lable at ScienceDirect

Biomaterials 30 (2009) 2489–2498

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Vertical alveolar ridge augmentation with b-tricalcium phosphateand autologous osteoblasts in canine mandible

Shaoyi Wang a, Zhiyuan Zhang a, Jun Zhao a, Xiuli Zhang b, Xiaojuan Sun c, Lunguo Xia a, Qing Chang d,Dongxia Ye a, Xinquan Jiang b,*

a Department of Oral and Maxillofacial Surgery, Ninth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Key Laboratory of Stomatology,Shanghai 200011, Chinab Oral Bioengineering Lab, Shanghai Research Institute of Stomatology, Ninth People’s Hospital Affiliated to Shanghai Jiao Tong University, School of Medicine,Shanghai Key Laboratory of Stomatology, Shanghai 200011, Chinac Department of Oral and Maxillofacial Surgery, Affiliated Hospital of Ningxia Medical University, Ningxia 750004, Chinad Wuxi Fourth People’s Hospital, Wuxi, Jiangsu 214000, China

a r t i c l e i n f o

Article history:Received 17 November 2008Accepted 26 December 2008Available online 14 January 2009

Keywords:Tissue engineeringAutologous osteoblastsb-Tricalcium phosphateAlveolar ridge augmentation

* Corresponding author. Tel.: þ86 21 6313 5412; faE-mail address: xinquanj@yahoo.com.cn (X. Jiang)

0142-9612/$ – see front matter � 2008 Elsevier Ltd.doi:10.1016/j.biomaterials.2008.12.067

a b s t r a c t

A tissue-engineered bone has become a viable alternative to autologous bone for bone augmentation inatrophy alveolar ridge. The aim of the present study was to evaluate porous b-tricalcium phosphate (b-TCP) combined with autologous osteoblasts to augment edentulous alveolar ridge in a canine model.Autologous osteoblasts were expanded and combined with b-TCP scaffold to fabricate a tissue-engi-neered bone. 12 bilateral alveolar ridge augmentation surgeries were carried out in 6 beagle dogs withthe following 3 groups: b-TCP/osteoblasts, b-TCP alone and autogenous iliac bone control (n¼ 4 pergroup). Sequential fluorescent labeling and radiographs were used to compare new bone formation andmineralization in each group. 24 weeks later, animals were sacrificed and non-decalcified and decalcifiedsections were evaluated histologically and histomorphometrically. Results indicated that the tissue-engineered bone dramatically enhanced new bone formation and mineralization, increase the new bonearea, and maintain the height and thickness of the augmented alveolar ridge when compared with b-TCPalone group. More importantly, the tissue-engineered bone achieved an elevated bone height andthickness comparable to that of autogenous iliac bone graft. This study demonstrated the potential ofporous b-TCP as a substrate for autogenous osteoblasts in bone tissue engineering for alveolar ridgeaugmentation.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Dealing with a resorbed edentulous mandible or maxillaremains a major challenge in modern dentistry [1,2]. A deficientalveolar ridge fails to provide sufficient support and retention fordentures. That will not only compromise the soft tissue support andthe lower anterior facial height, but also preclude dental implantsplacement [3,4], which may dramatically reduce the quality of lifefor patients. Alveolar augmentation of the deficient osseous ridgehave become an integral part of therapeutic procedures in pre-prosthetic, pre-implantology surgery [5].

Several techniques have been suggested to reconstruct orenlarge a deficient alveolar ridge, including autologous bone grafts

x: þ86 21 6313 6856..

All rights reserved.

[6–11], distraction osteogenesis [12], and guided bone regeneration(GBR) [13]. Autologous bone grafts remains the gold standard forregeneration of alveolar bone [14,15]; however, they requireextensive harvest of healthy tissue from a distant location. They arealso limited by the unpredictability of remodeling with reportsshowing the complication rate ranging from 33% to >70% using themandibular symphysis grafts [2,16,17]. Distraction osteogenesis,which was previously employed for lengthening of bones and nowapplied to increase alveolar crest, has a reported complication rateof w70%, including basal bone and transport segment fractures,fixation screw loss, non-union, premature consolidation, andlingual positioning of transport segment [18]. In addition, it istechnically demanding and often unacceptable for patients whocannot tolerate intraoral distraction devices, Finally, the guided-bone regeneration, using membranes to create a space for thedevelopment of new bone and subsequent implant placement, mayface an unexpected exposure of the barrier membrane, which may

S. Wang et al. / Biomaterials 30 (2009) 2489–24982490

cause secondary infections and consequently failure of boneregeneration [19].

Due to the disadvantage of these methods, other alternativetherapies need to be explored for alveolar bone augmentation.Recent technological advances have allowed researchers to exploreregeneration of bone by using novel tissue engineering techniques[20], where design, construction, modification and growth of livingtissue are the goal. Biomaterial and cells are among the mostimportant key elements for tissue engineering [21]. So far, there hasbeen a case report which used the principles of tissue engineeringto augment alveolar ridge. In Strietzel et al.’s report, a tissue-engineered bone obtained by autogenous periosteum cells andpolymer fleece was used for lateral alveolar augmentation [22].However, this study only provided preliminary data. Though itdemonstrated certain success, there was no sequential analysisconducted, and the mechanism of bone regeneration was largelyunknown. Thus comprehensive pre-clinical investigations arenecessary before extensive clinical applications with thetechniques.

In this study, we used a porous b-TCP block combined withautologous osteoblasts to explore its effect on vertical alveolar ridgeaugmentation in a canine model. Autogenous bone or b-TCP aloneserved as controls.

2. Materials and methods

2.1. Animals

A total of 6 adult beagle dogs in healthy condition, aged 18–24 months old withan average weight of 12.5 kg were used in this study. The experimental protocol wasapproved by the Animal Care and Experiment Committee of Ninth People’s Hospitalaffiliated to Shanghai Jiao Tong University, School of Medicine.

