Maxillary sinus floor elevation with a tissue-engineered bone composite of deciduous tooth stem cells and calcium phosphate cement in goats

Authors

  • Wei Zhao,

    1. Department of Stomatology, Sixth People's Hospital, Shanghai JiaoTong University, Shanghai, People's Republic of China
    Search for more papers by this author
    • These two authors contributed equally to this work

  • Jia-Yu Lu,

    1. Department of Stomatology, Sixth People's Hospital, Shanghai JiaoTong University, Shanghai, People's Republic of China
    Search for more papers by this author
    • These two authors contributed equally to this work

  • Yong-Ming Hao,

    1. Department of Stomatology, Sixth People's Hospital, Shanghai JiaoTong University, Shanghai, People's Republic of China
    Search for more papers by this author
  • Chun-Hua Cao,

    1. Department of Stomatology, Sixth People's Hospital, Shanghai JiaoTong University, Shanghai, People's Republic of China
    Search for more papers by this author
  • De-Rong Zou

    Corresponding author
    1. Department of Stomatology, Sixth People's Hospital, Shanghai JiaoTong University, Shanghai, People's Republic of China
    • Correspondence to: De-Rong Zou, Department of Stomatology, Shanghai Jiao Tong University Affiliated Sixth People's Hospital,600 Yishan Road,Shanghai 200233, People's Republic of China. E-mail: derongzou@gmail.com

    Search for more papers by this author

Abstract

The study aimed to assess the effect of maxillary sinus floor elevation with tissue-engineered bone constructed from deciduous tooth stem cells (DTSCs) and calcium phosphate cement (CPC). The stem cells from goat deciduous teeth (SGDs) were isolated and transfected by means of the adenovirus with an enhanced green fluorescent protein gene (AdEGFP). As many as 18 bilateral maxillary sinuses of nine goats were randomly allocated into three groups (n = 6/group): group A (SGDs–CPC compound), group B (CPC alone) and group C (autogenous bone obtained from an iliac crest). All the samples were evaluated by computed tomography (CT), histology and histomorphometric analysis. Furthermore, the fate of implanted SGDs was traced using an immunohistochemical staining method in the decalcified samples. SGDs might be differentiated into osteoblasts in an osteogenic medium. In the present study, three-dimensional CT analysis showed that the volume of newly formed bone in group A was greater than that in the other two groups. After a healing period of 3 months, sequential analyses of triad-colour fluorescence labelling, histology and histomorphology indicated that the SGDs–CPC compound primarily promoted bone formation and mineralization at 2 and 3 months after the operation. Moreover, the areas of new bone formation in elevated sinuses were 41.82 ± 6.24% in the SGDs–CPC group, which was significantly higher than the 30.11 ± 8.05% in the CPC-alone group or the 23.07 ± 10.21% in the autogenous bone group. Immunohistochemical staining revealed that GFP and OCN were both expressed in the new bone tissue for the samples with eGFP, which suggested that the implanted SGDs might have contributed to new bone formation on the elevated sinus floor. SGDs can promote new bone formation and maturation in the goat maxillary sinus, and the tissue-engineered bone composite of SGDs and CPC might be a potential substitute for existing maxillary sinus floor elevation methods. Copyright © 2014 John Wiley & Sons, Ltd.

1 Introduction

Patients who have lost the natural dentition in their posterior maxilla and do not have sufficient bone height because of the pneumatization of the maxillary sinus are often stripped of the choice to acquire an implant prosthesis. Fortunately, the procedure for elevating the sinus floor has been further developed to increase the height of the alveolar bone to support implants (Jensen et al., 1998). Various methods have been used for sinus floor elevation in clinics, such as the implantation of autogenous bone grafts, allogenous bone grafts or other substitute grafts (Klijn et al., 2010; Graziani et al., 2004; Whitesides et al., 2006). Although these grafts might lead to certain augmentations in the height of the alveolar bone, the transplants of autogenous bone grafts and allogenous bone grafts often lead to complications, such as donor site pain, bleeding, infections and ethical problems. In addition, the use of substitute materials has, on occasion, been unable to bring about the formation of sufficient bone volume. As a result, the possibility of restricting these grafts in further clinical applications was assessed (Engelke et al., 2003; Boyne et al., 2005; Ueda et al., 2001; Wiltfang et al., 2002). With the development of bioengineering techniques, we have learned that an ideal bone substitute, which is not only biocompatible but also has osteoconductive and/or osteoinductive properties, could be fabricated using tissue-engineering techniques (Salgado et al., 2004). Therefore, we believe that tissue-engineered bone may be a more suitable substitute for sinus floor elevation grafts.

