The aim of this study was to observe the effect of hypoxia-inducible factor (HIF)-1α on bone regeneration during distraction osteogenesis (DO). Fifty-one New Zealand white rabbits underwent mandibular lengthening with a distraction rate of 2 mm/day, and were divided randomly into three groups (17 in each). Group C rabbits received 20 μg rHIF-1α, group B received 10 μg rHIF-1α, and group A received 100 μl saline injection in the distraction gap every day for 10 days. Radionuclide bone imaging (RBI), computed tomography, dual energy X-ray absorptiometry, radiography, histology, and three-point bend testing were performed. RBI showed that the uptake ratio in group B (1.41 ± 0.25, P = 0.013) and group C (1.64 ± 0.37, P < 0.001) was higher than that in group A (1.01 ± 0.26). The bone mineralization density and bone mineralization content in group C were highest among the three groups. Radiology and histology findings indicated more callus regeneration in groups C and B. Mechanical testing demonstrated that the ultimate force in group C (289.71 ± 43.31 N, n = 6) was 1.49-fold ( P < 0.001) that of group A and 1.20-fold ( P = 0.012) that of group B. HIF-1α may represent a new agent to promote DO by accelerating osteogenesis and mineralization.
Distraction osteogenesis (DO) is a surgical tool that is used worldwide for limb lengthening, the correction of deformity, and the reconstruction of large bone defects. A unique advantage of this technique over other bone augmentation techniques is that bone lengthening is accompanied by simultaneous expansion of the functional soft tissues, including periosteum, blood vessels, nerves, muscles, skin, etc. However, the long treatment period and the potential for pin-track infection, stiffness of the joint, and fibrous union or non-union under certain circumstances, remain major limitations, especially for patients in a compromising osteogenic condition. Many growth factors and hormone proteins have been used in an attempt to enhance the regenerated bone quality and shorten the bone consolidation period, e.g. bone morphogenetic proteins (BMPs), insulin growth factor (IGF), basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF), nerve growth factor, adiponectin, platelet-rich plasma, etc. Most of these factors have given some positive results, but they are far from being ready for extensive clinical use.
Close temporal and spatial relationships between bone formation and angiogenesis have been observed during the process of DO. Previous studies have demonstrated that hypoxia-inducible factor (HIF) influences mesenchymal cells to differentiate along the osteoblast pathway, in part through activating osterix gene expression. HIF is also an active nucleoprotein that enhances the production of corresponding proteins, thereby launching the process of angiogenesis. It is therefore proposed that this agent could promote the DO process by simultaneously enhancing osteogenesis and angiogenesis.
This study was designed to observe the influence of recombinant human HIF (rHIF)-1α protein on bone regeneration during the process of DO in the rabbit mandible model. To simulate a poor osteogenic condition, short delay periods and a rapid distraction rate were adopted in the rabbit mandibular DO model.
Materials and methods
Fifty-one adult male New Zealand white rabbits weighing 3.5–4.0 kg were used in this study. Animal care and use was in accordance with the guidelines established by the animal ethics committee of the study university. Anaesthesia was administered with a mixture of ketamine (20 mg/kg; Sigma, Saint Louis, MO, USA) and xylazine (5 mg/kg; Sigma) by intravenous injection.
Osteodistraction and treatment protocol ( Fig. 1 )
A 3-cm skin incision was made longitudinally over the right inferior border of the mandible. After carefully separating the muscles and the periosteum, the right mandible was exposed. A custom-made external fixator was applied. Next, a vertical osteotomy was performed just between the mental foramen and the first molar tooth with an ultrafine drill-bit under continuous saline irrigation. The wound was then closed in layers with a suture (Mersilk, Ethicon, USA). At this point, the rabbits were divided randomly into three groups: group A ( n = 17), group B ( n = 17), and group C ( n = 17). After a 3-day delay period, the right mandible was distracted at a rate of 2.0 mm/day (1 mm every 12 h) for 5 days. From the third day of the distraction period, the animals in group C received 20 μg rHIF-1α (dissolved in 100 μl saline; ProSpec, Israel), those in the group B received 10 μg rHIF-1α (dissolved in 100 μl saline), and those in the group A received 100 μl saline injection in the distraction gap every day for 10 days .
Detecting protein factors in the distracted region tissue by Western blotting
Four weeks after the end of distraction, five rabbits selected randomly from each group were euthanized with an overdose of anaesthetic. The distracted callus was harvested immediately to detect the expression of growth factors by Western blotting. The total protein from the distracted tissue was collected using a commercially available kit (Micro; Sigma, Saint Louis, MO, USA). Fifty micrograms of total protein was separated on a sodium dodecyl sulphate (SDS)–polyacrylamide gel and transferred to polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA) by electrophoresis. After blocking the non-specific binding, the membranes were incubated overnight at 4 °C with primary antibodies: BMP-2 (1:750; Abcam, Hong Kong), bFGF (1:500; Boster, China), IGF (1:500; Boster, China), VEGF (1:1000; Abcam, Hong Kong), and runt-related transcription factor 2 (RUNX2) (1:500, Boster, China). After incubation of the membranes with horseradish peroxidase (HRP)-conjugated secondary antibody, the signals were detected by chemiluminescence detection (ECL; Bio-Rad, Hercules, CA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 1:500; Abcam, Hong Kong) served as the internal control.