2.2. Cell culture

Under general anesthesia with 5% sodium pentobarbital (0.5 ml/kg),a 3 mm� 3 mm bony sheet biopsy was obtained from the lateral cortex of themandibular body in the apical region of the second molar area by an intraoral buccalapproach. The explants were placed immediately into a solution of phosphatebuffered saline (PBS) containing of 100 units/ml penicillin and 100 units/ml strep-tomycin. After washing thoroughly by PBS to remove blood components, theexplants were cut into small fragments and cultured in 100 mm dishes (Falcon,Franklin Lakes, NJ) with DMEM:F12 (1:1) culture medium containing 10% (v/v) FBS(Hyclone), 100 units/ml penicillin and 100 units/ml streptomycin. Explants wereincubated at 37 �C under 95% humidity with 5% CO2. Explants were subsequentlymaintained by replacing the medium every 3–4 days until cell density reachedconfluence. Then cells were detached with 0.25% trypsin/EDTA, subcultured ata density of 1�105 cells/cm2 in 100 mm dishes. Cells at passage 2–3 were used forthe following experiments.

2.3. Phenotypes characterization of cultured cells

The osteoblastic phenotypes of these expanded cells were confirmed byfollowing tests.

2.3.1. Reverse transcription-polymerase chain reaction (RT-PCR)The transcription of osteoblastic marker of Collagen I, Osteopontin (OPN) and

Osteocalcin (OCN) was detected by RT-PCR. Briefly, the total cell RNA was prepared

Table 1Primers for RT-PCR.

Gene Primer sequence Product Size (bp)

Collagen type 1 50GAGCGGAGAATACTGGATTGA30

50CGGGAGGTCTTGGTGGTT30498

Osteopontin 50CACTGACATTCCAGCAAC30

50CTTCCATACTCGCACTTT30188

Osteocalcin 50TCACAGACCCAGACAGAACCG30

50AGCCCAGAGTCCAGGTAGCG30207

b-actin 50CCTGTGGC ATCCACGA AACT30

50GAAGCCATTTGCGGTGGACGA30307

by using TRIzol Plus RNA purification kit (Invitrogen) according to manufacturer’sinstructions. 1.0 mg total RNA was used as template for the synthesis of cDNA withOligodT and AMV reverse transcriptase (TaKaRa, Japan). The following PCR ampli-fication reaction utilized Taq polymerase and specific primers. All primer sequenceswere designed using Primer Premier 5.0 Software, synthesized commercially(Shenneng Co. Ltd. Shanghai, China) and the specific primers sets are outlined inTable 1.

2.3.2. Osteocalcin immunohistochemistryTo further evaluate osteoblastic differentiation level of expanded cells in vitro,

immunostaining with OCN antibodies was carried out. The cells were seeded andgrown on cover glass for 4 days and sections were subsequently fixed with 4%paraformaldehyde. Endogenous peroxidase was blocked with H2O2 for 30 min atroom temperature. Non-specific binding was blocked with goat normal serum for30 min at room temperature. Then the sections were incubated with mousemonoclonal antibodies against OCN (Abcam, Cambridge, UK) overnight at 4 �C ina moist chamber at 1:200 concentration and washed three times with PBS (PH 7.3).Goat biotinylated anti-mouse secondary antibody (Boster Co. Ltd, China) was incu-bated for 30 min at room temperature and washed three times with PBS (pH 7.3).Streptavidin biotin complex (Boster Co. Ltd, China) was incubated with the cells for20 min at room temperature and the cells were washed five times with PBS (pH 7.3).Staining was performed by DAB substrate (Boster Co. Ltd., China). The sections werecounterstained with hematoxylin, dehydrated through a series of alcohols andxylene, and then mounted with mounting medium. Positive cells stained brown,negative cells only showed blue/green nuclei but no brown cytoplasm.

2.3.3. Alkaline phosphatase and Von Kossa stainingAfter culturing for another 14 days, alkaline phosphatase (ALP) activity of the

cells were stained by fixing the cells for 10 min at 4 �C and incubating the cells witha mixture of naphthol AS–MX phosphate and fast blue BB salt (ALP kit, Hongqiao,Shanghai, China) [23]. The Von Kossa staining was performed by fixing the cells in70% ethanol and staining with 5% silver nitrate, and 5% Na2SO3 [24].

2.4. Preparation of osteoblasts/b-TCP construct

b-TCP scaffolds (Shanghai Bio-Lu Biomaterials Co. Ltd., Shanghai, China) weremolded into cuboids (20 mm� 6 mm� 6 mm) and were sterilized by 60Co irradia-tion before use. The scaffolds had volume porosity of 70% with average porediameter of 450� 50 mm. For cell seeding, osteoblasts were detached from culturedishes, centrifuged to remove supernatant, and then resuspended in the culturemedia without FBS at a density of 2�107 cells/ml. Cells in suspension were slowlyadded to the b-TCP cuboids till a final saturation. After incubation for an additional4 h to allow cell attachment, the scaffolds were used as described in Section 2.5.

In a parallel experiment, 3 mm� 3 mm� 3 mm cuboids were prepared andseeded with osteoblasts at an identical cell density. Four and 24 h later, theconstructs were fixed in 2% Glutaric dialdehyde for 2 h, cut into two halves, and thensubjected for scanning electron microscopy examination (Philips SEM XL-30,Amsterdam, Netherlands).

2.5. Surgical procedure

All mandibles were prepared by extraction of bilateral premolar as well as molarteeth 8 weeks before the alveolar augmentation surgeries. Under general anesthesia,12 alveolar augmentation surgeries in 6 animals were made bilaterally andrandomly divided into 3 groups: Group A consisted of tissue-engineered osteo-blasts/b-TCP complex (n¼ 4), Group B consisted of b-TCP alone (n¼ 4), and Group Cconsisted of autogenous bone obtained from iliac bone (n¼ 4) as a positive control.For the autogenous bone grafts, an incision of 5 cm was made on top of the anterioriliac crest and a corticocancellus bone blocks of the same size(20 mm� 6 mm� 6 mm) were harvested from the anterior iliac crest. During theoperation, full thickness mucoperiosteal flaps were elevated to expose the under-lying alveolar crests bone. All soft tissue remnants were carefully removed from thealveolar ridge and different groups of grafts were put into place to augment the

Denaturation Annealing Elongation PCR cycles

95 �C for 30 s 53 �C for 30 s 72 �C for 30 s 30

95 �C for 20 s 53 �C for 30 s 72 �C for 30 s 30

95 �C for 20 s 53 �C for 30 s 72 �C for 30 s 30

95 �C for 20 s 53 �C for 30 s 72 �C for 30 s 30

S. Wang et al. / Biomaterials 30 (2009) 2489–2498 2491

alveolar ridge. Great care was taken to avoid any puncture between the preparedgraft site and the oral cavity. Then osteoplasty for flat mandibular alveolar ridge wasperformed with a round carbide bur, to make grafted block closely adapted toalveolar ridge and stabilized by means of metal ligature wire. Then, the wound wasclosed in layers. All animals were kept on a soft diet during the study.

