- Email: [email protected]
Marrow-derived mesenchymal stem cells-directed bone regeneration in the dog mandible: a comparison between biphasic calcium phosphate and natural bone mineral
Marrow-derived mesenchymal stem cells-directed bone regeneration in the dog mandible: a comparison between biphasic calcium phosphate and natural bone mineral Mohammad Jafarian, DMD, MS,a Mohamadreza Baghaban Eslaminejad, PhD,b Arash Khojasteh, DMD,c Fatemeh Mashhadi Abbas, DDS, MS,d Mohammad Mehdi Dehghan, DVM, PhD,e Rahele Hassanizadeh, DDS,f and Bahar Houshmand, DDS,f Tehran, Iran SHAHID BEHESHTI UNIVERSITY OF MEDICAL SCIENCES, ROYAN INSTITUTE, AND UNIVERSITY OF TEHRAN
Objective. This study was designed to compare mesenchymal stem cell (MSC)-based alveolar bone regeneration in biphasic bone substitutes and natural bone mineral in a canine full-thickness alveolar defect model. Materials and methods. MSCs were isolated from bone marrow aspirates and culture expanded through 3 successive subcultures. The bone differentiation potential of third passage cells was evaluated and confirmed in vitro before cells were used in the transplantation experiment. Undifferentiated cells were then incubated with 3 ⫻ 3 ⫻ 3 mm3 hydroxyapatite/-tricalcium phosphate (HA/TCP) matrices (Kasios, Lanauguet, France) and 1- to 2-mm Bio-Oss spongiosa (Geistlich Biomaterials, Osteohealth, Switzerland), which is a natural bovine bone mineral (NBM). Kasios/ cell, Kasios alone, Bio-Oss/cell, and Bio-Oss alone were implanted in masseter muscle and 4 cylindrical (10-mm diameter) through-and-through bilateral mandibular body defects in 4 mongrel dogs. Histomorphometric analysis was performed 6 weeks after insertion of the scaffold loaded with MSCs. Results. H&E staining of the decalcified scaffold and scanning electron microscopy demonstrated large MSC coverage of the HA/TCP and Bio-Oss. Cell-loaded Kasios matrices showed the greatest amount of the bone regeneration among the groups in both the muscle (29.11%) and the bone specimens (65.78%). Cell-free biphasic scaffold revealed 44.9% bone fill in bone defects and 23.55% in muscle specimen, and Bio-Oss alone matrices had the least amount of new bone formation: 36.84% and 24.16% in bone and muscle specimens respectively. Kasios loaded with MSCs demonstrated more bone regeneration than Bio-Oss/cell but there was no significant statistical difference (P ⬎ .05). Conclusions. New biphasic synthetic bone substitutes may offer better conditions for bone regeneration than traditional bone substitute in combination with MSCs. They remained in the defect and contributed to bone regeneration. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:e14-e24)
a
Assistant Professor, Oral and Maxillofacial Surgery, Department of Oral and Maxillofacial Surgery, Taleghani University Hospital, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. b Assistant Professor of Anatomy and Embryology, Stem Cell Department, Royan Institute. c Chief Resident, Oral and Maxillofacial Surgery, Taleghani University Hospital, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. d Assistant Professor, Oral and Maxillofacial Pathology, Department of Oral and Maxillofacial Pathology, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. e Assistant Professor of Veterinary Surgery, Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Tehran. f Dentist, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. Received for publication Sep 25, 2007; returned for revision Dec 15, 2007; accepted for publication Jan 7, 2008. 1079-2104/$ - see front matter © 2008 Mosby, Inc. All rights reserved. doi:10.1016/j.tripleo.2008.01.010
e14
For the repair of bone defects, a tissue-engineering approach would be to combine cells capable of osteogenic activity with an appropriate scaffolding material to stimulate bone regeneration and repair. Dental and craniofacial surgeons must often restore aesthetics and function to their patients, despite an insufficient amount of bone. For this reason, bone regeneration procedures are an integral part of dental and craniofacial therapy.1 Although alloplastic materials including synthetic materials and natural bone– derived materials are used, autogenous bone remains the standard of care for regeneration of significant volumetric defects of the maxilla or mandible. Because of the limited amount of autogenous bone available for reconstructive use, other therapies are being explored.1,2 Bone marrow– derived mesenchymal stem cells (MSCs) are multipotential cells that are capable of differentiating into, at a minimum, osteoblasts, chondrocytes, adipocytes, tenocytes, and myoblasts. From a small volume of bone marrow, MSCs can be isolated and cultures expanded into large
OOOOE Volume 105, Number 5
Jafarian et al. e15
of Dulbecco modified Eagle medium (DMEM; GibcoBRL, Life Technologies, Grand Island, NY) supplemented with 15% fetal bovine serum (FBS; Dainippon Pharmaceutical, Osaka, Japan) and 100 g/mL penicillin-streptomycin (Gibco-BRL, Life Technologies). The bone marrow suspensions were cultured in polystyrene 6-well dishes and nonadherent cells were removed from the cultures after 2 days by a series of phosphatebuffered saline (PBS) washes and subsequent medium changes. Adherent cells were expanded as monolayer cultures in a 5% CO2/95% air atmosphere at 37°C with medium changes every 3 days. The confluent cells were dissociated with trypsin and subcultured in new 6-well culture dishes at a plating density of 6 ⫻ 104 cells/dish.
numbers because of their proliferative capacity, and ability to maintain their functionality after culture expansion and cryopreservation. Thus, MSCs are thought to be a readily available and abundant source of cells for tissue-engineering applications.3-5 Synthetic bone substitutes such as hydroxyapatite/-tricalcium phosphate (HA/TCP) and mixtures may be alternatives to these biological grafts with their bioactive and osteoconductive properties. Calcium phosphate bioceramics have shown good results in many clinical applications.6-8 MSCs, when combined with porous, biphasic calcium phosphate ceramics, namely the HA/TCP ceramics of the composition 60% HA/40% TCP, have been shown to conduct bone formation in large longbone defects.9,10 Yoshikawa et al.11 reported that HA loaded with MSCs has osteogenic potential comparable with autogenous particulate cancellous bone and marrow (PCBM). Bio-Oss is natural bovine bone mineral (NBM) with osteoconductive properties and high biocompatibility.12 It has been tested in more than 15 randomized clinical trials registered in the Cochrane library and thus is one of the best-documented biomaterials. Controversy remains as to whether this graft source is truly resorbable.12 However, these materials are not osteogenic (osteoinductive) and therefore applications are limited. This study was designed to compare MSC-based alveolar bone regeneration in a canine alveolar throughand-through defect model between HA/TCP and NBM. MSCs were loaded on the HA/TCP and natural bovine bone mineral matrices and compared histologically. This preclinical investigation was performed to define the safety and efficacy of the HA/TCP scaffold in alveolar bone repair.
In vitro osteogenic differentiation To evaluate in vitro osteogenic potential of the isolated cells, confluent passage 3 culture was used. For this purpose, the osteogenic medium consisted of the control medium supplemented with 0.2 mM ascorbic acid (Sigma, St. Louis, MO), 10 mM Na-b-glycerophosphate (Sigma), and 10⫺8 M dexamethasone (Sigma) was substituted as culture media. The cultures were then incubated at 37oC and 5% CO2 for 21 days. At the end of the cultivation period, the cells were fixed with 10% formalin for 10 minutes and stained with alizarin red (Sigma) for 15 minutes at room temperature so that the mineralized matrix of the bone could be examined. The cells were also used for RNA extraction and reverse transcriptase–polymerase chain reaction (RT-PCR) analysis of osteocytic gene expression. According to the results, osteopontin, Col I A1, and mRNA were largely produced after the third week in an osteoinductive medium (Fig. 1).