Tissue engineering is defined as an interdisciplinary field of research that applies the principles of life sciences and engineering to the design, construction, modification and growth of living tissue that are used to restore, maintain and improve tissue function with biomaterials, cells and other factors (alone or in combination with each other) (Sittinger et al., 1996; Langer and Vacanti, 1993; Salgado et al., 2004). In the bone tissue-engineering field, although bone marrow stromal cells (bMSCs) are known to differentiate into osteogenic cells and have been introduced as seed cells (Maniatopoulos et al., 1988; Petrakova et al., 1963), the percentage of bMSCs present in bone marrow is very low (1 in every 100 000 cells) (Caplan, 1994). Furthermore, bMSCs have a lower proliferation rate than some mesenchymal stem cells, which forces researchers to explore more proliferative stem cells for bone tissue engineering. Previous studies have shown that dental tissue contains various stem cells, which include dental pulp stem cells (DPSCs), stem cells from human exfoliated deciduous teeth (SHED) and periodontal ligament stem cells (PDLSCs). SHED are known to have a higher proliferation capacity than bMSCs and DPSCs (Seo et al., 2004; Gronthos et al., 2000; Miura et al., 2003). In addition, Miura et al. (2003) found that SHED can induce new bone formation by composing a supporting osteoinductive network for murine osteogenic cells after they are implanted into immunocompromised mice subcutaneously with HA/TCP. Interestingly, Seo et al. (2008) confirmed that SHED are able to differentiate into osteoblast cells and repair critical-size defects by the formation of substantial bone after being transplanted with HA/TCP. Zheng et al. (2009) applied SHED/β-TCP into critical-sized defects in mini-pig mandibles and confirmed that SHED/β-TCP can lead to good bone regeneration 6 months after transplantation with HA/TCP (Zheng et al., 2009). Furthermore, SHED are derived from a readily accessible tissue source, i.e. deciduous teeth, which are expendable and routinely exfoliated during childhood with little or no morbidity to the patient. Therefore, we can see that SHED could act as suitable seed cells for bone tissue regeneration.

Thus far, there have been no studies investigating SHED, which could be compounded with a calcium phosphate cement (CPC) scaffold material to construct tissue-engineered bone. Moreover, there are no reports of the application of such a method to maxillary sinus floor elevation. Our previous studies have demonstrated that goats have maxillary sinuses similar to those of humans and are appropriate large-animal models for maxillary sinus elevation (Zou et al., 2012; Derong et al., 2010). In addition, the stem cells from goat deciduous teeth (SGDs) have similar osteogenic differentiation patterns in vitro and bone-like tissue formation mechanisms in vivo to those of bMSCs. Moreover, SGDs may grow on CPC very well (Zhao et al., 2011). In this study, we hypothesized that SGDs compounded with CPC could be differentiated into osteogenic cells to form new bone under the sinus floor membrane to sufficiently support dental implants. A goat maxillary sinus floor elevation model was constructed to test this hypothesis and to verify the effect of SGDs–CPC in tissue-engineering applications in vivo.

2 Materials and methods

2.1 Animals

Nine healthy female goats, 6–8 months old, each weighing 18–22 kg, were used in this study, with an experimental protocol formally approved by the Animal Care and Experiment Committee of the Sixth People's Hospital affiliated with Shanghai Jiao Tong University.

2.2 CPC scaffold

CPC granules (3 mm diameter) were purchased from Rebone Biomaterial Co. Ltd (Shanghai, China) and prepared according to the product specifications, and were sterilized by 60Co irradiation before use. The average pore diameter of the CPC scaffold was 300–500 µm, with open porosity of 70% (Zou et al., 2012).