2.6. Radiographic observation

X-ray images were taken under general anesthesia at 4, 12, 20 weeks post-operation to follow up the scaffold degradation as well as the new bone formationand mineralization. Radiographs were made with a dental X-ray machine (Trophy,France), from a distance of 7 cm (230 V, 8 mA) with an exposure time of 0.28 s.Maxillofacial CT images were also obtained by a multi-slice spiral CT (GE LightspeedUltra 16, Milwaukee, WI) at three days before sacrifice.

2.7. Sequential fluorescent labeling

A polychrome sequential fluorescent labeling method was carried out to labelthe mineralized tissue and assess the time course of new bone formation andmineralization. At 4, 12 and 20 weeks after the operation, the animals were intra-peritoneally administered with 25 mg/kg tetracycline hydrochloride (TE, Sigma,USA), 20 mg/kg calcein (CA, Sigma, USA), and 30 mg/kg alizarin red s (AL, Sigma,USA), respectively.

2.8. Height and thickness of alveolar ridge bone

The height of the remaining augmented alveolar ridge was calculated by theaverage value of the three selected parallel lines defined at mesiodistal directions asshown in Fig. 1A. The mean value of three measurements (h1–h3) was calculated foreach graft and they were further used to calculate mean values for each group. Thethickness was evaluated by the average value of three selected parallel lines definedat coronal section as shown in Fig. 1B. The mean value of three measurements (t1–t3) were calculated for each graft, then they were further used to calculate meanvalues for each group. The augmented height and thickness were recorded andcompared as described previously by Schliephake et al. [25].

2.9. Sample preparation and histological and histomorphometric observation

The dogs were sacrificed at 24 weeks after surgery. The augmented alveolarbone was bisected mesiodistal direction and cut into the buccal and lingual halves,then fixed in 10% buffered formalin (pH7.4). One half was decalcified, embedded inparaffin, sectioned into 4 mm thick sections, and stained with hematoxylin–eosin.The other half was dehydrated in ascending concentrations of alcohols from 75% to100%, and finally embedded in polymethymetacrylate (PMMA). The specimens werecut in 150 mm thick sections using a microtome (Leica, Hamburg, Germany), andwere subsequently ground and polished to a final thickness of about 40 mm [24].Undecalcified sections were observed for fluorescent labeling using a microscopeunder confocal laser scanning microscope (CLSM) (Leica TCS Sp2 AOBS, Germany).Then the sections were stained with Van Gieson’s picro fuchsin for histologicobservation. Excitation and Emission Wavelengths of chelating fluorochromes wereused 405/560–590 nm, 488/500–550 nm, 543/580–670 nm for tetracycline(yellow), calcein (green), alizarin red s (red), respectively [26–28].

To analyze mineralization in alveolar bone, the fluorochrome staining of the newbone was quantified at nine locations: mesial, middle, distal along a sagittal sectionplane for top, middle, bottom area respectively. The mean value of the ninemeasurements was calculated per animal and they were used to calculate averagevalues for each group, according to the methodology of Lemperle et al. [29]. Fivephotographs were taken from the same area for each of the nine positions: threefluorescence microscopy images including the fluorochromes tetracycline, calcein,

Fig. 1. Schematic illustration of augmented alveol

and alizarin red s, one merging image of three fluorescent labeling to analyzemineralization of elevated alveolar ridge, and one image using transmission lightmicroscopy without a specific filter combining the former merging image. Themicroscope images were stored digitally and then evaluated histomorphometricallyusing a picture-analysis system (Image-Pro Plus�, Media Cybernetic, Silver Springs,MD, USA). Using this system, the number of pixels labeled with each fluorochromein each image was determined as a percentage of the mineralization area. This wasdone separately for yellow (tetracycline), green (calcein), and red (alizarin red s). Thedata on tetracycline, calcein and alizarin red s staining represent the bone regen-eration and mineralization 4, 12 and 20 weeks post-operation.

The measurements on undecalcified specimens were performed usinga personal computer-based image analysis system (image-Pro Plus�, MediaCybernetic, Silver Springs, MD, USA). Three randomly selected sections from theserial sections collected from each sample were analyzed. The percentage of newlyformed bone area in all augmented alveolar ridge area was calculated at lowmagnification (20�).

2.10. Statistical analysis

Statistically significant differences (p< 0.05) among the various groups weremeasured using ANOVA and SNK post hoc. All statistical analysis was carried outusing an SAS 6.12 statistical software package (Cary, NC, USA).

3. Results

3.1. Cell culture and osteoblastic phenotype

Cells were found growing around bone fragments 5–9 days afterinitial incubation in tissue culture medium (Fig. 2A), and prolifer-ated quickly afterwards to reach confluence in 4–7 days (Fig. 2B).Cells were then passaged, and these expanded cells were found toexpress the transcripts for the extracellular matrix proteins of Col I,OPN and OCN (Fig. 2C) based on the RT-PCR results. The OCNimmunohistochemistry staining further confirmed OCN expres-sion, which is a late marker of osteoblastic phenotype (Fig. 2D, E). Inaddition, ALP-positive staining (Fig. 2F) and mineralized calciumnodules (Fig. 2G) 14 days after seeding were confirmed in culturedcells, demonstrating that the expanded cells maintained the oste-oblastic phenotype.

3.2. Adhesion and spreading of osteoblasts on scaffold

The b-TCP scaffold was fabricated into 20 mm� 6 mm� 6 mmblocks (Fig. 3A). Scanning electron microscope was used to deter-mine all pores interconnected with interpore diameter of150� 50 mm (Fig. 3B). 4 h after the osteoblasts were combined withthe material, cells attached to the surface of the scaffold in vitro(Fig. 3C). After 1 day, cells spreading on the implant surfaces wereobserved (Fig. 3D), indicating the suitability of the material forosteoblasts adhesion.

ar ridge height and thickness measurement.