MATERIALS AND METHODS Mesenchymal stem cell isolation and cultivation Mongrel dogs with average weight of 15 to 25 kg were used for this experiment. This study was performed in accordance with the regulations and approval of the Institutional Animal Care and Use Committee of the Tehran University of Medical Sciences and conformed to standards of Association for Assessment and Accreditation of Laboratory Animal Care. The animals were housed for 1 week to become acclimatized to housing and diet. Throughout the experiments the animals were monitored for general appearance, activity, exertion, and weight. Under general anesthesia, bone marrow aspirate (about 10 mL) was drawn from the canine humerus, collected into a 50-mL tube containing 7500 units heparin, and transferred on ice to the cell culture facility of Royan Institute. In the cell culture lab, canine MSCs were isolated according to Kadiyala et al.13 with some modification. The medium consisted
Implant preparation Porous HA/TCP ceramic matrices were manufactured to 3 ⫻ 3 ⫻ 3 mm (width ⫻ 1ength ⫻ height) diameter. The material consisted of 60% HA and 40% -TCP with a mean pore size of 300 to 500 m. Bio-Oss (Geistlich, Osteohealth Biomaterials, Bern, Switzerland) retains the natural mineral content of the bone, which has a more complex composition than dose hydroxyapatites.4 The measurement of crystalline size of natural bone mineral (Bio-Oss) has shown the same characteristic of the tiny crystal size observed in normal human bone. The spongiosa structure demonstrates a wide interconnected pore system (300 to 1500 ) with the crystalline size of 10 to 60 nm. HA/TCP preparation. In all cases, 1 day before transplantation, implants were loaded by the cells obtained from the third subculture. HA/TCP cubes (3 ⫻ 3 ⫻ 3 mm) were first coated with purified bovine collagen I (Vitrogen, Chohesion, Sweden) and then loaded
e16
Jafarian et al.
OOOOE May 2008
Fig. 1. Reverse transcription–polymerase chain reaction analysis. Osteopontin and collagen expression after 3 weeks.
with MSCs by placing 5 ⫻ 105 cells in 0.2 mL DMEM on the top of the cubes. Cells were allowed to attach to the ceramics for 2 hours at 37°C before implantation. Collagen-coated HA/TCP cubes without cells were taken as controls. Bio-Oss preparation. Bio-Oss granules (1-2 mm) were similarly coated with purified bovine collagen I and for each amount of granule (3-4 granules) equal to a 3-mm cube of HA/TCP, 5 ⫻ 105 cells in 0.2 mL DMEM were added. Cells were allowed to attach to the ceramics for 2 hours at 37°C before implantation. Collagen-coated Bio-Oss granules without cells were taken as controls. Scaffold decalcification Samples were fixed overnight in 10% formaldehyde in PBS buffer decalcified in 10% EDTA for 24 hours, washed with tap water, incubated in an ascending row of isopropanol (Merck, Darmstadt, Germany), and finally embedded in paraffin wax (Leica, Bensheim, Germany). Sections were stained with hematoxylin and eosin (H&E) (Waldeck GmbH&Co Division Chroma, Munster, Germany). Three representative sequential sections with 5-mm thickness at a defined depth of 0.75 mm were selected for evaluation. The samples were examined at a magnification of ⫻100 with light microscopy (Fig. 2). Scanning electron microscopy analysis HA/TCP blocks subcultured with MSCs 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 a scanning electron microscope (SEM). SEM images were obtained and digitally recorded using a Jeol 660 instrument (Jeol USA Inc., Peabody, MA). Implantation of scaffold/ MSCs constructs in canine mandible defects In vivo bone formation is the result of a complex process that can be influenced by (1) the biomaterial itself, (2) the host-dependent parameters, and (3) the donor cells including their osteogenic cell commitment. In vivo bone formation can be influenced by the type and form of the biomaterial, the pore size (macro and micropores) and pore distribution, the interconnectivity, and the resorption of the biomaterial. Four adult mongrel dogs with healthy teeth, weighing between 20 and 30 kg, were used in this study. The dogs were premedicated with XylazineHCl (1 mg/kg) (Xylazine 2%, Alfasan, Woerden, The Netherlands) intramuscularly and atropine sulphate (0.05 mg/kg) (Atropin 0.5, Daroupakhsh Pharmaceutical Mfg Co, Tehran, Iran) subcutaneously. This was followed by general anesthesia with sodium thiopental (10 mg/kg) (Nesdonal, Specia, France) intravenously and oroendotracheal intubation. After induction of general anesthesia, Infiltration anesthesia was applied to the submandibular body area. A submandibular incision was made in the bilateral mandibular angle area and layered dissection was performed through the mandibular bone. Two 10-mm circular
OOOOE Volume 105, Number 5
Jafarian et al. e17
Fig. 2. A, Alizarin red staining. The arrow shows nodule of in vitro bone formation. B, Undifferentiated canine mesenchymal stem cells 13 days after cultivation. C, Clusters of bone marrow stem cells could be seen on the surfaces of decalcified Kasios (1/100). D, Higher magnification of decalcified Kasios cubes loaded with canine MSCs (1/30).
through-and-through defects were created in each side of the mandible using a dental hand piece and a trephine bur (Messeinger, Dusseldorf, Germany) under constant irrigation with sterile saline. The trephined bone was removed from the surgical field. The right mandibular defect was filled with 3 Kasios blocks with and without MSCs respectively and the left side was filled with Bio-Oss– enriched MSCs and pure Bio-Oss scaffold. All samples were prepared with purified bovine collagen I (Vitrogen). The periosteum and skin over the defects were approximated in a layered fashion (Vicryls 3.0; Nylon 5.0; Ethicon GmbH & Co. KG, Norderstedt, Germany). For the evaluation of the ectopic bone formation, layered dissection was performed through investing fascia of the masseter muscle. Blunt dissection with a curved hemostat created a tunnel pouch that measured 5 ⫻ 5 mm. In the right side of the mandible, 2 cubes of HA/TCP granules were embedded with microforceps and in the other side the same amount of the Bio-Oss granules were embedded. The pouch was closed in the layered fashion with resorbable sutures. At the
end of the surgery the animals received a single intramuscular injection of antibiotics (ampicillin 100 mg/kg). Each dog was humanely killed 6 weeks after insertion of the scaffold loaded with MSCs. The animals were killed with an overdose of sodium thiopental and subsequently perfused through the carotid arteries with a fixative consisting of a mixture of glutaraldehyde (5%) and formaldehyde (4%) buffered to pH 7.2. The mandibles were removed and placed in 10% formalin for an additional 10 days and decalcified in formic acid for 24 days. The specimens were washed with tap water, dehydrated with ascending concentrations of ethyl alcohol, cleared in xylene, infiltrated with paraffin, and processed for histologic evaluation. Decalcified coronal 5-m sections that incorporated the whole boney defect and muscular implants were prepared and stained using H&E. The histomorphometric data obtained from each specimen were recorded with a computerized image analysis system (Image-Pro Plust, Media Cybernetic, Silver Spring, MD). Slides taken at ⫻10 magnification were digitized with a solid-state,
e18
Jafarian et al.