2.3 Isolation and culture of primary SGDs

The deciduous pulp tissues derived from first and second deciduous premolars were harvested from nine female goats. The isolation and culture of primary SGDs followed an established protocol (Miura et al., 2003; Jiang et al., 2008). The primary cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco) with 20% fetal bovine serum (FBS; Gibco) in an incubator and followed for the first passage, when they reached 90% confluence. It took approximately 7 days for an additional passage to occur. SGDs acquired from the second passage were used for further studies, and the medium was replaced by osteogenic medium (DMEM, 10% FBS, 50 µg/ml ascorbic acid, 10 mm glycerophosphate and 100 nm dexamethasone; Sigma).

2.4 Immunohistochemical staining

To identify the mesenchymal origin of SGDs, cells were fixed in 4% paraformaldehyde for 30 min, incubated with 0.1% trypsin solution for 30 min for antigen retrieval, blocked with non-specific antibodies, and incubated with either anti-CD29 or anti-CD146 for 1 h, according to the manufacturer's protocol. Subsequently, the samples were incubated with rabbit secondary antibodies for 1 h and then counterstained with DAPI. The results were observed via fluorescence microscopy.

2.5 Transfection of eGFP

Conditional retroviral supernatants derived from the stable retrovirus-producing cell lines PG13/eGFP were used in this study. For transfection, 2 × 105 SGDs cultured in six-well plates were incubated with a mixture of viral supernatants and DMEM at equal volumes for 48 h, and then the transfected cells were selected with G418 (100 µg/ml, Sigma). The transfer efficiency was determined by calculating the percentage of eGFP-expressing cells among all the cells observed, as described previously. The SGDs were collected for further study when they reached 90% confluence.

2.6 Osteogenic differentiation of SGDs

Real-time polymerase chain reaction (real-time PCR) assays were performed to assess osteogenic gene expression (i.e. ALP, COL-I, OCN, OPN) in SGDs cultured in osteogenic medium in vitro. SGDs were lysed to extract total RNA, using the reagent Trizol (Invitrogen), on days 7 and 14. cDNA synthesis was performed, using a PrimeScript II 1st Strand cDNA Synthesis Kit (TaKaRa). Real-time PCR analyses were performed using a Bio-Rad real-time PCR system (Bio-Rad) with markers of ALP, COL-I, OCN and OPN. β-Actin was analysed as a housekeeping gene for normalization. The primer sequences used in this experiment were designed using the Primer Premier 5.0 program, and are shown in Table 1.

Table 1. Real-time PCR primer sequences used in this study
GenePrimerSize of PCR product (bp)
  1. F, forward primer; R, reverse primer.
β-ActinF: AGAAAATCTGGCACCACACC142
R: GGGGTGTTGAAGGTCTCAAA
ALPF: GATGGCCGAGGTGATAGTGT200
R: TCTTTGGGAAGCTCGTCACT
COL-IF: CCAGTCACCTGCGTACAGAA243
R: AGTGTCTCCTTTGGGTCCCT
OPNF: CAGAAATACCAGTCCAGCAA135
R: GAAAGACAGCACCTCAAA
OCNF: AAAGCCTGGTGATGCAGAGT165
R: GGGTTGAGCTCACACACCTC

On day 14, SGDs cultured in the osteogenic medium were measured for their alkaline phosphatase (ALP) activity, using the ALP assay as described by Wu et al (Wu et al., 2007). On day 21, the level of calcium deposition was detected with 0.1% alizarin red S staining solution. Briefly, the cells were washed twice with phosphate-buffered saline (PBS) and fixed with 4% polyoxymethylene for 15 min, and then stained with 0.1% alizarin red S solution at 37 °C for 30 min.

2.7 Preparation of the SGDs–CPC complex

After being cultured in the osteogenic medium for 7 days, SGDs were collected and resuspended in osteogenic medium without FBS and then pipetted onto CPC at a density of 2 × 107 cells/ml for maxillary sinus elevation in vivo. The extra cell-seeded CPCs were cultured for 7 days, fixed with 2.5% glutaraldehyde in 0.1 m sodium cacodylate buffer, pH 7.2, for 4 h at 4 °C, and then examined for cell adhesion on CPC, using scanning electron microscopy (SEM; Philips SEM XL-30).