Fig. 2. Cell culture and osteoblastic phenotype. Cells were found growing around bone fragments after initial incubation (A), and proliferated quickly afterwards to reacha confluence (B). RT-PCR revealed the expanded cells expressing the transcripts of extracellular matrix proteins Col I, OPN and OCN (C). Immunocytochemical staining demonstratesthe expression of OCN (D). No positive staining appears in negative control (E) (200�). Alkaline phosphatase-positive stain area (F) (50�) and mineralized calcium nodules shownby Von Kossa stain (G) (16�), incubated for 14 days in vitro.

S. Wang et al. / Biomaterials 30 (2009) 2489–24982492

3.3. General observations and radiographic analyses

After surgery, all dogs recovered well, with no sign of infectionat anytime. Slight post-surgical edema was observed at the recip-ient site in each animal, which disappeared at 2–3 days post-operation.

Radiographic evidence of scaffold resorption as well as newbone formation was highly variable among the three groups. InGroup A at 4 weeks post-operation, the implant was radioopaquedue to the property of b-TCP scaffold itself. However, the interfacebetween the implant and the host bone began to show radiopacityindicating new bone formation and remodeling at this area(Fig. 4A1). At 12 weeks post-operation, certain degree of degrada-tion was evident based on the appearance of radiotranslucent areainside the graft. The interface between the graft and the host boneremained radioopaque. Interestingly, triangle radioopaque areashowing newly formed bone could be detected at the two ends ofthe grafts (Fig. 4A2). At 20 weeks post-operation, radiopacity ofelevated alveolar began to increase and the remodeling wasobvious by bone texture, indicating newly formed trabecular bone,and at this time point augmented alveolar ridge maintained theoriginal height, and the interfaces between the grafts and hostbones were hard to distinguish (Fig. 4A3–A5).

The radiopacity in Group B implants (b-TCP alone) could bevisualized easily at 4 weeks after implantation (Fig. 4B1), but itdecreased dramatically overtime due to scaffold degradation.Distinct radiolucent areas at the interfaces were obvious at weeks 4and weeks 12 (Fig. 4B1, B2), but it was no longer clear at 20 weekspost-operation, suggesting that b-TCP alone could also integratewith the host bone given sufficient time (Fig. 4B3). Nevertheless,most of the b-TCP ceramic implants had been resorbed at 20 weekspost-operation with a lower radiopacity and minor augmentedalveolar ridge remained (Fig. 4B3–B5).

In Group C, at 4 weeks post-operation, radiolucent zone at theinterface between the implant and the host bone was not obvious.

The new bone remodeling was seen, and the implants kept theiroriginal shape (Fig. 4C1). At 12 weeks post-operation, the appear-ance of the implant became smoother and more radiopaque. Theinterface between the implant and the host bone could no longer beeasily discriminated on the radiographs (Fig. 4C2). At 20 weekspost-operation, radiopacity of newly formed bone and remodelingwere obvious, and augmented alveolar ridge was maintainedthroughout the observation. Triangle radioopaque area showingnewly formed bone could also be detected at the two ends of thegrafts (Fig. 4C3–C5).

3.4. Height and thickness of alveolar ridge bone

To evaluate the effects of augmented alveolar bone, the heightwas measured. As the graph shows in Fig. 4D, the augmentedheight decreased significantly at 24 weeks to only 3.3� 0.24 mmfor Group B (b-TCP control). On the contrary, tissue-engineeredalveolar augmentation height remained 5.50� 0.16 mm which wascomparable to the autogenous bone group at 5.61�0.16 mm(p> 0.05 between Group A and C, but p< 0.05 for Group A and C vs.B). Similarly, at 24 weeks post-operation, the bone thickness ofGroups A, B, and C were 5.33� 0.36 mm, 3.47� 0.41 mm, and4.90� 0.23 mm, respectively (Fig. 4E). The thickness was signifi-cantly (p< 0.05) higher in Groups C and A as compared to Group B,but there was no significant difference between Groups A and C,indicating the equivalent bone formation between the tissueengineered construct and the autologous bone implants.

3.5. Fluorochrome labeling histomorphometrical analysis

New bone formation and mineralization were determinedhistomorphometrically by TE, CA and AL fluorescent quantifica-tion, which represented the mineralization level at different timeperiods. At 4 weeks, the percentage of TE labeling (yellow) inGroup A (osteoblasts/b-TCP) was 1.88� 0.51%, which was more

Fig. 3. Gross view of b-TCP block and scanning electron microscopic evaluation of the scaffold microstructure and biocompatibility. b-TCP with a dimension of 20 mm long, 6 mmhigh and 6 mm thick (A). Three-dimensional porous structures of b-TCP (B) (50�). 4 h after the osteoblasts were combined with b-TCP, they could be seen attaching to the surface ofthe scaffold (C) (1500�) and 1 day after they can spread well on the surface of the scaffold (D) (1500�).

S. Wang et al. / Biomaterials 30 (2009) 2489–2498 2493

than Group B (b-TCP) of 0.73� 0.23%, but less than Group C(autologous bone) 3.91�1.07% (Fig. 5A1, B1, C1). There weresignificant differences among three groups (p< 0.05; Fig. 5D). At12 weeks, the percentage of CA labeling (green) was 1.17� 0.37%,0.51�0.19%, and 1.35� 0.54%, for Groups A, B and C respectively(Fig. 5A2, B2, C2). There were significant statistical differencesbetween Groups A and B (p< 0.05), and Groups B and C(p< 0.05), but no significant difference between Groups A and C(p> 0.05) (Fig. 5D). At 20 weeks, the percentage of AL labeling(red) was 1.32� 0.46%, 1.43� 0.34%, 1.1�0.33%, respectively(Fig. 5A3, B3, C3), with no significant differences among threegroups (Fig. 5D). Taken together, this data indicated that, for theb-TCP scaffold control, fluorescence obviously detected till 20weeks after operation, demonstrated that osteo-conductive b-TCPalone can be used to achieve only a delayed and less effectivemineralization in alveolar ridge augmentation. In Group A(osteoblasts/b-TCP), seeding osteoblasts with osteo-conductive b-TCP, however, could promote new bone formation and mineral-ization inside the elevated space at a much earlier stage, as morefluorescent area and intensity in the center of grafts 4, 12 weeksafter operation, than in Group B. This tissue-engineered constructremained inferior to autologous graft in Group C at 4 weeks butcaught up with at 12 weeks.