35-mm slide scanner and a CCD linear photo diode array interfaced with a computer. The percent area of newly generated tissue that consisted of mineralized bone and marrow space in each histological section of the control site was calculated relative to the circular area of the host bone defect; this latter volume was taken as 100%. Immunohistochemistry The immunohistochemical procedure was performed on sections in which bone formation could be observed histologically. After the samples had been cut to 5 m, they were dyed using antibodies for osteonectin (Novacastra Laboratories Ltd., Newcastle, England) and osteopontin (Novacastra Laboratories Ltd.), then the primary antibodies against osteonectin and osteopontin were added (concentrations: osteonectin 1:500 and osteopontin 1:100). To enable a colored presentation, a secondary antibody (DAKO, Glostrup, Denmark) had to be added. Finally, the addition of StreptAB/HRP (DAKO) made it possible to bind the actual dye AEC⫹ (DAKO). The procedure was completed by H&E counterstaining. All samples were accompanied by a negative control. The stained sections were analyzed by optical microscope and were photographed. Statistical analysis All data are presented as means and standard deviations. The data were subjected to statistical analysis using the Mann-Whitney test following the KruskalWallis test (intergroup comparison) and the Wilcoxon test (intragroup comparison). Differences at P ⬍ .05 were considered significant. Paired differences were calculated and tested for normality using the Kolmogorov-Smirnoff goodness-of-fit test. Calculations were performed using the SPSS statistical package (SPSS 11, SPSS Inc., Chicago, IL). RESULTS SEM evaluation Qualitative evaluation of cell loading from SEM images of representative samples demonstrated greater cellular adherence of cMSCs in the HA/TCP ceramics. Cell adhesion after this relatively short loading period in serum-free conditions was represented by independent cell attachment to the matrix with limited spreading. Bio-Oss granules showed less cellular adhesion and the distribution of the cells were lower than the Kasios block (Fig. 3). Histologic evaluation Muscle specimens. A histological examination of the nodules revealed that they were encapsulated in a fibrous capsule, and that there was trabecular bone as
OOOOE May 2008
well as an amorphous calcified matrix in the nodules of all the samples. The trabeculae included many osteocytes and were regularly lined with many osteoblasts, indicating bone-forming activity. At the periphery of the nodules, the bone had a laminar pattern similar to normal bone. There was no evidence of inflammation or foreign-body reaction in the host tissue adjacent to the new bone, nor was there any evidence of cartilage generation. The percentages of newly formed bone were 29.12% in the HA/TCP/MSCs implants and 28.18% in the Bio-Oss/MSCs implants, while in the control cell free matrices were 23.57% and 24.17% respectively (Fig. 4). This demonstrated that there were differences in the amount of new bone formed in response to the different cells, but it is not statistically significant (P ⬎ .05). In the pores, bone tissue together with cuboidal active osteoblasts was observed in contact with the surfaces of the pores in HA/TCP. The osteoblast layer was in close contact with the implanted material. There was no intervening fibrous tissue between the ceramic and de novo bone. Regenerated bone marrow was observed in association with the new bone formation inside some porous regions of the HA/TCP (Fig. 5). In this bone-formation process, typical cartilage formation was not found at any time, and thus the process was the so-called intramembranous bone formation. There was a generalized absence of inflammation in these sections with the pores of the matrix filled with a vascularized dense connective tissue devoid of adipose or cartilage tissue. In contrast, woven bone formation within the pores was evident in both of the HA/TCP and Bio-Oss matrix groups. Bone defect specimens. Healing was uneventful in all of the surgical sites. In all of the groups, the newly generated tissue was of varying size, and it consisted of thin pieces of mineralized bone and large marrow spaces with fat cells and some hematopoietic cells. Cuboidal osteoblast-like cells were generally found to be actively laying down bone to varying degrees along the mineralized bone in all of the groups. More obvious findings of osteoblast-like cells were often found in the MSC-seeded groups. Neither foreign body reaction nor severe inflammation was seen in each of the specimens evaluated. Also, direct bone-biomaterial contact was seen whenever new bone formation occurred more in the NBM ⫹ MSCs and HA/TCP⫹ MSCs groups. Degree of trabecular bone formation evaluated with light microscopy showed significant differences between Kasios ⫹ MSCs and NBM ⫹ MSCs with control groups respectively. Adding MSCs caused a slight tendency toward more frequent lamellar trabeculae. In this study, a total of 4 experimental animals and 16 throughand-through defects were analyzed for both histological and histochemical evaluation of mandibular defect. At
OOOOE Volume 105, Number 5
Jafarian et al. e19
Fig. 3. Cells cultured on the biphasic bone substitute. A, Clusters of bone marrow stem cells could be seen everywhere on the surface pores of Kasios (1/100). B, Higher magnification of Kasios cubes. The microprosisty can springe for the MSCs and harvest them well. Bone marrow mesenchymal cell trapped in the collagen network and Kasios (1/10). C, Cells cultured on Bio-Oss; low magnification 1/300; MSCs spread more slowly and formed a confluent monolayer by Day 2 on cell culture plate (arrow). D, Minimal settling of the cells on the Bio-Oss granules (arrow) (high magnification 1/100).
6 weeks, Bio-Oss and HA/TCP were found within the osteogenic matrix in the main portion of the mandibular bone defects. The newly formed bone with osteocytes could be mainly seen at the periphery and osteoblastlike cells exhibiting dense arrangement adjacent to newly formed bone suggest continuing bone formation through the center of the defect (Fig. 6). However, less bone-formative activity was evident at the center of the defects. There was no evidence of fatty marrow or cartilage formation. The newly formed bone consisted of lamellar and trabecular bone. Histomorphometric evaluation The results of the histomorphometric analysis for animals implanted with Bio-Oss, HA/TCP, HA/TCP/ cell, and Bio-Oss/cell are shown in Table I. In the present study, new bone (%) in the defects was defined as bone trabeculae and marrow spaces, since at 6 weeks postsurgery Bio-Oss and HA/TCP have shown mild resorption activity with the osteoclasts at the periphery
and occupy a substantial portion of the defect space. The following were measured: area of new bone (BF), reported as a percentage of the area of the whole section, excluding the original cortical bone. The defect sites exhibited variable healing between specimens, ranging from limited bone formation to 70% of whole specimens. At 6 weeks postsurgery, the mean percentage of bone formation for HA/TCP scaffold loaded with MSCs amounted to 65.78% and for the control defects with pure HA/TCP scaffold coated with collagen type I was 44.90%. The mean percentage of bone fill in Bio-Oss loaded with MSCs was 50.31% and in control specimens was 36.83% (Fig. 7). The percentage of bone fill had statistically significant differences between the HA/TCP/MSC group and pure HA/TCP scaffold (P ⬍ .05). The MSC-loaded scaffold showed more bone fill than the pure scaffold group. HA/TCP cultured with MSCs indicated greater percentage of bone fill than Bio-Oss loaded with MSCs but there was no statistically significant difference (P ⬎ .05).
e20
Jafarian et al.
OOOOE May 2008
Fig. 4. Histomorphometeric data of the percentage of bone fill (BF) percent in each group revealed the higher percentage of bone fill in the Biphasic bone substitutes (Kasios). 1 ⫽ Biphasic, 2 ⫽ NBM, 3 ⫽ Biphasic ⫹ cells, 4 ⫽ NBM ⫹ cells.
Immunohistochemical observation Positive immunoreaction for osteopontin and osteonectin was evident in all 4 groups. Intense osteopontin and osteonectin were observed near newly formed bone and cells within it. An even immunoreaction was seen throughout the defect. Together with histological observation suggests bone formation and mineralization proceed in the central direction from the margin of the defect (Fig. 8). DISCUSSION The current trend in reconstruction of the facial skeleton is directed toward the enhancement of bone regeneration to avoid the need for harvesting autogenous bone grafts. One of these innovative approaches is to use tissue-engineered growth of bone, by which cultivated bone cells in scaffolds are delivered to a skeletal defect to form bone at the site of implantation. For this purpose, the scaffold should be biocompatible and serve as a proper matrix for the cells to produce the new structural environment of extracellular matrix ex vivo.14 The tissue-specific differentiation of bone marrow– derived MSCs in a controlled manner (e.g., osteogenesis) is dependent on the matrix or local environment in which the hMSC resides.15 Selection of a matrix carrier involves consideration of the matrix’s role as a scaffold for physical support and host tissue integration, as well as its ability to support or synergize
the osteoinductive program of the implanted MSCs. A majority of the currently available matrix materials for bone grafting are osteoconductive materials that support bone formation by acting as a scaffold for angiogenesis, cell recruitment, and ultimately osteogenesis by host cells.16 For MSC-directed bone repair to be clinically successful, a scaffold must be identified and optimized to support cellular adherence, cell recruitment, osteoinduction, and osteoconduction. These results contribute to a growing literature that demonstrates that the MSCs are able to form bone after implantation on porous HA/TCP matrices.11,17 In this study, to enhance cell-loading efficacy, the scaffolds were coated with collagen I gel a day before they were loaded by MSCs; light and scanning electron microscopic study confirmed that the cells occupied the spaces of the scaffolds. The cells were not treated by osteogenic medium for the implantation in the animal models and used as undifferentiated cells. The osteogenic medium only demonstrated the in vitro capacity of osteogenic differentiation. It was thought that the occurrence of osteogenesis in the implantation site of mandibular bone defect would be as a result of combined action of bone microenvironment and MSCs loaded on scaffold materials. It is clearly beneficial for any tissueengineered device to allow the host cells to contribute to regeneration of the desired tissue. Comparison
OOOOE Volume 105, Number 5
Jafarian et al. e21
Fig. 5. Representative histomicrograph of the bone formation around the cuboidal osteoblast-like cells in the masseter muscle. Bone is stained pink, ceramic has a shadowy white appearance, loose connective tissue is stained light pink, cells are stained a dark pink (hematoxylin-eosin stain). A, magnification ⫻10. B, Magnification ⫻100. C, Magnification ⫻10). D, Bone formation within the NBM (magnification ⫻10). Arrows indicate the new bone formation in direct bone contact with the bone substitutes.