2.8 Maxillary sinus floor elevation procedure

Our previous studies have established a suitable surgical-operative plan and a procedure for goat maxillary sinus floor elevation (Figure 1) (Zou et al., 2012). Briefly, each goat was fasted for 48 h before the operation. All surgical procedures were conducted under general anaesthesia with an intravenous administration of pentobarbital sodium (30 mg/kg). A 6 cm paramedian oblique sagittal skin incision was made, the bony facial wall was exposed, and then a bone window of 1.5 × 1.5 cm was created with a round bur under continued sterile saline solution irrigation. To avoid perforation of the antral membrane, bent dissectors (Frios SinusSet) were used to gently push the sinus membrane inward and upward from the maxillary sinus floor. The membrane was lifted from the sinus floor and the infraorbital canal to form a large space (1.5 × 1.0 × 1.5 cm). The space was randomly filled with the following three groups of grafts: group A, the SGDs–CPC compound (n = 6); group B, CPC alone (n = 6); and group C, autogenous bone obtained from an iliac crest (n = 6). In addition, the bony bulk split was restored on the bony window. Finally, the mucoperiosteal flap and skin were repositioned and sutured. Moreover, to avoid postoperative infection, the goats received 0.8 mU/day of penicillin for 3 days.

Figure 1.

Surgical procedure for goat maxillary sinus elevation: (A) after anaesthesia and cannulation; (B) after skin preparation and draping; (C) bone window created on the maxillary sinus lateral wall; (D) the bone wall was lifted; (E) the sinus floor membrane was lifted with bent dissectors; (F) the elevated maxillary sinus area; (G) materials were implanted into the elevated maxillary sinus area; (H) the process of suturing the operative incisions; (I) the implanted materials of CPC incorporated with SGDs

2.9 Quantitative computed tomography assessment of bone volume in the maxillary sinus

Maxillofacial computed tomography (CT) images were acquired at 1, 2 and 3 months after surgical operation, using spiral CT (GE Lightspeed VCT 64) with a 120 kV voltage tension, 350 mA tube current, 1.25 mm slice thickness and 1.25 mm slice length. Volumetric measurements of the bones were obtained from all selected CT scans using Simplant 13.0 (Materialise), and were implemented according to the method described by Johansson et al. (2001).

2.10 Polyfluorochrome sequential labelling

A polyfluorochrome sequential labelling method was carried out to assess the time course of new bone formation and mineralization. Four and 8 weeks after the operation, the goats were subcutaneously injected with 25 mg/kg tetracycline (TE; Sigma) and 30 mg/kg alizarin red (AL; Sigma), respectively. Two weeks before sacrifice, 20 mg/kg calcein (CA; Sigma) was administered subcutaneously.

2.11 Sample preparation and general observations

The goats were sacrificed at 3 months after the surgical operation was conducted, according to the intravital intracarotid perfusion method described by Fickl et al. (2008) and Wetzel et al. (1995). The maxillary sinus was dissected, and the overlying soft tissues were cleared away and immersed in 10% neutral-buffered formalin for 7 days. Subsequently, all the samples (except those with the transfected eGFP genes, dehydrated in ascending concentrations of ethyl alcohol from 75% to 100%) were embedded in polymethylmethacrylate, cut into 200 µm-thick sections with a microtome (Leica) and ground to a thickness of 50 µm, which were further used for Van Gieson's picric acid–fuchsin staining. The samples with transfected eGFP genes were decalcified, embedded in paraffin and sectioned into 4 µm-thick sections, which were further used in haematoxylin and eosin (H&E) and immunohistochemistry staining.

2.12 Histological and histomorphometric observations

Four sections from the serial sections collected from each sample were selected randomly to analyse the new bone in augmented sinus. Fluorescent labelling was observed using the undecalcified sections under a confocal laser-scanning microscope (Leica TCS Sp2 AOBS). The excitation/emission wavelengths for each of the fluorochromes were as follows: 405/580 nm (TE, yellow), 543/617 nm (AL, red), and 488/517 nm (CA, green) (Suzuki and Mathews, 1966; Garcia et al., 1992; Traviesa-Alvarez et al., 2007). To quantify the bone formation and mineralization in the augmented sinus, five areas 1000 µm length × 1000 µm width were determined at the top, mesial, centre, distal and bottom for each section, and fluorescence microscopy images were taken in these areas, as described by Xia et al. (2011). These included images of tetracycline, alizarin red and calcein. A merged image of the three fluorescent labels was obtained to reflect the mineralization of the augmented sinus floor, and another image was formed under transmission light microscopy (without a specific filter) and combined with the former merged image to exhibit the newly formed bone (Xia et al., 2011). The images were stored and evaluated histologically and histomorphometrically with an image analysis system (Image-Pro PlusTM 6.0). The parameters included a single-labelled surface (sLS), a double-labelled surface, a bone surface (BS), a mineralizing surface (MS/BS, %), mineral apposition rate (MAR, µm3/µm2/day) and bone formation rate (BFR/BS; µm3/µm2/day), according to the American Society for Bone and Mineral Research Histomorphometry Nomenclature Committee (Parfitt et al., 1987). In addition, the single fluorochrome-stained area (%; i.e. the number of pixels labelled with one fluorochrome, determined as a percentage of the total bone-regeneration surface in the image) was measured and compared as previously described by Weibrich et al. (2004).