3.6. Histological findings

Under light microscopy, the undecalcified specimens in Group Ademonstrated a significant amount of new bone formation, even atthe center portion of the implant. The percentage of newly formedbone area among initially augmented area (62.9� 6.33%) wassignificantly higher than that of Group B (28.2� 3.50%) (p< 0.05),in which only very thin layer of newly formed bone was found incontinuity with the native basal bone. In Group C, bone trabeculaecould be seen with a percentage of new bone area at 55.2� 6.65%,with no significant difference between Groups A and C (Fig. 6).

Decalcified sections stained with HE showed that bone forma-tion in the center of augmented area was more frequently observedin Group A. The woven to lamellar bone transition could bedetected in the newly formed bone area. The osteoid can also beseen in the center of augmented alveolar ridge. A line of cuboidal-shaped active osteoblasts was seen in the pores as well as bloodvessels that support progressive bone formation. Blood vesselswere also found close to the surface of b-TCP, possibly contributingto its biodegradation. Multinuclear cells were hardly observed inGroups A and B. In Group B, mature bone was less frequent at thecenter pores of b-TCP scaffold while immature bone and fibroustissue are more frequently observed, cuboidal-shaped active

Fig. 4. Sequential radiographs and gross observation post-operation. Group A: the interface between the implant and the host bone began to show radiopacity at 4 weeks (A1).Certain degree of degradation was observed by radiotranslucent area inside the scaffold and the interface between the graft and the host bone was radioopaque at 12 weeks (A2).Radiopacity of elevated alveolar began to increase and the interface between the grafts and host bone was hard to distinguish at 20 weeks (A3). Group B: b-TCP ceramic could alsobe visualized easily as the radiopacity of the material at 4 weeks (B1). Distinct radiolucent area at the interfaces could still be obviously observed atweek 12 (B2). b-TCP had beendramatically resorbed at 20 weeks with a lower radiopacity and minor augmented alveolar ridge remained (B3). Group C: Radiolucent zone at the interface between the implantand the host bone was not obvious and new bone remodeling can be seen at 4 weeks (C1). At 12 weeks, the appearance of the implant became smoother and more radiopaque. Theinterface between the implant and the host bone could no longer be discriminated easily on the radiographs (C2). At 20 weeks, radiopacity of newly formed bone and remodelingwere obvious (C3). Coronal maxillofacial CT images demonstrate the thickness of augmented alveolar ridge at 24 weeks post-operation. The thickness of b-TCP alone (B4) is muchthinner than b-TCP/osteoblasts complex (A4), or autogenous iliac bone (C4). Gross view of augmented alveolar ridge at 24 weeks post-operation, b-TCP/osteoblasts complex, b-TCPalone, autogenous iliac bone (A5, B5, C5) (scale bar¼ 1 cm). Dash lines indicate the augmented alveolar ridge. The graphs show the different height (D) and thickness (E) ofaugmented alveolar ridge among three groups at 24 weeks after implant, and there is significant difference between Group B and Group A or Group C. (* indicates significantdifferences p< 0.05).

S. Wang et al. / Biomaterials 30 (2009) 2489–24982494

osteoblasts was seldom seen at the surface in the pores (Fig. 7). InGroup C, trabecular bone was dominant in augmented alveolarridge, and mature lamellar bone could be observed. There are alsofew multinucleated giant cells in alveolar bone indicating theremodeling.

4. Discussion

Using a standardized dog vertical alveolar ridge augmentationmodel [30], we demonstrated that tissue-engineered bone withautogenous osteoblasts and a biodegradable b-TCP scaffold ach-ieved an effective elevation of the alveolar ridge which wascomparable to autologous bone grafts.

Autogenous osteoblasts derived from periosteum or bone, hasbeen successfully used previously to regenerate bones in animalexperiments [31], or in clinic [22,32,33]. They are easy to harvest,and compared with mixed properties of bone marrow stromal cells,they are relatively pure, and have inherent bone formation capa-bility [34,35]. This reduces the necessity for cell exposure to oste-ogenic medium and may reduce the risk of differentiationuncertainty observed in bone marrow derived stem cells. We havebeen able to isolate the osteoblasts from a small sheet of corticalbone in mandible through an intraoral surgical procedure. We wereable to obtain a total number of 30�106 cells in 3 passages withdefinite osteoblastic phenotypes, which was sufficient forconstructs to fill a void volume of as large as 3 cm3. Such a surgicalbiopsy is minimally invasive, and convenient for oral surgeons.

The b-TCP has good biocompatibility and osteo-conductivecapacity [36–38]. Compared with other bone substitutes (e.g.collagen scaffolds), b-TCP is characterized by its precisely definedphysical and chemo-crystalline properties, high level of purity and

uniformity of chemical composition, so that its biological reactionscan be predicted reliably [39]. It can be fabricated into high porosityscaffolds with good interconnectivity, which will ensure intercel-lular communication among osteogenic cells rested in lacunae[40,41]. The macro-porosity of the material will facilitate cellsadhesion and growth, and facilitate bone ingrowth and especiallyvascularization [42]. It has been used in oral and maxillofacialsurgery such as sinus floor augmentation and repair of deficientalveolar bone at immediate implant [39,43,44]. However, itspotential as a tissue engineering scaffold is not well established foraugmentation of alveolar ridge.

In this study, we found that osteoblasts can spread well on thescaffold surfaces and proliferate based on microscopic evaluation invitro. There was no obvious foreign body reactions based on theexplant analysis, confirming the biocompatibility of this biomate-rial. In vivo, through polychrome sequential fluorescent labeling,new bone deposition and mineralization can be clearly seen insidescaffold throughout the observation period for this scaffold/cellgroup. Three-labeled fluorescence was intensively and extensivelyexpressed since 4 weeks after operation, and fluorescence CA, Alwas comparable to autologous bone at a later stage of weeks 12 and20. These findings suggested that b-TCP had served as a goodscaffold for osteoblasts to increase the bone area and mineraliza-tion which has led the tissue-engineered complex to maintain theheight and thickness of the augmented alveolar ridge throughoutthe experiment.