of cell-free and MSC-loaded matrices in this study demonstrated a clear difference in bone formation, confirming the osteogenic potency of MSC-loaded constructs. The MSC-loaded implants were well integrated with the surrounding host bone. In most cases, the cell-free implants showed slight evidence of osteoconduction from the host tissue, which may have been influenced by collagen type I as an inductive factor, but were loose in the defect site (particularly in the Bio-Oss group) and had not formed significant bonding with the surrounding bone. It is possible that one important effect of the MSC/matrix is the encouragement of the regional or local osteogenesis. Bone was formed by an intramembranous pathway exclusive of evidence of chondrogenesis at 6 weeks. Under the current conditions, MSC osteogenesis in orthopedic sites occurred by a direct conversion of mesenchymal cells into osteoblasts rather than by an endochondral sequence.16 The muscle specimen showed that cell-loaded Kasios biphasic
matrices have more bone formation percentage (BF ⫽ 29.11%) at the end of the study but there is no significant statistically difference between groups (P ⬎ .05). The ranges of data between the maximum amounts of the BF in each group were lower than 20%; the authors could not find a positive inductive factor for MSCs seeded on the natural and synthetic scaffold. This minimal amount of bone formation in a non-mineral environment may be due to the collagen I, which acts as a cell carrier and presented in the control groups. The fate of the distant ectopic bone formation with the use of the MSCs may decrease with the findings of this study. The authors suggested MSCs could not impose a great influence on the bone formation in a muscular environment. Immunohistochemical analysis revealed bone matrix protein expression through the defect in all groups with higher enhancement in the periphery. The bone defect analysis demonstrated Kasios loaded with MSCs showed greater bone formation (65.78%) after 6 weeks
e22
OOOOE May 2008
Jafarian et al.
Fig. 6. A, Cuboidal osteoblast-like cells were generally found to be actively laying down bone to varying degrees along the mineralized bone in Kasios group. B, Bone formation around the cube of Kasios in the center of the defect far away from the margin of the original defect. C, D, Osteoblastic activity was observed at the test site in all groups. Large marrow spaces with fat cells and some hematopoietic cells were found in all of the groups. Arrow shows the original bone around the defect and the newly formed bone through the center of the defect. (hematoxylin and eosin, original magnification, ⫻100).
Table I. Mean percentage of bone fill in the dog mandibular bone through-and-through defects
HA/TCP Bio-Oss HA/TCP ⫹ cell BioOss ⫹ cell
Mean of bone formation in bone defect ⫾ SD
Mean of bone formation in muscle specimen ⫾ SD
44.9% ⫾ 13.45% 36.84% ⫾ 8.73% 65.78% ⫾ 4.94% 50.31% ⫾ 6.97%
23.55% ⫾ 4.99% 24.16% ⫾ 4.22% 29.11% ⫾ 6.10% 28.18% ⫾ 5.20%
The mean percentage of bone fill in biphasic bone substitute is more than other groups. HA/TCP, hydroxyapatite/-tricalcium phosphate.
(Fig. 7). Bone formation between the MSC groups (Kasios and Bio-Oss) and between pure scaffold groups (Kasios and Bio-Oss) did not reveal significant differences (P ⬎ .05). Kasios plus MSC had a significant statistical difference with other groups (P ⬍ .05). In the present investigation, the resorption process of the demineralized bone mineral and
HA/TCP particles did not follow a specific pattern. Implantation of culture-expanded autologous MSCs offers the advantage of directly delivering the cellular machinery responsible for synthesizing new bone and circumventing the otherwise slow steps leading to natural or enhanced bone repair. By incorporating living cells with specifically designed matrices, the shortcomings of osteoinductive factors alone to affect permanent bone repair may be overcome. CONCLUSION In the present study, MSCs from canine marrow were isolated, culture expanded, and loaded into either of 2 commercially available osteoconductive bone substitutes in order to compare their bone regeneration potential in through-and-thorough bone defects in combination with MSCs. Our results indicated that HA/ TCP matrices loaded with MSCs provides better conditions for bone regeneration than the other, although the difference was not statistically significant.
OOOOE Volume 105, Number 5
Jafarian et al. e23
Fig. 7. Histomorphometeric data of the percentage of bone fill (BF). Percent in each group revealed the higher percentage of bone fill in the biphasic bone substitutes (Kasios). 1 ⫽ Biphasic, 2 ⫽ NBM, 3 ⫽ Biphasic ⫹ cells, 4 ⫽ NBM ⫹ cells.
Fig. 8. Positive immunoreactions around the edges of bone substitutes and central direction of the bone formation through the center of the defect (arrow).
Special thanks to Reza Vahid for his professional editing of the article. This study was supported by a grant-in-aid of the Iranian Center for Dental Research (ICDR), Shahid Beheshti University of Medical Sciences and Royan Institute.
REFERENCES 1. Block MS, Kent JN. Sinus augmentation for dental implants: the use of autogenous bone. J Oral Maxillofac Surg 1997;55: 1281-6. 2. Hollinger JO, Winn S, Bonadio J. Options for tissue engineering
e24
3.
4.
5.
6.
7.
8.
9.
10.
OOOOE May 2008
Jafarian et al.
to address challenges of the aging skeleton. Tissue Engineering 2000;6:341-50. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143-7. Dong J, Kojima H, Uemura T, Kikuchi M, Tateishi T, Tanaka J. In vivo evaluation of a novel porous hydroxyapatite to sustain osteogenesis of transplanted bone marrow-derived osteoblastic cells. J Biomed Mater Res 2001;57:208-16. Al Ruhaimi KA. Bone graft substitutes: a comperative qualitative histologic review of current osteoconductive grafting materials. Int J Oral Maxillofac Implants 2001;16:105-14. Meyer U, Joos U, Wiesmann HP. Biological and biophysical principles in extracorporal bone tissue engineering. Part I. Int J Oral Maxillofac Surg 2004;33:325-32. Niemeyer P, Krause U, Kasten P, Kreuz PC, Henle P, Südkamp NP, et al. Mesenchymal stem cell-based HLA-independent cell therapy for tissue engineering of bone and cartilage. Current Stem Cell Res Therapy 2006;1:21-7. Ito K, Yamada Y, Nagasaka T, Baba S, Ueda M. Osteogenic potential of injectable tissue-engineered bone: a comparison among autogenous bone, bone substitute (Bio-Oss), platelet-rich plasma, and tissue-engineered bone with respect to their mechanical properties and histological findings. J Biomed Mater Res 2005;73:63-72. Ogura N, Kawada M, Chang WJ, Zhang Q, Lee SY, Kondoh T, et al. Differentiation of the human mesenchymal stem cells derived from bone marrow and enhancement of cell attachment by fibronectin. J Oral Sci 2004;46:207-13. Jenssen SS, Broggini N, Weibrich G, Hjorting-Hanssen E, Shenk R, Buser D. Bone regeneration in standardized bone defects with autograft or bone substitute in combination with palatlete concentrate. A histologic and histomorphometeric study in the
11.
12.
13.
14.
15.
16.
17.
mandible of minipigs. Int J Oral Maxillofac Implants 2005;20: 703-12. Yoshikawa T, Ohgishi H, Tamai S. Immediate bone forming capability of prefabricated osteogenic hydroxyapatite. J Biomed Mater Res 1996;32:481-92. Piatteli M, Favero GA, Scranio A, Orsinic G, Piatteli A. Bone reaction to anorganic bovine bone (Bio-Oss) used in sinus augmentation procedure: a histologic long term report of 20 cases in humans. Int J Oral Maxillofac Implants 1999;14:835-40. Kadiyala S, Jaiswal N, Bruder SP. Culture-expanded, bone marrow-derived mesenchymal stem cells can regenerate a criticalsized segmental bone defect. Tissue Eng 1997;3(2):173-85. Wiedmann-Al-Ahmad M, Gutwald R, Gellrich NC, Hubner U, Schmelzeisen R. Search for ideal biomaterials to cultivate human osteoblast-like cells for reconstructive surgery. J Mater Sci Mater Med 2005;16:57-66. Cooper LF, Harris CT, Bruder SP, Kowalski R, Kadiyala S. Incipient analysis of mesenchymal stem-cell-derived osteogenesis. J Dent Res 2001;80:314-20. Nomura T, Katz L, Powers MP, Saito C. Evaluation of micromechanical elastic properties of potential bone-grafting materials. J Biomed Mater Res B Appl Biomater 2004;73:29-34. Schopper C, Ziya-Ghazvini F, Goriwoda W, Moser D, Wanschitz F, Spassova E, et al. HA/TCP compounding of a porous CaP biomaterial improves bone formation and scaffold degradation—a long-term histological study. J Biomed Mater Res B Appl Biomater 2005;74:458-67.