The undecalcified sections were subsequently stained with Van Gieson's picric acid–fuchsin for histological observation, and the percentage of newly formed bone tissue for each group in the augmented space was calculated, using an image analysis system (Image-Pro PlusTM 6.0).

To explore the effect of SGDs on newly formed bone, and whether the inducted SGDs could sustain osteogenic differentiation in the augmented sinus area, decalcified sections were used to detect the expression of osteocalcin (OCN) and eGFP in the endochylema of bone cells by immunohistochemistry staining, according to an image analysis system (Image-Pro Plus TM 6.0).

2.13 Statistical analysis

All data are shown as mean ± standard deviation (SD). Statistical analyses were performed with SPSS 16.0 statistical analysis software, and all statistically significant differences (p < 0.05) were calculated by ANOVA and SNK post hoc or the Kruskal–Wallis non-parametric procedure, followed by the Mann–Whitney U-test for multiple comparisons, based on the normal distribution and equal variance assumption tests.

3 Results

3.1 Culture of SGDs and expression of the eGFP gene

Cell clones were formed 5–7 days after the initial seeding, and they reached confluence approximately 14 days later. SGDs at passage 2 presented a type of fibroblastic cell morphology (Figure 2A) and were positive for the mesenchymal surface markers CD29 and CD146 (Figure 2B, C). Two days after the retroviral transfection was performed, the transfer efficiency reached > 80%. Subsequently, transfected SGDs, which were selected with G418, continued to express eGFP after several passages (Figure 2D).

Figure 2.

Characterization of stem cells from goat deciduous teeth (SGDs). (A) Second passage SGDs displayed the typical fibroblastic phenotype. (B, C) SGDs expressed CD29 (B) or CD146 (C), detected by immunohistochemical staining with anti-CD29 or anti-CD146. (D) 48 h after transfection with AdEGFP, SGDs still emitted bright and intensive green fluorescence. (E) Alkaline phosphatase-positive expression of SGDs after 14 days of culture in osteogenic medium. (F) Calcium deposition-positive expression of SGDs after 21 days of culture in osteogenic medium. (G) Scanning electron microscopy revealed that SGDs attached and spread to the surface of the CPC

3.2 Osteogenic differentiation of SGDs in vitro

ALP staining (Figure 2E) and alizarin red staining (Figure 2F) of calcified deposits were strongly positive for SGDs cultured in the osteogenic medium. Moreover, the real-time PCR results revealed the that the expression of the osteogenic genes (i.e. ALP and OCN) was significantly upregulated on days 7 and 14, but the upregulated extent of the expression for OPN was lower than the above (Figure 3). The expression of COL-I mRNA on day 7 did not increase significantly. In addition, the expression of ALP mRNA was downregulated from day 7 to day 14, which was contradictory to the results for OCN mRNA. In summary, these results suggest that SGDs cultured in osteogenic medium are committed to the osteogenic differentiation pathway.

Figure 3.

Comparisons of gene expression of alkaline phosphatase (ALP), collagen type I (COL-I), osteopontin (OPN) and osteocalcin (OCN) for SGDs after 7 and 14 days of culture in osteogenic medium

3.3 Adhesion and spreading of SGDs on CPC

SEM analysis of CPC indicated that CPC scaffolds were porous in structure, with the average pores in the size range 300–500 µm. The interconnection pores were ~250 µm, which suggests that the CPC scaffolds are similar to human cancellous bones. At 7 days after the SGDs were seeded onto CPC, observations indicated that the cells had firmly adhered to, and were well spread onto, the scaffold surfaces, suggesting that the CPC scaffolds were suitable for further in vivo studies (Figure 2G).

