USA); and control, no treatment ( Fig. 1 B–D). Seeded cells were cultured in standard medium for 24 h. The medium was then changed to the osteogenic inducing type and the cells were incubated for 3 days. After the treatments, the flaps were repositioned and sutured. Following surgery, flunixin meglumine (Banamines, Schering-Plough Veterinary, Rio de Janeiro, RJ, Brazil) was administered for 3 consecutive days. Penicillin/erythromycin (Pentabiotic, Wyeth-Whitehall, São Paulo, SP, Brazil) was administrated on the first and fourth days. Postoperative plaque control was performed during the whole experimental phase.

Histomorphometric analysis

After 3 months, the animals were killed and the oral tissues were fixed by perfusion with 10% buffered formalin. One undecalcified buccolingual section was prepared for each implant . The sections were stained with 1% toluidine blue. Histometric analysis was performed using light microscopy and a PC-based image analysis system to record the following parameters : bone-to-implant contact (BIC), percentage of BIC along the threads of the implant surface within the defect area; bone fill (BF), percentage of mineralized bone formed within the threads of the implant located into the defect region; new bone area (BA), total area (μm ² ) of new bone formation out of the threads of the implant into the defect region; new bone height (BH), linear vertical distance (mm) from the bottom of the defect to the most coronal level of the newly formed bone; new bone weight (BW), linear horizontal distance (mm) from the thread of the implant surface to the most external level of the newly formed bone at bottom of the bone defect.

All measurements were carried out by the same calibrated masked examiner, after intra-examiner calibration by evaluating 7 non-study photomicrographs presenting peri-implant dehiscence defects. The examiner measured each parameter of all photomicrographs twice within 24 h. The intra class correlation showed 97% reproducibility for BF, 92% for BA, 95% for BIC and BW, and 98% for BH.

Statistical analysis

A randomized block design was used in the study, as each dog received all three treatments. The data were statistically analysed using the two-way ANOVA ( α = 5%) to test the hypothesis that there were no intergroup differences for the assessed parameters. When statistical difference was found, analysis of the difference was determined using Tukey’s method. Values of P < 0.05 were considered statistically significant.