Reprint requests: Mohamadreza Baghaban Eslaminejad, PhD Royan Institute PO Box 19395-4644 Tehran, Iran [email protected]
Objective. This study was designed to compare mesenchymal stem cell (MSC)-based alveolar bone regeneration in biphasic bone substitutes and natural bone mineral in a canine full-thickness alveolar defect model. Materials and methods. MSCs were isolated from bone marrow aspirates and culture expanded through 3 successive subcultures. The bone differentiation potential of third passage cells was evaluated and confirmed in vitro before cells were used in the transplantation experiment. Undifferentiated cells were then incubated with 3 ⫻ 3 ⫻ 3 mm3 hydroxyapatite/-tricalcium phosphate (HA/TCP) matrices (Kasios, Lanauguet, France) and 1- to 2-mm Bio-Oss spongiosa (Geistlich Biomaterials, Osteohealth, Switzerland), which is a natural bovine bone mineral (NBM). Kasios/ cell, Kasios alone, Bio-Oss/cell, and Bio-Oss alone were implanted in masseter muscle and 4 cylindrical (10-mm diameter) through-and-through bilateral mandibular body defects in 4 mongrel dogs. Histomorphometric analysis was performed 6 weeks after insertion of the scaffold loaded with MSCs. Results. H&E staining of the decalcified scaffold and scanning electron microscopy demonstrated large MSC coverage of the HA/TCP and Bio-Oss. Cell-loaded Kasios matrices showed the greatest amount of the bone regeneration among the groups in both the muscle (29.11%) and the bone specimens (65.78%). Cell-free biphasic scaffold revealed 44.9% bone fill in bone defects and 23.55% in muscle specimen, and Bio-Oss alone matrices had the least amount of new bone formation: 36.84% and 24.16% in bone and muscle specimens respectively. Kasios loaded with MSCs demonstrated more bone regeneration than Bio-Oss/cell but there was no significant statistical difference (P ⬎ .05). Conclusions. New biphasic synthetic bone substitutes may offer better conditions for bone regeneration than traditional bone substitute in combination with MSCs. They remained in the defect and contributed to bone regeneration. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:e14-e24)
a
Assistant Professor, Oral and Maxillofacial Surgery, Department of Oral and Maxillofacial Surgery, Taleghani University Hospital, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. b Assistant Professor of Anatomy and Embryology, Stem Cell Department, Royan Institute. c Chief Resident, Oral and Maxillofacial Surgery, Taleghani University Hospital, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. d Assistant Professor, Oral and Maxillofacial Pathology, Department of Oral and Maxillofacial Pathology, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. e Assistant Professor of Veterinary Surgery, Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Tehran. f Dentist, Iranian Center for Dental Research, Shahid Beheshti University of Medical Sciences. Received for publication Sep 25, 2007; returned for revision Dec 15, 2007; accepted for publication Jan 7, 2008. 1079-2104/$ - see front matter © 2008 Mosby, Inc. All rights reserved. doi:10.1016/j.tripleo.2008.01.010
e14
For the repair of bone defects, a tissue-engineering approach would be to combine cells capable of osteogenic activity with an appropriate scaffolding material to stimulate bone regeneration and repair. Dental and craniofacial surgeons must often restore aesthetics and function to their patients, despite an insufficient amount of bone. For this reason, bone regeneration procedures are an integral part of dental and craniofacial therapy.1 Although alloplastic materials including synthetic materials and natural bone– derived materials are used, autogenous bone remains the standard of care for regeneration of significant volumetric defects of the maxilla or mandible. Because of the limited amount of autogenous bone available for reconstructive use, other therapies are being explored.1,2 Bone marrow– derived mesenchymal stem cells (MSCs) are multipotential cells that are capable of differentiating into, at a minimum, osteoblasts, chondrocytes, adipocytes, tenocytes, and myoblasts. From a small volume of bone marrow, MSCs can be isolated and cultures expanded into large
OOOOE Volume 105, Number 5
Jafarian et al. e15
of Dulbecco modified Eagle medium (DMEM; GibcoBRL, Life Technologies, Grand Island, NY) supplemented with 15% fetal bovine serum (FBS; Dainippon Pharmaceutical, Osaka, Japan) and 100 g/mL penicillin-streptomycin (Gibco-BRL, Life Technologies). The bone marrow suspensions were cultured in polystyrene 6-well dishes and nonadherent cells were removed from the cultures after 2 days by a series of phosphatebuffered saline (PBS) washes and subsequent medium changes. Adherent cells were expanded as monolayer cultures in a 5% CO2/95% air atmosphere at 37°C with medium changes every 3 days. The confluent cells were dissociated with trypsin and subcultured in new 6-well culture dishes at a plating density of 6 ⫻ 104 cells/dish.
numbers because of their proliferative capacity, and ability to maintain their functionality after culture expansion and cryopreservation. Thus, MSCs are thought to be a readily available and abundant source of cells for tissue-engineering applications.3-5 Synthetic bone substitutes such as hydroxyapatite/-tricalcium phosphate (HA/TCP) and mixtures may be alternatives to these biological grafts with their bioactive and osteoconductive properties. Calcium phosphate bioceramics have shown good results in many clinical applications.6-8 MSCs, when combined with porous, biphasic calcium phosphate ceramics, namely the HA/TCP ceramics of the composition 60% HA/40% TCP, have been shown to conduct bone formation in large longbone defects.9,10 Yoshikawa et al.11 reported that HA loaded with MSCs has osteogenic potential comparable with autogenous particulate cancellous bone and marrow (PCBM). Bio-Oss is natural bovine bone mineral (NBM) with osteoconductive properties and high biocompatibility.12 It has been tested in more than 15 randomized clinical trials registered in the Cochrane library and thus is one of the best-documented biomaterials. Controversy remains as to whether this graft source is truly resorbable.12 However, these materials are not osteogenic (osteoinductive) and therefore applications are limited. This study was designed to compare MSC-based alveolar bone regeneration in a canine alveolar throughand-through defect model between HA/TCP and NBM. MSCs were loaded on the HA/TCP and natural bovine bone mineral matrices and compared histologically. This preclinical investigation was performed to define the safety and efficacy of the HA/TCP scaffold in alveolar bone repair.
In vitro osteogenic differentiation To evaluate in vitro osteogenic potential of the isolated cells, confluent passage 3 culture was used. For this purpose, the osteogenic medium consisted of the control medium supplemented with 0.2 mM ascorbic acid (Sigma, St. Louis, MO), 10 mM Na-b-glycerophosphate (Sigma), and 10⫺8 M dexamethasone (Sigma) was substituted as culture media. The cultures were then incubated at 37oC and 5% CO2 for 21 days. At the end of the cultivation period, the cells were fixed with 10% formalin for 10 minutes and stained with alizarin red (Sigma) for 15 minutes at room temperature so that the mineralized matrix of the bone could be examined. The cells were also used for RNA extraction and reverse transcriptase–polymerase chain reaction (RT-PCR) analysis of osteocytic gene expression. According to the results, osteopontin, Col I A1, and mRNA were largely produced after the third week in an osteoinductive medium (Fig. 1).
MATERIALS AND METHODS Mesenchymal stem cell isolation and cultivation Mongrel dogs with average weight of 15 to 25 kg were used for this experiment. This study was performed in accordance with the regulations and approval of the Institutional Animal Care and Use Committee of the Tehran University of Medical Sciences and conformed to standards of Association for Assessment and Accreditation of Laboratory Animal Care. The animals were housed for 1 week to become acclimatized to housing and diet. Throughout the experiments the animals were monitored for general appearance, activity, exertion, and weight. Under general anesthesia, bone marrow aspirate (about 10 mL) was drawn from the canine humerus, collected into a 50-mL tube containing 7500 units heparin, and transferred on ice to the cell culture facility of Royan Institute. In the cell culture lab, canine MSCs were isolated according to Kadiyala et al.13 with some modification. The medium consisted
Implant preparation Porous HA/TCP ceramic matrices were manufactured to 3 ⫻ 3 ⫻ 3 mm (width ⫻ 1ength ⫻ height) diameter. The material consisted of 60% HA and 40% -TCP with a mean pore size of 300 to 500 m. Bio-Oss (Geistlich, Osteohealth Biomaterials, Bern, Switzerland) retains the natural mineral content of the bone, which has a more complex composition than dose hydroxyapatites.4 The measurement of crystalline size of natural bone mineral (Bio-Oss) has shown the same characteristic of the tiny crystal size observed in normal human bone. The spongiosa structure demonstrates a wide interconnected pore system (300 to 1500 ) with the crystalline size of 10 to 60 nm. HA/TCP preparation. In all cases, 1 day before transplantation, implants were loaded by the cells obtained from the third subculture. HA/TCP cubes (3 ⫻ 3 ⫻ 3 mm) were first coated with purified bovine collagen I (Vitrogen, Chohesion, Sweden) and then loaded
e16
Jafarian et al.