3.4 Volume analysis using CT

The results of the three-dimensional (3D) CT analyses for newly formed bone volume are shown in Figure 4. The volumes of the bone tissues in group C for months 1 and 2 were significantly less than those in groups A and C. However, the volume in group A at 3 months after the operation was significantly more than that observed for the other groups. It is possible that the resorption ratio of autogenous bone was higher than that of tissue-engineered bone or a bone substitute alone, that the transplanted materials were constantly absorbed in vivo, and/or that the implanted seeding cells promoted bone formation.

Figure 4.

The total volume of the augmented maxillary sinus was reconstructed by 3D CT at 3 months after the operations (A1–2, B1–2, C1–2 represented by groups A, B and C, respectively)

3.5 Fluorochrome-labelling analysis

Bone histomorphometric indexes were evaluated initially with fluorochrome-labelling analyses, and were compared among the three groups (A, B and C). The percentage of the AL-stained area (red) in group A was 2.84 ± 0.62%, which was significantly greater than that in group B (2.18 ± 0.64%) and group C (2.01 ± 0.21%). The percentage of the CA-stained area (green) in group A was 2.94 ± 0.65%, which was also significantly greater than that in group B (2.32 ± 0.57%) and group C (2.01 ± 0.21%). However, the comparisons among these three groups for the percentage of the TE-stained area (yellow) were not found to have any statistical differences (Figure 5).

Figure 5.

Three-colour polychrome sequential fluorescent labelling with TE, AL and CA at different periods. The images of (A1, B1, C1), (A2, B2, C2) and (A3, B3, C3) were labelled at 1, 2 and 3 months after operation, respectively. (A) SGDs–CPC; (B) CPC; (C) autogenous bone (× 50). (A4, B4, C4) represent merged images of the three fluorochromes for the same group (×50). (A5, B5, C5) represent merged images of the three fluorochromes, together with plain confocal laser microscope images for the same group (×50). (D) Graph shows the area of each fluorescent labelling in the three groups; *significant differences (p < 0.05)

In addition, the values of MS/BS MAR and BFR/BS in group A were greater than their corresponding values in groups B or C at 8–12 weeks after the operation, and there were statistical differences (p < 0.05) between group A and groups B and C within these parameters. The statistical analysis of bone histomorphometric indexes suggested that the bone remodelling course prompted by the SGDs in the sinus area was incurred at 8–12 weeks after the operation (Figure 6).

Figure 6.

New bone formation and mineralization parameters, including mineralizing surface (MS/BS, %), bone formation rate (MAR, µm/day) and bone formation rate (µm3/µm2/day) were calculated for each group during an observation period of 1–2 and 2–3 months postoperation; *p < 0.05

3.6 Histological findings

The results of the undecalcified specimens stained with Van Gieson's picro–fuchsin showed that there were newly formed trabeculae inside the augmented sinus in all groups at 3 months after the operation. The areas of new bone formation were 41.82 ± 6.24%, 30.11 ± 8.05% and 23.07 ± 10.21% for groups A, B and C, respectively; and there were significant differences between group A and groups B and C (p < 0.05). Under light microscopy, a number of new bones with newly formed and thicker trabeculae were found in group A; the CPC scaffolds were almost degraded and absorbed. In group B, new trabeculae were also observed. Moreover, the scaffolds were gradually degraded into small granules, and new bone was formed between the small remaining graft particles or on the surfaces of these particles. In addition, only small pieces of the slender trabecular bones were observed in the augmented space with the autogenous bone (Figure 7).

Figure 7.