Radionuclide bone imaging
Four weeks after the completion of distraction, eight animals selected randomly from each group received an injection of 2 mCi/kg 99m Tc-methylenediphosphonate ( 99m Tc-MDP; Syncor, Shanghai, China) in the right marginal ear vein, flushing with 1 ml physiological saline. Three hours later, the accumulation of radionuclide in the DO region was assessed under anaesthesia. Images were obtained with 256 × 256 matrices and the data were obtained using a large field-of-view gamma camera equipped with a low-energy, high-resolution, parallel-hole collimator (SKYlight SPECT; Philips, Netherlands). Energy discrimination was centred on 140 keV with a 20% window. The distracted segment was set as the region of interest (ROI) and the symmetrical region on the contralateral side served as the control. The uptake ratio was calculated as the uptake in the ROI of the distraction segment divided by the uptake in the contralateral normal area.
Computed tomography and DXA examination
After radionuclide bone scanning, the eight rabbits selected randomly from each group underwent computed tomography scanning (SOMATOM Sensation, 120 kV, 100 mA; Siemens, Germany) in the supine position under anaesthesia. Bone mineral density (BMD) and bone mineral content (BMC) were also assessed at 4 and 8 weeks after distraction by dual-energy X-ray absorptiometry (DXA) analysis (Lunar iDXA; GE Healthcare, USA). The distraction regions were selected as the ROIs. After 8 weeks, all of the animals were euthanized with an overdose of anaesthetic for subsequent investigations.
Eight right mandibles selected randomly from each group were positioned 30 cm from the X-ray tube. The X-ray unit was set at 70 kV and 7 mA with a 0.06-ms exposure time (DEN S-O-MAT; Gendex, Milano, Italy).
Six samples selected randomly from each group underwent mechanical testing. The three-point bending strength was measured with a support span of 5 mm using a commercial materials testing system (Instron, Norwood, MA, USA) with a 1-kN load cell under displacement control (1 mm/min). The ultimate load at failure (N) was recorded.
The remaining samples (five in each group) were dissected 5 mm outside the distracted region for histological examination. After fixing in 4% paraformaldehyde solution for 24 h, the samples were decalcified in 14.5% ethylenediaminetetraacetic acid buffer (pH 7.2) at room temperature for 20–30 days. The specimens were sectioned longitudinally along the axial plane and embedded in paraffin, then cut with a microtome (Leica, Germany) into 5-μm sections for haematoxylin and eosin staining.
All data are presented as the mean ± standard deviation (SD) and were analyzed statistically using SPSS v. 13.0 software (SPSS Inc., Chicago, IL, USA). One-way analysis of variance (ANOVA) was conducted to assess the differences between groups. P -values of less than 0.05 were considered to be statistically significant.
Only one rabbit died from postoperative infection (in group A); the rest of the animals tolerated the experimental procedures well. After the distraction period, an apparent cross-bite was present in each animal.
Western blotting results
BMP-2, bFGF, VEGF, RUNX2, and IGF protein levels of the distracted tissue were detected in all groups. With the increasing dosage of rHIF-1α, levels of BMP-2, bFGF, and VEGF were noticeably elevated; levels of RUNX2 were slightly up-regulated. However, there was no obvious change in IGF expression ( Fig. 2 ).
Radionuclide bone imaging results
99m Tc-MDP was deposited mostly in the distracted gaps in all groups. The uptake ratios for group B (1.41 ± 0.25, P = 0.013) and group C (1.64 ± 0.37, P < 0.001) rabbits were significantly higher than that of group A animals (1.01 ± 0.26), while there was no significant difference between group B and group C ( P = 0.129) ( Fig. 3 ).
BMD and BMC evaluation
DXA measurements of the regenerate regions revealed that the BMD at week 4 of the consolidation period was 0.138 ± 0.021 g/cm 2 in group B, which was 1.32-fold that in group A (0.105 ± 0.015 g/cm 2 , P = 0.005). The BMD in group C (0.152 ± 0.026 g/cm 2 ) was 1.44-fold that in group A ( P < 0.001). However, there was no significant difference between group B and group C ( P = 0.208). The BMC in group B at week 4 of the consolidation period was 0.102 ± 0.009 g, which was 1.16-fold that in group A (0.088 ± 0.010 g, P = 0.004). The BMC in group C (0.113 ± 0.007 g) was 1.28-fold that in group A ( P < 0.001) and 1.11-fold that in group B ( P = 0.019).
BMD at week 8 of the consolidation period was 0.173 ± 0.018 g/cm 2 in group B, which was 1.22-fold that in group A (0.142 ± 0.023 g/cm 2 , P = 0.027). The BMD in group C (0.216 ± 0.032 g/cm 2 ) was 1.52-fold that in group A ( P < 0.001) and 1.25-fold that in group B ( P = 0.002). BMC in group B at 8 weeks of consolidation was 0.149 ± 0.036 g, which was 1.43-fold that in group A (0.104 ± 0.027 g, P = 0.004). The BMC in group C (0.192 ± 0.019 g) was 1.85-fold that in group A ( P < 0.001) and 1.29-fold that in group B ( P = 0.006).
Radiological examination showed callus regeneration in the distracted gap of all animals at 8 weeks after the end of distraction ( Fig. 4 ). However, the intensity of regeneration differed among the three groups. In group A, the lengthened segment showed a characteristic zone structure consisting of a central radiolucent zone and two sclerotic zones. The osteotomy site showed bony continuity, but the distracted area was less radiodense than the host bone and the distracted region was easy to discern. In group B, greater callus regeneration was observed and the distracted area was difficult to discern. Radiodensity was highest in the distraction gap in group C, suggesting greater mineralization. The inferior and superior borders of the distracted area in group C showed obvious cortical bone. Bone regeneration within the distracted callus appeared to increase much earlier in this group, and the newly forming callus was the most abundant and homogeneous in its structure among the three groups.