OOOOE May 2008
Fig. 1. Reverse transcription–polymerase chain reaction analysis. Osteopontin and collagen expression after 3 weeks.
with MSCs by placing 5 ⫻ 105 cells in 0.2 mL DMEM on the top of the cubes. Cells were allowed to attach to the ceramics for 2 hours at 37°C before implantation. Collagen-coated HA/TCP cubes without cells were taken as controls. Bio-Oss preparation. Bio-Oss granules (1-2 mm) were similarly coated with purified bovine collagen I and for each amount of granule (3-4 granules) equal to a 3-mm cube of HA/TCP, 5 ⫻ 105 cells in 0.2 mL DMEM were added. Cells were allowed to attach to the ceramics for 2 hours at 37°C before implantation. Collagen-coated Bio-Oss granules without cells were taken as controls. Scaffold decalcification Samples were fixed overnight in 10% formaldehyde in PBS buffer decalcified in 10% EDTA for 24 hours, washed with tap water, incubated in an ascending row of isopropanol (Merck, Darmstadt, Germany), and finally embedded in paraffin wax (Leica, Bensheim, Germany). Sections were stained with hematoxylin and eosin (H&E) (Waldeck GmbH&Co Division Chroma, Munster, Germany). Three representative sequential sections with 5-mm thickness at a defined depth of 0.75 mm were selected for evaluation. The samples were examined at a magnification of ⫻100 with light microscopy (Fig. 2). Scanning electron microscopy analysis HA/TCP blocks subcultured with MSCs 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 a scanning electron microscope (SEM). SEM images were obtained and digitally recorded using a Jeol 660 instrument (Jeol USA Inc., Peabody, MA). Implantation of scaffold/ MSCs constructs in canine mandible defects In vivo bone formation is the result of a complex process that can be influenced by (1) the biomaterial itself, (2) the host-dependent parameters, and (3) the donor cells including their osteogenic cell commitment. In vivo bone formation can be influenced by the type and form of the biomaterial, the pore size (macro and micropores) and pore distribution, the interconnectivity, and the resorption of the biomaterial. Four adult mongrel dogs with healthy teeth, weighing between 20 and 30 kg, were used in this study. The dogs were premedicated with XylazineHCl (1 mg/kg) (Xylazine 2%, Alfasan, Woerden, The Netherlands) intramuscularly and atropine sulphate (0.05 mg/kg) (Atropin 0.5, Daroupakhsh Pharmaceutical Mfg Co, Tehran, Iran) subcutaneously. This was followed by general anesthesia with sodium thiopental (10 mg/kg) (Nesdonal, Specia, France) intravenously and oroendotracheal intubation. After induction of general anesthesia, Infiltration anesthesia was applied to the submandibular body area. A submandibular incision was made in the bilateral mandibular angle area and layered dissection was performed through the mandibular bone. Two 10-mm circular
OOOOE Volume 105, Number 5
Jafarian et al. e17
Fig. 2. A, Alizarin red staining. The arrow shows nodule of in vitro bone formation. B, Undifferentiated canine mesenchymal stem cells 13 days after cultivation. C, Clusters of bone marrow stem cells could be seen on the surfaces of decalcified Kasios (1/100). D, Higher magnification of decalcified Kasios cubes loaded with canine MSCs (1/30).
through-and-through defects were created in each side of the mandible using a dental hand piece and a trephine bur (Messeinger, Dusseldorf, Germany) under constant irrigation with sterile saline. The trephined bone was removed from the surgical field. The right mandibular defect was filled with 3 Kasios blocks with and without MSCs respectively and the left side was filled with Bio-Oss– enriched MSCs and pure Bio-Oss scaffold. All samples were prepared with purified bovine collagen I (Vitrogen). The periosteum and skin over the defects were approximated in a layered fashion (Vicryls 3.0; Nylon 5.0; Ethicon GmbH & Co. KG, Norderstedt, Germany). For the evaluation of the ectopic bone formation, layered dissection was performed through investing fascia of the masseter muscle. Blunt dissection with a curved hemostat created a tunnel pouch that measured 5 ⫻ 5 mm. In the right side of the mandible, 2 cubes of HA/TCP granules were embedded with microforceps and in the other side the same amount of the Bio-Oss granules were embedded. The pouch was closed in the layered fashion with resorbable sutures. At the
end of the surgery the animals received a single intramuscular injection of antibiotics (ampicillin 100 mg/kg). Each dog was humanely killed 6 weeks after insertion of the scaffold loaded with MSCs. The animals were killed with an overdose of sodium thiopental and subsequently perfused through the carotid arteries with a fixative consisting of a mixture of glutaraldehyde (5%) and formaldehyde (4%) buffered to pH 7.2. The mandibles were removed and placed in 10% formalin for an additional 10 days and decalcified in formic acid for 24 days. The specimens were washed with tap water, dehydrated with ascending concentrations of ethyl alcohol, cleared in xylene, infiltrated with paraffin, and processed for histologic evaluation. Decalcified coronal 5-m sections that incorporated the whole boney defect and muscular implants were prepared and stained using H&E. The histomorphometric data obtained from each specimen were recorded with a computerized image analysis system (Image-Pro Plust, Media Cybernetic, Silver Spring, MD). Slides taken at ⫻10 magnification were digitized with a solid-state,
e18
Jafarian et al.