Photomicrograph of new bone formation in the augmented maxillary sinus for non-decalcified samples in SHED/CPC, CPC and the autologous bone group (groups A, B and C; A1–C1, ×1.25; A2–C2, ×4; A3–C3, ×20). (A) SHED/CPC group: a number of newly formed bones with thicker and more numerous trabeculae were observed, and a small number of CPC scaffolds were found. (B) CPC group: new trabeculae were observed, and small granules degraded from CPC were detected. (C) Autologous bone group: only a small number of slender trabeculae were observed under the elevated sinus membrane. The area of new bone formation in the remaining augmented region in group A (SHED/CPC) was significantly higher than that of group B (CPC) or group C (autologous bone group); no significant difference was observed between groups B and C. (D) Graph presents an analysis of new bone formation area in the three groups; *significant differences (p < 0.05)

Through the analysis of immunohistochemistry staining with OCN and GFP proteins for decalcified specimens, we observed that the cells lining the trabeculae or located inside the bone lacunae presented with a strong positive result for OCN proteins and a weak positive result for GFP proteins (Figure 8). This suggested that SGDs continued to differentiate into osteoblasts and play an important part in the process of bone remodelling and new bone formation in the augmented sinus space. It was also confirmed that eGFP was useful for tracing the effects and locations of transplanted cells in vivo.

Figure 8.

Immunohistochemical staining characterization of OCN and GFP proteins in decalcified sections. (A) Black arrow indicates an OCN-positive osteoblast (×40). (B) Black arrow indicates positive eGFP-SGDs (×40)

4 Discussion

A meta-analysis of cell-based approaches in maxillary sinus augmentation procedures demonstrated that, although the bMSC-based approach returned a larger amount of newly formed bone in the lifted sinus, no significant differences were seen between the bMSC-based approach and the control groups. These data suggest that the selected cells should have advantages, such as easy acquisition, proliferation and differentiation (Park, 2010). As it is known that bMSCs have some drawbacks, e.g. acquisition and proliferation (Muraglia et al., 2000; Caplan, 1994), the work on exploring new, appropriate seeding cells is still a very important in tissue engineering. Miura et al. (2003) initially identified stem cells from human exfoliated deciduous teeth (SHED) and indicated that SHED have a higher degree of proliferation and a higher rate of population doubling. They also showed that SHED compounded with HA in vivo to produce bone-like tissue after transplantation into nude mice. Seo et al. (2008) successfully repaired mice calvaria defects with a SHED–HA–TCP complex. Hence, we can see that SHED, which are easily accessible and highly proliferative stem cell sources, are appropriate seeding cells in tissue engineering.

It is worth noting the debate on whether SHED differentiation into osteoblasts and their direct participation in bone formation is ongoing. Miura et al. (2003) suggested that SHED do not differentiate in vivo into osteoblasts but, instead, produce new bone by inducing recipient murine cells to differentiate into osteoblasts. Furthermore, Zheng et al. (2009) implanted SHED/β-TCP into critical-sized defects in mini-pig mandibles for reconstruction and obtained good bone regeneration. Another study indicated that induced osteogenesis to repair bone defects in a cell–scaffold approach was derived from donor progenitor cells and osteogenesis recruitment by vascularization (Zhang et al., 2005). Accordingly, there is an urgent need to test the ability of SHED with respect to osteogenic differentiation and to explore the mechanisms of how SHED participate in new bone regeneration in vivo.

In the present study, we chose SGDs as seeding cells and combined them with CPC for implantation into lifted maxillary sinuses in goats. Immunohistochemistry staining results showed that SGDs were positive for CD29 and CD146 in this study. Therefore, we hypothesized that SGDs have a similar tissue origin to SHED or bMSCs, because they are all derived from neural crest cells. After being cultured in osteogenic media, including dexmethasone, ascorbic acid and β-glycerolphosphate, SGDs ex vivo secreted a mineralized extracellular matrix and intracellular ALP proteins, which suggested that SGDs have mineralizing properties. Yamada et al. (2006, 2010) compared gene expression profiles of SHED and bMSCs via cDNA microarray analysis and found that the two cell lines had similar gene profiles, with high expression of genes associated with the initiation of mineralization and bone homeostasis. During the osteoblast differentiation and matrix formation phases, ALP was an early marker and OCN was a late marker (Higuchi et al., 2009; Ohara et al., 2004). In real-time PCR analyses, the expression levels of the osteogenic genes (i.e. ALP, COL-I, OCN and OPN) were upregulated during the course of our observations. A noticeable, upregulated level of ALP mRNA on day 7 was higher than that on day 14, and the expression level of OCN mRNA maintained this increase for 14 days, which corresponded to the above studies. In addition, we labelled SGDs with the GFP gene and transplanted SGDs into the augmented sinus area. Green fluorescence expression was found inside the newly formed bone lacunae and cells lining the trabeculae, where OCN was also expressed. Thus, it appears that SGDs transplanted into the augmented sinus area may maintain their state of differentiation into osteoblasts and directly take part in bone remodelling, regardless of the osteogenic medium or transplant area.