In the b-TCP alone group, however, the augmented alveolarridge decreased overtime and the intensively expressed fluores-cence was not obviously observed till 20 weeks after operation,indicating that b-TCP alone has served only as an osteoconductivityscaffold. However, once seeded with autogenous osteoblasts in

Fig. 5. New bone formation and mineralization was determined histomorphometrically by TE, CA and AL fluorescent quantification, which represented the mineralization level at4(A1, B1, C1), 12(A2, B2, C2), and 20(A3, B3, C3) weeks after operation. A, B and C represents confocal LASER microscope for each group. A4, B4, C4 represented merged images of thethree fluorochromes for the same group. A5, B5, C5 represent the merged images of the three fluorochromes together with plain confocal laser microscope image for the samegroup; (D) The graph shows the percentage of each fluorochrome area for different group (* indicates significant differences p< 0.05).

S. Wang et al. / Biomaterials 30 (2009) 2489–2498 2495

Group A, these cells could filtrate into the pores of the scaffold andpromote more bone formation and mineralization much earlier andmore extensively in the inner area. This demonstrated the impor-tant role of tissue engineering technique has played during theapplication of porous ceramics like b-TCP in bone regeneration.

Fig. 6. The photomicrograph of bone formation in augmented alveolar ridge from non-deca(A1), Group B (B1), Group C (C1) (original magnification 20�). (A2) (B2) (C2) (original magnformation of three groups (* indicates significant differences p< 0.05).

As for the fate and role of the implanted osteoblasts, we do nothave direct evidence for their in vivo fate in this study. In a parallelexperiment, however, in which green fluorescence protein (GFP)labeled osteoblasts were used in the identical canine alveolar ridgeaugmentation model, we have evidence showing that the newly

lcified samples. The whole images of the representative slices of three groups: Group Aification 100�) (B) The graph shows the histomorphometric analysis of the new bone

Fig. 7. Decalcified sections stained with HE showed that, in Group A, bone formation in the center of augmented area was more frequently observed (A) (100�). The osteoid (OS) canalso be seen in the center of augmented alveolar ridge (B, arrow shows OS) (200�). A line of cuboidal-shaped active osteoblasts (OB) was seen in the pores (B, arrow shows OB)(200�), blood vessel (BV) can also be seen in the fibrous connective tissue (FCT) (B, arrow shows BV) (200�). In Group B, less mature bone existed in the center pores of b-TCPscaffold while fibrous connective tissue are more frequently observed (C) (100�), cuboidal-shaped active osteoblasts was seldom seen at the surface in the pores (D) (200�).

S. Wang et al. / Biomaterials 30 (2009) 2489–24982496

formed bone inside the b-TCP scaffold could express GFP 8 weeksafter surgery (data not shown), suggested their origin and roles ofthe seeded osteoblasts to form new bone in vivo. Long term celllabeling by either retroviral system or Lentiviral system are stillbeing explored for further evaluation by the authors.

With regard to the degradation rate of porous b-TCP, variousreports have showed that it depends not only on its porosity andpore size [45,46], but also at implant sites [47–50]. For example,Yuan J et al. implanted porous b-TCP to repair canine mandibularbone segmental defects, and found that most of the material wasdegraded in load-bearing area 26 weeks post-operation. Lu J et al.has reported that 55% of b-TCP could be degraded after 24 weeks ofimplantation in a rabbit model of femoral condyle implantation[45]. In our experiment, from the dramatically decreased heightand thickness of augmented alveolar ridge based on X-ray imagesand general observations, we found b-TCP degraded rapidly, frag-menting into pieces at 12 weeks post-operation, and most of whichwas resorbed at 20 weeks. The degradation of b-TCP in vivo isbelieved to involve in two pathways: solution-mediated dissolutionand a cell-mediated resorption process [42,51]. In our study, bloodvessels close to the surface of the pores were detected but we failedto find multinucleated giant cells in the scaffolds. Solution-medi-ated dissolution seems the predominant cause of b-TCP degrada-tion [52].

Once seeded with cells, however, the degradation rate of b-TCPin Group A was found reduced remarkably to such a level that itmatched the growth of newly formed bone to maintain theaugmented alveolar ridge in the current study. This phenomenonthat cells seeding into b-TCP could remarkably slow down the

degradation of b-TCP was also observed previously by Kurashinaet al. in an ectopic intramuscular implantation model and others[42,53]. The reason may be explained in this way: early boneformation and mineralization deposited could cover the innersurface which might have limited its exposure to surrounding bodyfluid. This might have prevented the implanted grafts exposed tosolution, slowed down of material degradation. Another explana-tion might be associated with mechanical forces. Handschel et al.reported that b-TCP was hardly resorbed in a non-loading calvarialmodel, while it could be easily degraded in the load-bearing area asother researchers observed [47,54]. Since augmented alveolar ridgein oral cavity was considered as a mechanically loaded area, itwould facilitate the degradation of b-TCP scaffold [48]. However,once combined with osteoblasts, earlier new bone formationoccurred in the center pores would have enhanced augmentedalveolar strength to withstand the mechanical force, which in turnreduced the degradation rate of b-TCP.

5. Conclusions

In summary, b-TCP alone could not be successfully used for theaugmentation of alveolar ridge in dogs. However, the combinationof b-TCP block with osteoblasts could be an ideal bone graft forvertical alveolar ridge augmentation. Sequential fluorescentlabeling and X-ray images indicated that through a tissue engi-neering technique, earlier new bone formation and mineralizationoccurred in the center pores inside b-TCP which has led to themaintenance of the height and thickness of alveolar bone. This

S. Wang et al. / Biomaterials 30 (2009) 2489–2498 2497

tissue engineered bone complex has provided a basis for furthersupport of denture or dental implants.

Acknowledgements

The authors acknowledge Dr. Hasan Uludag (Department ofChemical Engineering, University of Alberta, Canada) for hisinsightful comments and polish for the manuscript. This work wassupported by National Natural Science Foundation of China30400502, 30772431; Science and Technology Commission ofShanghai Municipality 07DZ22007, 08410706400, 08JC1414400,08DZ2271100, S30206; Shanghai Rising-star Program 05QMX1426,08QH14017; Shanghai Education Committee Y0203, 07SG19.

Appendix

Figures with essential colour discrimination. The majority offigures in this article may be difficult to interpret in black andwhite. The full colour image can be found in the on-line version, atdoi:10.1016/j.biomaterials.2008.12.067.