35-mm slide scanner and a CCD linear photo diode array interfaced with a computer. The percent area of newly generated tissue that consisted of mineralized bone and marrow space in each histological section of the control site was calculated relative to the circular area of the host bone defect; this latter volume was taken as 100%. Immunohistochemistry The immunohistochemical procedure was performed on sections in which bone formation could be observed histologically. After the samples had been cut to 5 m, they were dyed using antibodies for osteonectin (Novacastra Laboratories Ltd., Newcastle, England) and osteopontin (Novacastra Laboratories Ltd.), then the primary antibodies against osteonectin and osteopontin were added (concentrations: osteonectin 1:500 and osteopontin 1:100). To enable a colored presentation, a secondary antibody (DAKO, Glostrup, Denmark) had to be added. Finally, the addition of StreptAB/HRP (DAKO) made it possible to bind the actual dye AEC⫹ (DAKO). The procedure was completed by H&E counterstaining. All samples were accompanied by a negative control. The stained sections were analyzed by optical microscope and were photographed. Statistical analysis All data are presented as means and standard deviations. The data were subjected to statistical analysis using the Mann-Whitney test following the KruskalWallis test (intergroup comparison) and the Wilcoxon test (intragroup comparison). Differences at P ⬍ .05 were considered significant. Paired differences were calculated and tested for normality using the Kolmogorov-Smirnoff goodness-of-fit test. Calculations were performed using the SPSS statistical package (SPSS 11, SPSS Inc., Chicago, IL). RESULTS SEM evaluation Qualitative evaluation of cell loading from SEM images of representative samples demonstrated greater cellular adherence of cMSCs in the HA/TCP ceramics. Cell adhesion after this relatively short loading period in serum-free conditions was represented by independent cell attachment to the matrix with limited spreading. Bio-Oss granules showed less cellular adhesion and the distribution of the cells were lower than the Kasios block (Fig. 3). Histologic evaluation Muscle specimens. A histological examination of the nodules revealed that they were encapsulated in a fibrous capsule, and that there was trabecular bone as
OOOOE May 2008
well as an amorphous calcified matrix in the nodules of all the samples. The trabeculae included many osteocytes and were regularly lined with many osteoblasts, indicating bone-forming activity. At the periphery of the nodules, the bone had a laminar pattern similar to normal bone. There was no evidence of inflammation or foreign-body reaction in the host tissue adjacent to the new bone, nor was there any evidence of cartilage generation. The percentages of newly formed bone were 29.12% in the HA/TCP/MSCs implants and 28.18% in the Bio-Oss/MSCs implants, while in the control cell free matrices were 23.57% and 24.17% respectively (Fig. 4). This demonstrated that there were differences in the amount of new bone formed in response to the different cells, but it is not statistically significant (P ⬎ .05). In the pores, bone tissue together with cuboidal active osteoblasts was observed in contact with the surfaces of the pores in HA/TCP. The osteoblast layer was in close contact with the implanted material. There was no intervening fibrous tissue between the ceramic and de novo bone. Regenerated bone marrow was observed in association with the new bone formation inside some porous regions of the HA/TCP (Fig. 5). In this bone-formation process, typical cartilage formation was not found at any time, and thus the process was the so-called intramembranous bone formation. There was a generalized absence of inflammation in these sections with the pores of the matrix filled with a vascularized dense connective tissue devoid of adipose or cartilage tissue. In contrast, woven bone formation within the pores was evident in both of the HA/TCP and Bio-Oss matrix groups. Bone defect specimens. Healing was uneventful in all of the surgical sites. In all of the groups, the newly generated tissue was of varying size, and it consisted of thin pieces of mineralized bone and large marrow spaces with fat cells and some hematopoietic cells. Cuboidal osteoblast-like cells were generally found to be actively laying down bone to varying degrees along the mineralized bone in all of the groups. More obvious findings of osteoblast-like cells were often found in the MSC-seeded groups. Neither foreign body reaction nor severe inflammation was seen in each of the specimens evaluated. Also, direct bone-biomaterial contact was seen whenever new bone formation occurred more in the NBM ⫹ MSCs and HA/TCP⫹ MSCs groups. Degree of trabecular bone formation evaluated with light microscopy showed significant differences between Kasios ⫹ MSCs and NBM ⫹ MSCs with control groups respectively. Adding MSCs caused a slight tendency toward more frequent lamellar trabeculae. In this study, a total of 4 experimental animals and 16 throughand-through defects were analyzed for both histological and histochemical evaluation of mandibular defect. At
OOOOE Volume 105, Number 5
Jafarian et al. e19
Fig. 3. Cells cultured on the biphasic bone substitute. A, Clusters of bone marrow stem cells could be seen everywhere on the surface pores of Kasios (1/100). B, Higher magnification of Kasios cubes. The microprosisty can springe for the MSCs and harvest them well. Bone marrow mesenchymal cell trapped in the collagen network and Kasios (1/10). C, Cells cultured on Bio-Oss; low magnification 1/300; MSCs spread more slowly and formed a confluent monolayer by Day 2 on cell culture plate (arrow). D, Minimal settling of the cells on the Bio-Oss granules (arrow) (high magnification 1/100).
6 weeks, Bio-Oss and HA/TCP were found within the osteogenic matrix in the main portion of the mandibular bone defects. The newly formed bone with osteocytes could be mainly seen at the periphery and osteoblastlike cells exhibiting dense arrangement adjacent to newly formed bone suggest continuing bone formation through the center of the defect (Fig. 6). However, less bone-formative activity was evident at the center of the defects. There was no evidence of fatty marrow or cartilage formation. The newly formed bone consisted of lamellar and trabecular bone. Histomorphometric evaluation The results of the histomorphometric analysis for animals implanted with Bio-Oss, HA/TCP, HA/TCP/ cell, and Bio-Oss/cell are shown in Table I. In the present study, new bone (%) in the defects was defined as bone trabeculae and marrow spaces, since at 6 weeks postsurgery Bio-Oss and HA/TCP have shown mild resorption activity with the osteoclasts at the periphery
and occupy a substantial portion of the defect space. The following were measured: area of new bone (BF), reported as a percentage of the area of the whole section, excluding the original cortical bone. The defect sites exhibited variable healing between specimens, ranging from limited bone formation to 70% of whole specimens. At 6 weeks postsurgery, the mean percentage of bone formation for HA/TCP scaffold loaded with MSCs amounted to 65.78% and for the control defects with pure HA/TCP scaffold coated with collagen type I was 44.90%. The mean percentage of bone fill in Bio-Oss loaded with MSCs was 50.31% and in control specimens was 36.83% (Fig. 7). The percentage of bone fill had statistically significant differences between the HA/TCP/MSC group and pure HA/TCP scaffold (P ⬍ .05). The MSC-loaded scaffold showed more bone fill than the pure scaffold group. HA/TCP cultured with MSCs indicated greater percentage of bone fill than Bio-Oss loaded with MSCs but there was no statistically significant difference (P ⬎ .05).
e20
Jafarian et al.
OOOOE May 2008
Fig. 4. Histomorphometeric data of the percentage of bone fill (BF) percent in each group revealed the higher percentage of bone fill in the Biphasic bone substitutes (Kasios). 1 ⫽ Biphasic, 2 ⫽ NBM, 3 ⫽ Biphasic ⫹ cells, 4 ⫽ NBM ⫹ cells.
Immunohistochemical observation Positive immunoreaction for osteopontin and osteonectin was evident in all 4 groups. Intense osteopontin and osteonectin were observed near newly formed bone and cells within it. An even immunoreaction was seen throughout the defect. Together with histological observation suggests bone formation and mineralization proceed in the central direction from the margin of the defect (Fig. 8). DISCUSSION The current trend in reconstruction of the facial skeleton is directed toward the enhancement of bone regeneration to avoid the need for harvesting autogenous bone grafts. One of these innovative approaches is to use tissue-engineered growth of bone, by which cultivated bone cells in scaffolds are delivered to a skeletal defect to form bone at the site of implantation. For this purpose, the scaffold should be biocompatible and serve as a proper matrix for the cells to produce the new structural environment of extracellular matrix ex vivo.14 The tissue-specific differentiation of bone marrow– derived MSCs in a controlled manner (e.g., osteogenesis) is dependent on the matrix or local environment in which the hMSC resides.15 Selection of a matrix carrier involves consideration of the matrix’s role as a scaffold for physical support and host tissue integration, as well as its ability to support or synergize
the osteoinductive program of the implanted MSCs. A majority of the currently available matrix materials for bone grafting are osteoconductive materials that support bone formation by acting as a scaffold for angiogenesis, cell recruitment, and ultimately osteogenesis by host cells.16 For MSC-directed bone repair to be clinically successful, a scaffold must be identified and optimized to support cellular adherence, cell recruitment, osteoinduction, and osteoconduction. These results contribute to a growing literature that demonstrates that the MSCs are able to form bone after implantation on porous HA/TCP matrices.11,17 In this study, to enhance cell-loading efficacy, the scaffolds were coated with collagen I gel a day before they were loaded by MSCs; light and scanning electron microscopic study confirmed that the cells occupied the spaces of the scaffolds. The cells were not treated by osteogenic medium for the implantation in the animal models and used as undifferentiated cells. The osteogenic medium only demonstrated the in vitro capacity of osteogenic differentiation. It was thought that the occurrence of osteogenesis in the implantation site of mandibular bone defect would be as a result of combined action of bone microenvironment and MSCs loaded on scaffold materials. It is clearly beneficial for any tissueengineered device to allow the host cells to contribute to regeneration of the desired tissue. Comparison
OOOOE Volume 105, Number 5
Jafarian et al. e21
Fig. 5. Representative histomicrograph of the bone formation around the cuboidal osteoblast-like cells in the masseter muscle. Bone is stained pink, ceramic has a shadowy white appearance, loose connective tissue is stained light pink, cells are stained a dark pink (hematoxylin-eosin stain). A, magnification ⫻10. B, Magnification ⫻100. C, Magnification ⫻10). D, Bone formation within the NBM (magnification ⫻10). Arrows indicate the new bone formation in direct bone contact with the bone substitutes.
of cell-free and MSC-loaded matrices in this study demonstrated a clear difference in bone formation, confirming the osteogenic potency of MSC-loaded constructs. The MSC-loaded implants were well integrated with the surrounding host bone. In most cases, the cell-free implants showed slight evidence of osteoconduction from the host tissue, which may have been influenced by collagen type I as an inductive factor, but were loose in the defect site (particularly in the Bio-Oss group) and had not formed significant bonding with the surrounding bone. It is possible that one important effect of the MSC/matrix is the encouragement of the regional or local osteogenesis. Bone was formed by an intramembranous pathway exclusive of evidence of chondrogenesis at 6 weeks. Under the current conditions, MSC osteogenesis in orthopedic sites occurred by a direct conversion of mesenchymal cells into osteoblasts rather than by an endochondral sequence.16 The muscle specimen showed that cell-loaded Kasios biphasic
matrices have more bone formation percentage (BF ⫽ 29.11%) at the end of the study but there is no significant statistically difference between groups (P ⬎ .05). The ranges of data between the maximum amounts of the BF in each group were lower than 20%; the authors could not find a positive inductive factor for MSCs seeded on the natural and synthetic scaffold. This minimal amount of bone formation in a non-mineral environment may be due to the collagen I, which acts as a cell carrier and presented in the control groups. The fate of the distant ectopic bone formation with the use of the MSCs may decrease with the findings of this study. The authors suggested MSCs could not impose a great influence on the bone formation in a muscular environment. Immunohistochemical analysis revealed bone matrix protein expression through the defect in all groups with higher enhancement in the periphery. The bone defect analysis demonstrated Kasios loaded with MSCs showed greater bone formation (65.78%) after 6 weeks
e22
OOOOE May 2008
Jafarian et al.