Sequential polychrome fluorescent labelling, 3D CT and histological and histomorphometric analyses all confirmed that the effect of bone mineralization and regeneration in SGDs–CPC was superior to that in autogenous bone or CPC alone. Fluorescent labelling using calcium-binding fluorescent dyes can clearly demonstrate the rate of new bone mineralization, which may be used as an indicator for assessing bone formation and remodelling in vivo (Pautke et al., 2005). According to sequential fluorescent labelling at 1, 2 and 3 months post-operation, the rate of bone mineralization and deposition was accelerated at 2 months in SGDs–CPC and was significantly higher than autogenous bone or CPC alone; a constant, high rate for SGDs–CPC was subsequently seen. The percentages of the AL-stained or CA-stained areas in the SGDs–CPC group were 2.84 ± 0.62% and 2.94 ± 0.65%, respectively; all percentages were significantly greater than those of CPC alone or autologous bone. The results of the MS/BS, MAR and BFR/BS measurements were in accordance with the above results. These findings indicate that the complex of SGDs and CPC transplanted into the sinus notably formed bone-like tissue 2 months after the operation. Although the key timing of bone formation in the augmented sinus from this study was a little later than the time reported in related studies, this difference might be attributed to differences in the transplant environment, scaffolds or other experimental conditions. A microenvironment with available oxygen, nutrients and osteogenic factors could benefit the survival and function of transplanted osteoblast-like cells (Liu et al., 2008). Zawicki et al. (1981) found that the development of new vessels in the augmented area might be slow, as was indicated by a vascularization rate of 0.09–0.25 mm/day in a rabbit model. Liu et al. (2008) transplanted autogenous osteoblast-like cells into augmented mini-pig sinuses but failed to find the enhanced bone formation seen with the cell-free group, possibly resulting from the existence of few osteoinductive factors in the maxillary sinus. In the present study, we concluded that the slow vascularization rate and low number of osteoinductive factors in the augmented area inhibited bone remodelling remarkably in the first month after the operation. It also confirmed that bone regeneration was the result of a continuous interplay between growth factors, cytokines and osteogenic cells (Lieberman et al., 2002; Reddi, 1998).

With the exception of the above distinctive microenvironment, the maxillary sinus was a special environment in which consistent air pressure co-existed with breathing. In such a natural physical environment, scaffolds with sufficient compressive strength may withstand consistent air pressure to maximally maintain the height and volume of an augmented maxillary sinus. A clinical research study reported that autogenous bone grafts could absorb up to 47% within 6–7 months after maxillary sinus lifting (Johansson et al., 2001), and the rapid absorption of autologous bone could not maintain the elevated height and would not lead to many newly formed bones in sinuses. The results of the current study showed that the bone volume of augmented sinuses in all autogenous bone was less than that in CPC alone or SGDs–CPC throughout our observations. In addition, the areas of new bone formation indicated by Van Gieson's picric acid–fuchsin staining at 3 months post-operation were 41.82 ± 6.24%, 30.11 ± 8.05% and 23.07 ± 10.21% for SGDs–CPC, CPC-alone and autogenous bone, respectively. The above results demonstrate that tissue-engineered bone (SGDs–CPC) maximally maintained the height and volume of augmented sinuses compared with CPC alone or autogenous bone.

Based on our findings, we may conclude that SGDs can promote the formation and maturation of new bone, and that the tissue-engineered bone composite of SGDs and CPC might be potential substitutions for existing maxillary sinus floor elevation methods.

Conflict of interest

The authors have declared that there is no conflict of interest.

Acknowledgements

This study was supported by the Science and Technology Commission of Shanghai Municipality (Grant No.12JC1407300; No.12ZR1447200) Shanghai Education Committee of China (Grant No.12JC1407301); Shanghai Municipal Commission of Health and Family Planning(Grant No.20134Y192); Shanghai Jiao Tong University Affiliated Sixth People's Hospital (Grant No.1582).

Ancillary