This study was designed to compare bone regeneration of tissue-engineered bone from adipose-derived stem cell and autogenous bone graft in a canine maxillary alveolar cleft model. In this prospective clinical trial, mesenchymal stem cells (MSCs) were isolated from subcutaneous canine adipose tissue. Undifferentiated cells were incubated with a 3 mm × 3 mm × 3 mm hydroxyapatite/beta-tricalcium phosphate scaffold, in specific osteogenic medium for 21 days. Four mongrel dogs were prepared by removal of two of the three incisors bilaterally and a 15 mm defect in bone was created from crest to nasal floor. After healing, repair was followed by a tissue engineered bone graft from adipose-derived stem cells on one side and corticocancellous tibial auto graft on the other side. Bone regeneration was evaluated by histomorphometry on days 15 and 60 after implantation. The data were analysed with descriptive and t test methods ( α = 0.05). Bone formation on the autograft sides was higher than on the stem cell sides at 15 and 60 days, 45% and 96% versus 5% and 70%, respectively. Differences between the two groups at 15 and 60 days were significant ( p = 0.004 and 0.001, respectively). Although autograft is still the gold standard for bone regeneration, tissue engineered bone may provide an acceptable alternative.
Repair of bony defects remains a challenging part of many reconstructive procedures. The use of autogenous bone is the gold standard for grafting bone defects. The reconstruction of alveolar cleft defects is well established, with the most widely accepted approach being secondary alveolar cleft osteoplasty in the mixed dentition phase. In conventional methods, autogenous bone grafting has become an essential step in treating patients with alveolar cleft, but harvesting autogenous bone graft for an alveolar cleft defect may cause short- and long-term complications in donor sites.
To reduce these complications, substitute biomaterials, such as hydroxyapatites, α- and β-tricalciumphosphates, and demineralized bone matrices, are in clinical use. Combining them with osteoblasts and growth factors (i.e. tissue engineering) provides a new alternative in which bone cells are seeded on 3-dimensional bone-like scaffolds of natural and artificial origin. Both synthetic and allograft materials allow adhesion and growth of osteoblastic cells, or osteogenic differentiation of precursor cells in vitro .
Tissue engineering of bone is a rapidly growing field and is a promising approach. In tissue engineering, cells, the extra cellular matrix, and growth factors are combined to design novel graft materials which can induce tissue regeneration and repair based on natural healing potential. There is much debate about the ideal source of osteoprogenitor cells for use in skeletal tissue engineering.
Embryonic stem cells, the gold standard of multipotency, are derived from the inner cell mass of the preimplantation blastocyte. Embryonic stem cells are known to have the ability to differentiate into multiple tissues type from all three embryonic germ layers. Concerns about the possibility of infection, immunogenicity, and tumourgenicity have limited the application of embryonic stem cells. Adult stem cells, derived from different tissues, have the unique ability to self-renew and differentiate into various phenotypes. These cells have the potential for cell therapy and tissue engineering. Adipose tissue is an appropriate source of mesenchymal stem cells (MSCs) with wide differentiation potential.
Owing to the abundance of stem cells, the ease with which they can be procured, and their rapid expansion in vitro , adipose derived stem cells (ADSCs) are particularly desirable candidates for skeletal tissue engineering applications. Zuk originally characterized this population of cells isolated from adipose tissue and found that they were able to differentiate towards osteogenic, adipogenic, myogenic, and chondrogenic lineages in vitro when treated with the appropriate inducing factors. Cowan et al. demonstrated the ability of ADSC seeded onto apatite-coated scaffolds to heal critical-sized (4 mm) calvarial defects. This was the first published report of the ability of ADSC to heal critical-sized bony defects. Shi et al. compared the biological differences and osteogenic ability between juvenile and adult mice and although differences were demonstrated, namely greater attachment and proliferation in juvenile mice, adult ADSCs also exhibited robust terminal osteogenic differentiation. The dog offers a valuable experimental model and current studies have mainly focused on the osteogenic potential of canine MSCs in vivo and in vitro .
MSCs when combined with porous, biphasic calcium phosphate ceramics, namely hydroxyapatite/β-tricalcium phosphate (HA–TCP) ceramics with the composition 60% HA/40%TCP (in weight%), have been shown to induce bone formation in large, long bone defects. Yoshikawa et al. reported that HA loaded with MSCs has osteogenic potential comparable with autogenous particulate cancellous bone.
This study was designed to compare ADSCs based alveolar cleft regeneration with traditional autogenous bone grafting in a through-and-through canine alveolar cleft model, histologically. ADSCs were loaded on HA/TCP the bone regeneration of which has been compared histologically with traditional autogenous bone grafts.
Materials and methods
Isolation and cultivation of ADSCs
This study was performed in accordance with the regulation and approval of the Institutional Animal Care and Use Committee of the Isfahan University of Medical Sciences and conformed to its standard of animal care. Under general anaesthesia, 20 mg scapular subcutaneous adipose tissue was isolated from four mongrel dogs that had undergone maxillary alveolar cleft creation surgery. The isolated adipose tissues were cut into small pieces and washed with phosphate buffered saline (PBS; Gibco, UK). To process them, 1.5 mg of collagenase type I (Sigma, USA) per gram of fat tissue was added and incubated with continuous shaking for 1 h at 37 °C. To separate stromal cells from floating adipocytes multiple centrifugation and washing steps were applied before removing red blood cells by application of lysis buffer. The separated stromal cells were counted using a haemocytometer and were plated in tissue culture flasks (3000 cells/cm 2 ) containing Dulbecco’s modified Eagle’s medium (DMEM; Gibco, UK) supplemented with 10% foetal bovine serum (FBS; Dainippon Pharmaceutical, Osaka, Japan), 1% penicillin–streptomycin (Gibco-BRL, Life Technologies) and incubated at 37 °C with 5% carbon dioxide. After 24 h the non-adherent cells were discarded and the medium changed. Culture media were replaced every 2–3 days and trypsinization and replating was carried out when the cultures reached about 80% confluency.