References

[1] Davo R, Malevez C, Rojas J. Immediate function in the atrophic maxilla usingzygoma implants: a preliminary study. J Prosthet Dent 2007;97(6 Suppl.):S44–51.

[2] Sjostrom M, Sennerby L, Nilson H, Lundgren S. Reconstruction of the atrophicedentulous maxilla with free iliac crest grafts and implants: a 3-year report ofa prospective clinical study. Clin Implant Dent Relat Res 2007;9(1):46–59.

[3] Cawood JI, Howell RA. A classification of the edentulous jaws. Int J OralMaxillofac Surg 1988;17(4):232–6.

[4] Cawood JI, Howell RA. Reconstructive preprosthetic surgery. I. Anatomicalconsiderations. Int J Oral Maxillofac Surg 1991;20(2):75–82.

[5] McAllister BS, Haghighat K. Bone augmentation techniques. J Periodontol2007;78(3):377–96.

[6] Cordaro L, Amade DS, Cordaro M. Clinical results of alveolar ridge augmen-tation with mandibular block bone grafts in partially edentulous patients priorto implant placement. Clin Oral Implants Res 2002;13(1):103–11.

[7] Lundgren S, Nystrom E, Nilson H, Gunne J, Lindhagen O. Bone grafting to themaxillary sinuses, nasal floor and anterior maxilla in the atrophic edentulousmaxilla: a two-stage technique. Int J Oral Maxillofac Surg 1997;26(6):428–34.

[8] Donovan MG, Dickerson NC, Hanson LJ, Gustafson RB. Maxillary andmandibular reconstruction using calvarial bone grafts and Branemarkimplants: a preliminary report. J Oral Maxillofac Surg 1994;52(6):588–94.

[9] Kondell PA, Nordenram A, Moberg LE, Eliasson S, Nyberg B. Reconstruction ofthe resorbed edentulous maxilla using autogenous rib grafts and osseointe-grated implants. Clin Oral Implants Res 1996;7(3):286–90.

[10] Tolman DE. Reconstructive procedures with endosseous implants in graftedbone: a review of the literature. Int J Oral Maxillofac Implants 1995;10(3):275–94.

[11] Schliephake H, Neukam FW, Wichmann M. Survival analysis of endosseousimplants in bone grafts used for the treatment of severe alveolar ridge atrophy.J Oral Maxillofac Surg 1997;55(11):1227–33 [discussion 1233–4].

[12] Chiapasco M, Romeo E, Vogel G. Vertical distraction osteogenesis of edentu-lous ridges for improvement of oral implant positioning: a clinical report ofpreliminary results. Int J Oral Maxillofac Implants 2001;16(1):43–51.

[13] Buser D, Bragger U, Lang NP, Nyman S. Regeneration and enlargement of jawbone using guided tissue regeneration. Clin Oral Implants Res 1990;1(1):22–32.

[14] Wojtowicz A, Chaberek S, Urbanowska E, Ostrowski K. Comparison of effi-ciency of platelet rich plasma, hematopoieic stem cells and bone marrow inaugmentation of mandibular bone defects. N Y State Dent J 2007;73(2):41–5.

[15] Misch CM, Misch CE, Resnik RR, Ismail YH. Reconstruction of maxillary alve-olar defects with mandibular symphysis grafts for dental implants: a prelimi-nary procedural report. Int J Oral Maxillofac Implants 1992;7(3):360–6.

[16] Joshi A. An investigation of post-operative morbidity following chin graftsurgery. Br Dent J 2004;196(4):215–8 [discussion 211].

[17] Clavero J, Lundgren S. Ramus or chin grafts for maxillary sinus inlay and localonlay augmentation: comparison of donor site morbidity and complications.Clin Implant Dent Relat Res 2003;5(3):154–60.

[18] McAllister BS, Gaffaney TE. Distraction osteogenesis for vertical bone augmen-tation prior to oral implant reconstruction. Periodontol 2000 2003;33:54–66.

[19] Alliot B, Piotrowski B, Marin P, Zahedi S, Brunel G. Regeneration procedures inimmediate transmucosal implants: an animal study. Int J Oral MaxillofacImplants 1999;14(6):841–8.

[20] Hollinger JO, Winn S, Bonadio J. Options for tissue engineering to addresschallenges of the aging skeleton. Tissue Eng 2000;6(4):341–50.

[21] Langer R, Vacanti JP. Tissue engineering. Science 1993;260(5110):920–6.

[22] Strietzel FP. Tissue-engineered bone for lateral alveolar ridge augmentation:a case report. Int J Oral Maxillofac Implants 2006;21(1):131–5.

[23] Jiang X, Gittens SA, Chang Q, Zhang X, Chen C, Zhang Z. The use of tissue-engineered bone with human bone morphogenetic protein-4-modified bone-marrow stromal cells in repairing mandibular defects in rabbits. Int J OralMaxillofac Surg 2006;35(12):1133–9.

[24] Sun XJ, Zhang ZY, Wang SY, Gittens SA, Jiang XQ, Chou LL. Maxillary sinus floorelevation using a tissue-engineered bone complex with osteobone and bMSCsin rabbits. Clin Oral Implants Res 2008;19(8):804–13.

[25] Schliephake H, Weich HA, Dullin C, Gruber R, Frahse S. Mandibular bone repairby implantation of rhBMP-2 in a slow release carrier of polylactic acid: anexperimental study in rats. Biomaterials 2008;29(1):103–10.

[26] Traviesa-Alvarez JM, Costa-Fernandez JM, Pereiro R, Sanz-Medel A. Directscreening of tetracyclines in water and bovine milk using room temperaturephosphorescence detection. Anal Chim Acta 2007;589(1):51–8.

[27] Suzuki HK, Mathews A. Two-color fluorescent labeling of mineralizing tissueswith tetracycline and 2,4-bis [N,N0-di-(carbomethyl)aminomethyl] fluores-cein. Stain Technol 1966;41(1):57–60.

[28] Garcia Campana AM, Ales Barrero F, Roman Ceba M. Spectrofluorimetricdetermination of boron in soils, plants and natural waters with alizarin red s.Analyst 1992;117(7):1189–91.

[29] Lemperle SM, Calhoun CJ, Curran RW, Holmes RE. Bony healing of largecranial and mandibular defects protected from soft-tissue interposition: Acomparative study of spontaneous bone regeneration, osteoconduction,and cancellous autografting in dogs. Plast Reconstr Surg 1998;101(3):660–72.