Fig. 6. A, Cuboidal osteoblast-like cells were generally found to be actively laying down bone to varying degrees along the mineralized bone in Kasios group. B, Bone formation around the cube of Kasios in the center of the defect far away from the margin of the original defect. C, D, Osteoblastic activity was observed at the test site in all groups. Large marrow spaces with fat cells and some hematopoietic cells were found in all of the groups. Arrow shows the original bone around the defect and the newly formed bone through the center of the defect. (hematoxylin and eosin, original magnification, ⫻100).
Table I. Mean percentage of bone fill in the dog mandibular bone through-and-through defects
HA/TCP Bio-Oss HA/TCP ⫹ cell BioOss ⫹ cell
Mean of bone formation in bone defect ⫾ SD
Mean of bone formation in muscle specimen ⫾ SD
44.9% ⫾ 13.45% 36.84% ⫾ 8.73% 65.78% ⫾ 4.94% 50.31% ⫾ 6.97%
23.55% ⫾ 4.99% 24.16% ⫾ 4.22% 29.11% ⫾ 6.10% 28.18% ⫾ 5.20%
The mean percentage of bone fill in biphasic bone substitute is more than other groups. HA/TCP, hydroxyapatite/-tricalcium phosphate.
(Fig. 7). Bone formation between the MSC groups (Kasios and Bio-Oss) and between pure scaffold groups (Kasios and Bio-Oss) did not reveal significant differences (P ⬎ .05). Kasios plus MSC had a significant statistical difference with other groups (P ⬍ .05). In the present investigation, the resorption process of the demineralized bone mineral and
HA/TCP particles did not follow a specific pattern. Implantation of culture-expanded autologous MSCs offers the advantage of directly delivering the cellular machinery responsible for synthesizing new bone and circumventing the otherwise slow steps leading to natural or enhanced bone repair. By incorporating living cells with specifically designed matrices, the shortcomings of osteoinductive factors alone to affect permanent bone repair may be overcome. CONCLUSION In the present study, MSCs from canine marrow were isolated, culture expanded, and loaded into either of 2 commercially available osteoconductive bone substitutes in order to compare their bone regeneration potential in through-and-thorough bone defects in combination with MSCs. Our results indicated that HA/ TCP matrices loaded with MSCs provides better conditions for bone regeneration than the other, although the difference was not statistically significant.
OOOOE Volume 105, Number 5
Jafarian et al. e23
Fig. 7. Histomorphometeric data of the percentage of bone fill (BF). Percent in each group revealed the higher percentage of bone fill in the biphasic bone substitutes (Kasios). 1 ⫽ Biphasic, 2 ⫽ NBM, 3 ⫽ Biphasic ⫹ cells, 4 ⫽ NBM ⫹ cells.
Fig. 8. Positive immunoreactions around the edges of bone substitutes and central direction of the bone formation through the center of the defect (arrow).
Special thanks to Reza Vahid for his professional editing of the article. This study was supported by a grant-in-aid of the Iranian Center for Dental Research (ICDR), Shahid Beheshti University of Medical Sciences and Royan Institute.
REFERENCES 1. Block MS, Kent JN. Sinus augmentation for dental implants: the use of autogenous bone. J Oral Maxillofac Surg 1997;55: 1281-6. 2. Hollinger JO, Winn S, Bonadio J. Options for tissue engineering
e24
3.
4.
5.
6.
7.
8.
9.
10.
OOOOE May 2008
Jafarian et al.
to address challenges of the aging skeleton. Tissue Engineering 2000;6:341-50. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143-7. Dong J, Kojima H, Uemura T, Kikuchi M, Tateishi T, Tanaka J. In vivo evaluation of a novel porous hydroxyapatite to sustain osteogenesis of transplanted bone marrow-derived osteoblastic cells. J Biomed Mater Res 2001;57:208-16. Al Ruhaimi KA. Bone graft substitutes: a comperative qualitative histologic review of current osteoconductive grafting materials. Int J Oral Maxillofac Implants 2001;16:105-14. Meyer U, Joos U, Wiesmann HP. Biological and biophysical principles in extracorporal bone tissue engineering. Part I. Int J Oral Maxillofac Surg 2004;33:325-32. Niemeyer P, Krause U, Kasten P, Kreuz PC, Henle P, Südkamp NP, et al. Mesenchymal stem cell-based HLA-independent cell therapy for tissue engineering of bone and cartilage. Current Stem Cell Res Therapy 2006;1:21-7. Ito K, Yamada Y, Nagasaka T, Baba S, Ueda M. Osteogenic potential of injectable tissue-engineered bone: a comparison among autogenous bone, bone substitute (Bio-Oss), platelet-rich plasma, and tissue-engineered bone with respect to their mechanical properties and histological findings. J Biomed Mater Res 2005;73:63-72. Ogura N, Kawada M, Chang WJ, Zhang Q, Lee SY, Kondoh T, et al. Differentiation of the human mesenchymal stem cells derived from bone marrow and enhancement of cell attachment by fibronectin. J Oral Sci 2004;46:207-13. Jenssen SS, Broggini N, Weibrich G, Hjorting-Hanssen E, Shenk R, Buser D. Bone regeneration in standardized bone defects with autograft or bone substitute in combination with palatlete concentrate. A histologic and histomorphometeric study in the
11.
12.
13.
14.
15.
16.
17.
mandible of minipigs. Int J Oral Maxillofac Implants 2005;20: 703-12. Yoshikawa T, Ohgishi H, Tamai S. Immediate bone forming capability of prefabricated osteogenic hydroxyapatite. J Biomed Mater Res 1996;32:481-92. Piatteli M, Favero GA, Scranio A, Orsinic G, Piatteli A. Bone reaction to anorganic bovine bone (Bio-Oss) used in sinus augmentation procedure: a histologic long term report of 20 cases in humans. Int J Oral Maxillofac Implants 1999;14:835-40. Kadiyala S, Jaiswal N, Bruder SP. Culture-expanded, bone marrow-derived mesenchymal stem cells can regenerate a criticalsized segmental bone defect. Tissue Eng 1997;3(2):173-85. Wiedmann-Al-Ahmad M, Gutwald R, Gellrich NC, Hubner U, Schmelzeisen R. Search for ideal biomaterials to cultivate human osteoblast-like cells for reconstructive surgery. J Mater Sci Mater Med 2005;16:57-66. Cooper LF, Harris CT, Bruder SP, Kowalski R, Kadiyala S. Incipient analysis of mesenchymal stem-cell-derived osteogenesis. J Dent Res 2001;80:314-20. Nomura T, Katz L, Powers MP, Saito C. Evaluation of micromechanical elastic properties of potential bone-grafting materials. J Biomed Mater Res B Appl Biomater 2004;73:29-34. Schopper C, Ziya-Ghazvini F, Goriwoda W, Moser D, Wanschitz F, Spassova E, et al. HA/TCP compounding of a porous CaP biomaterial improves bone formation and scaffold degradation—a long-term histological study. J Biomed Mater Res B Appl Biomater 2005;74:458-67.
Reprint requests: Mohamadreza Baghaban Eslaminejad, PhD Royan Institute PO Box 19395-4644 Tehran, Iran [email protected]