There is no definitive marker to identify MSCs, so the gold standard procedure to prove their stem cell identity is: their adherence to cell culture plates after isolation; their expression of specific markers; and their differentiation potential to osteoblast, chondrocyte and in vitro . In this study, the MSC character was proven by flow cytometrical analysis and by the ability of the cell to differentiate into various lineages. In one tube 1 × 10 5 cells were stained simultaneously with phycoerythrin (PE) conjugated monoclonal antibody to CD 44 (ab58754; ABCAM Antibodies, Cambridge Science Park, UK) and fluorecin isothiocyanate conjugated (FITC) monoclonal antibody to CD 90 (ab22541 ; ABCAM Antibodies, Cambridge Science Park, UK). After incubation at room temperature for 15 min the specimen was analysed by FACS caliber 488 (Becton Dickenson, CA, USA). The FACS analysis showed a distinct population of CD 44 and CD 90 positive cells. The result of FACS analysis with the result of the differentiation assay proves that actual MSCs were transplanted ( Fig. 1 A and B ).
In vitro osteogenic differentiation
For in vitro osteogenic differentiation, confluent passage 3 culture was used. 5 × 10 6 cells were incubated with 3 mm × 3 mm × 3 mm HA/TCP (Ceraform, Teknimed, France) (60% HA and 40% β-TCP with a mean pore size of 200–800 μm) in specific osteogenic medium at 37 °C and 5% carbon dioxide for 21 days. Osteogenic medium consisting of 50 μml ascorbic acid 2-phosphate (Sigma, USA), 10 μml β-glycerophosphate (Sigma, USA) and 100 nmg dexamethasone (Sigma, USA).
Cell differentiation was evaluated by reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of osteogenic gen expression. Osteocalcin and collagen I were largely produced after 21 days in an osteoinductive medium. The selected housekeeping gene was GAPDH ( Fig. 2 A and B ).
Scanning electron microscopy
The morphology of the HA/TCP scaffolds, with and without cells, was observed by scanning electron microscopy (SEM, Vega Tescan, Philadelphia, PA, USA) at an accelerating voltage of 20 kV. Before the observation, samples of cell-scaffold constructs were fixed with 2% paraformaldehyde/2.5% glutaraldehyde in 0.1 M Na cacodylate buffer (PH 7.4), then dehydrated in graded alcohols, and examined with SEM ( Fig. 3 ).
Surgical procedure for alveolar cleft creation
Four adult mongrel dogs (mean age 22 month) weighing 20–30 kg were used in this study. The animals were kept for 2 week to become acclimatized to the housing and diet. Throughout the experiments they were monitored for general appearance and weight. They were starved 24 h before each surgery and for 24 h after surgery. During this period they were given serum therapy with lactated Ringer’s solution. After surgery, 1 mg ceftriaxon was administered once a day. The dogs were given a soft diet up to 3 weeks after surgery. Under general anaesthesia with ketamine (20 mg/kg) and rampone (2 mg/kg), the animals were prepared by removal of two of the three incisors bilaterally and a 15 mm wide defect was created from the alveolar crest to the nasal floor using a dental handpiece. Nasal bone and membrane were removed to create complete alveolar cleft penetration to the nasal cavity. The nasal mucosa was sutured to the oral mucosa (Ethicon, Norderstedt, Germany) and a stent was placed (endotracheal tube no. 7, as the stent was placed through the cleft on one side, exiting through the opposite cleft side, and fixed by wires to the maxillary canine teeth, which had been notched on their distal aspect) ( Fig. 4 ). Two months were allowed for healing and in this way bilateral clefts were prepared with functional teeth on each side that were expected not to heal spontaneously with new bone. To approximate human maxillary alveolar cleft more closely, the experimentally created clefts had to fulfil the following five criteria ( Fig. 5 ): bilateral maxillary alveolar cleft had to exist in each research animal; each cleft had to have 15 mm bony width; a demonstrable oronasal communication had to be present; each cleft had to be lined by healthy epithelialized mucosa; and there had to be functional teeth on each side of every cleft.
Implantation of scaffold/MSCs constructs and autograft
After removal of the stent, 2 weeks were allowed for local inflammation to subside. Following crestal incision at the level of the gingival sulcus, the scar tissues were dissected to reach the bony surface of the cleft walls. The tissue was elevated in the subperiostal plane. The flaps of the nasal floor and the oral mucosa formed the ceiling and the floor of the cleft, respectively. For repair, one side was grafted with tissue engineered bone from ADSCs (the scaffolds with cells were transfer to the defect by microforceps) and the other side was repaired with corticocancellous tibial autograft harvested at the same session, as the conventional method ( Fig. 6 A and B ). The wound was closed in a watertight manner.