[30] Simion M, Rocchietta I, Kim D, Nevins M, Fiorellini J. Vertical ridge augmen-tation by means of deproteinized bovine bone block and recombinant humanplatelet-derived growth factor-BB: a histologic study in a dog model. Int JPeriodontics Restorative Dent 2006;26(5):415–23.

[31] Rai B, Teoh SH, Ho KH, Hutmacher DW, Cao T, Chen F, et al. The effect ofrhBMP-2 on canine osteoblasts seeded onto 3D bioactive polycaprolactonescaffolds. Biomaterials 2004;25(24):5499–506.

[32] Schimming R, Schmelzeisen R. Tissue-engineered bone for maxillary sinusaugmentation. J Oral Maxillofac Surg 2004;62(6):724–9.

[33] Schmelzeisen R, Schimming R, Sittinger M. Making bone: implant insertioninto tissue-engineered bone for maxillary sinus floor augmentation-a prelim-inary report. J Craniomaxillofac Surg 2003;31(1):34–9.

[34] Lee MK, Lim SW, Yang H, Sung SH, Lee HS, Park MJ, et al. Osteoblast differ-entiation stimulating activity of biflavonoids from Cephalotaxus koreana.Bioorg Med Chem Lett 2006;16(11):2850–4.

[35] Hu Y, Chan E, Wang SX, Li B. Activation of p38 mitogen-activated proteinkinase is required for osteoblast differentiation. Endocrinology 2003;144(5):2068–74.

[36] Koepp HE, Schorlemmer S, Kessler S, Brenner RE, Claes L, Gunther KP, et al.Biocompatibility and osseointegration of beta-TCP: histomorphological andbiomechanical studies in a weight-bearing sheep model. J Biomed Mater Res BAppl Biomater 2004;70(2):209–17.

[37] Xin R, Leng Y, Chen J, Zhang Q. A comparative study of calcium phosphateformation on bioceramics in vitro and in vivo. Biomaterials 2005;26(33):6477–86.

[38] Sun JS, Tsuang YH, Liao CJ, Liu HC, Hang YS, Lin FH. The effects of calciumphosphate particles on the growth of osteoblasts. J Biomed Mater Res1997;37(3):324–34.

[39] Horch HH, Sader R, Pautke C, Neff A, Deppe H, Kolk A. Synthetic, pure-phasebeta-tricalcium phosphate ceramic granules (Cerasorb) for bone regenerationin the reconstructive surgery of the jaws. Int J Oral Maxillofac Surg2006;35(8):708–13.

[40] Ogose A, Hotta T, Hatano H, Kawashima H, Tokunaga K, Endo N, et al. Histo-logical examination of beta-tricalcium phosphate graft in human femur.J Biomed Mater Res 2002;63(5):601–4.

[41] Boo JS, Yamada Y, Okazaki Y, Hibino Y, Okada K, Hata K, et al. Tissue-engi-neered bone using mesenchymal stem cells and a biodegradable scaffold.J Craniofac Surg 2002;13(2):231–9 [discussion 240–3].

[42] Dong J, Uemura T, Shirasaki Y, Tateishi T. Promotion of bone formation usinghighly pure porous beta-TCP combined with bone marrow-derived osteo-progenitor cells. Biomaterials 2002;23(23):4493–502.

[43] Zerbo IR, Bronckers AL, de Lange GL, van Beek GJ, Burger EH. Histology ofhuman alveolar bone regeneration with a porous tricalcium phosphate:a report of two cases. Clin Oral Implants Res 2001;12(4):379–84.

[44] Ormianer Z, Palti A, Shifman A. Survival of immediately loaded dentalimplants in deficient alveolar bone sites augmented with beta-tricalciumphosphate. Implant Dent 2006;15(4):395–403.

[45] Lu J, Descamps M, Dejou J, Koubi G, Hardouin P, Lemaitre J, et al. Thebiodegradation mechanism of calcium phosphate biomaterials in bone.J Biomed Mater Res 2002;63(4):408–12.

[46] Yokozeki H, Hayashi T, Nakagawa T, Kurosawa H, Shibuya K, Ioku K. Influenceof surface microstructure on the reaction of the active ceramics in vivo.J Mater Sci Mater Med 1998;9(7):381–4.

[47] Handschel J, Wiesmann HP, Stratmann U, Kleinheinz J, Meyer U, Joos U. TCP ishardly resorbed and not osteoconductive in a non-loading calvarial model.Biomaterials 2002;23(7):1689–95.

[48] Yuan J, Cui L, Zhang WJ, Liu W, Cao Y. Repair of canine mandibular bonedefects with bone marrow stromal cells and porous beta-tricalcium phos-phate. Biomaterials 2007;28(6):1005–13.

S. Wang et al. / Biomaterials 30 (2009) 2489–24982498

[49] Merten HA, Wiltfang J, Grohmann U, Hoenig JF. Intraindividual comparativeanimal study of alpha- and beta-tricalcium phosphate degradation inconjunction with simultaneous insertion of dental implants. J Craniofac Surg2001;12(1):59–68.

[50] Nicholas RW, Lange TA. Granular tricalcium phosphate grafting of cavitarylesions in human bone. Clin Orthop Relat Res 1994;306:197–203.

[51] Kurashina K, Kurita H, Kotani A, Takeuchi H, Hirano M. In vivo study of a calciumphosphate cement consisting of alpha-tricalcium phosphate/dicalcium phosphatedibasic/tetracalcium phosphate monoxide. Biomaterials 1997;18(2):147–51.

[52] Zerbo IR, Bronckers AL, de Lange G, Burger EH. Localisation of osteogenic andosteoclastic cells in porous beta-tricalcium phosphate particles used forhuman maxillary sinus floor elevation. Biomaterials 2005;26(12):1445–51.

[53] Kurashina K, Kurita H, Wu Q, Ohtsuka A, Kobayashi H. Ectopic osteogenesiswith biphasic ceramics of hydroxyapatite and tricalcium phosphate in rabbits.Biomaterials 2002;23(2):407–12.

[54] Liu G, Zhao L, Zhang W, Cui L, Liu W, Cao Y. Repair of goat tibial defects withbone marrow stromal cells and beta-tricalcium phosphate. J Mater Sci MaterMed 2008;19(6):2367–76.