Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Jul 5.
Published in final edited form as: Mol Ther. 2007 Jul 24;15(10):1872–1880. doi: 10.1038/sj.mt.6300270

A Study of the Role of Nell-1 Gene Modified Goat Bone Marrow Stromal Cells in Promoting New Bone Formation

Tara Aghaloo 1,2, Xinquan Jiang 1,3, Chia Soo 4, Zhiyuan Zhang 3, Xiuli Zhang 3, Jingzhou Hu 3, Hongya Pan 3, Tiffany Hsu 1, Benjamin Wu 5,6, Kang Ting 1,2,7,*, Xinli Zhang 1,7,*
PMCID: PMC2705762  NIHMSID: NIHMS113876  PMID: 17653100

Abstract

Nell-1 is a recently discovered secreted protein with the capacity to promote osteoblastic calvarial cell differentiation and mineralization and induce calvarial bone overgrowth and regeneration in various rodent models. However, the extent of Nell-1 osteoinductivity in large animal cells remains unknown. The objective of the study was to evaluate the feasibility of adenoviral encoding Nell-1 (AdNell-1) gene transfer into primary adult goat bone marrow stromal cells (BMSCs) in vitro and in vivo and to compare the osteoinductive effects with those produced by bone morphogenetic protein-2 (BMP-2), a well established osteoinductive molecule currently utilized for regional gene therapy. AdNell-1-transduced BMSCs expressed Nell-1 protein and underwent osteoblastic differentiation within 2 weeks in vitro, which is comparable to AdBMP-2. After intramuscular injection of nude mice, the AdNell-1- and AdBMP-2-transduced BMSCs revealed new bone formation, while untransduced or AdLacZ-transduced BMSCs showed mainly fibrotic tissue proliferation. At 4 weeks, BMP-2 induced significantly larger bone mass with a mature bone margin and central cavity filled with primarily fatty marrow tissue. Nell-1 samples had significantly less bone mass but were histologically similar to newly formed trabecular bone mixed with chondroid bone-like areas verified by type X collagen (ColX) immunohistochemistry. This distinct difference in histomorphology from the bone mass induced by BMP-2 suggests that there is a potential clinical role/advantage for Nell-1 in skeletal tissue engineering and regeneration.

INTRODUCTION

Clinical problems requiring bone regeneration are diverse and challenging. Causes for deficient bone include trauma, surgical resection and reconstruction, neoplasia, and degenerative disorders. These deformities usually require extensive bone grafting surgery. Although autogenous bone grafting is the ‘gold standard’ in this type of surgery, it is associated with many complications including donor site morbidity,1 and therefore alternative biotechnological approaches are under investigation.

Therapy with recombinant proteins such as bone morphogenetic proteins (BMPs) has shown potential as a clinically useful alternative to autogenous bone grafting. BMPs have been able to repair bony defects in various animal models.2,3 Recently, BMPs have obtained Food and Drug Administration approval for human lumbar spinal fusion as well as acute and non-union tibial fracture treatment.4,5 However, supraphysiological doses of 0.4-1.5 mg/ml BMP-2 recombinant protein are required for osteoinduction in non-human primates and also in humans, as shown in clinical trials.6 When such high doses of BMP-2 are used, potential systemic side effects are of concern.7 In addition, ectopic bone formation and cyst formation due to BMP’s pleiotropic effect are undesirable outcomes that have been documented.8,9

Ex vivo gene therapy has been investigated with a view to transducing target cells with osteoinductive genes such as BMP, in order to enhance bone healing and repair.10,11 Ex vivo gene therapy allows delivery of gene products to be localized and target-oriented, thereby minimizing systemic side effects and maximizing local therapeutic effects.12 Bone marrow stromal cells (BMSCs), also referred to as mesenchymal stem cells are most frequently utilized because they are multipotent and can be easily obtained and manipulated.13,14 Furthermore, BMSCs have shown success in regenerating bone and repairing cartilage defects,15,16 and are therefore under intense investigation for regeneration of bone to repair local defects, for combination with gene therapy for osteoinduction, and for use in tissue engineering.17,18 BMPs have also been used to drive the differentiation of BMSCs toward the osteoblast lineage.19,20 Specifically, ex vivo adenoviral gene transfer of BMP-2 into mouse mesenchymal cells forms bone when injected intramuscularly,21 and repairs local bony defects.10,11,22

Most recently, the transcription factor Runx2, the master gene and a marker of osteoblasts with high specificity, has been evaluated for its osteoinductive efficacy when transduced into murine BMSCs both in vitro and in vivo.23 The osteoblast specificity of Runx2 may restrict some of the unwanted side effects previously seen with the BMPs since it is a downstream target of BMP-2. However, the transduction efficiency of BMSCs with Runx2 is relatively low, and only transduced cells will exhibit a direct response to the gene therapy.24 Ideally, a secreted molecule with a relatively high specificity to preosteoblasts and osteoblasts and potent osteoinductive properties would be beneficial for ex vivo gene therapy in skeletal tissue engineering.

NELL-1 (Nel-like molecule-1; Nel (a protein strongly expressed in neural tissue encoding epidermal growth factor-like domain)), was previously isolated from human calvarial bones with prematurely fused/fusing coronal sutures.25 At the structural level, NELL-1 contains several highly conserved motifs including a secretory signal peptide, an NH2-terminal thrombospondin-1-like module, five chordin-like cysteine-rich domains and six epidermal growth factor-like domains. The protein can be cytoplasmic or secretory.26 Adenoviral encoding Nell-1 (AdNell-1) overexpression in calvarial cells strongly promoted their osteoblastic differentiation and mineralization,27 while Nell-1 overexpression in vivo increased premature bone formation in the calvarial sutures of transgenic animals. Coincidently, the major defects of an N-ethyl-N-nitrosourea mutant mouse on the Nell-1 locus were also related to skeletal tissue.28 Our previous work has demonstrated that ex vivo treatment with Nell-1 protein can induce secondary cartilage hypertrophy and bone formation in distracted rat intermaxillary sutures.29 Furthermore, our recent data have confirmed that Nell-1 is a downstream target of Runx2 and is directly regulated by Runx2, suggesting that Nell-1 may act more selectively on the cells of the osteogenic lineage.30 In addition, in a rat calvarial bone defect regeneration study, we have demonstrated that Nell-1 is a stimulant for bone formation with a potency comparable to BMP-2.31 Because it is a secretory protein, Nell-1 is also expected to be more beneficial than Runx2 for ex vivo gene therapy in skeletal tissue engineering.

The objective of this study was to evaluate Nell-1’s osteoinductive properties in comparison with BMP-2’s well-established osteoinductive properties, using an ex vivo regional gene therapy approach involving adenoviral gene transfer into higher mammal—(goat) BMSCs. When compared against AdLacZ control cells, it was found that AdNell-1-transduced BMSCs significantly raised alkaline phosphatase production and mineralization in vitro, to levels comparable to those produced by AdBMP-2 transduced BMSCs. Furthermore, most of the AdNell-1-transduced BMSCs demonstrated new bone formation 4 weeks after intramuscular injection, as compared to the formation of only fibrotic tissue and/or small amounts of early-stage cartilaginous tissue in most of the control samples. Therefore, Nell-1, a novel Runx2 downstream molecule, could be a new candidate for ex vivo gene therapy in regeneration of bone. The relatively high osteoinductivity and localized bone formation shown by Nell-1, and the differences in volume and structure of new bone formation produced by Nell-1 and BMP-2 suggest that Nell-1 could also be used to complement BMP-2 in promoting bone regeneration while controlling undesired excessive ectopic bone formation.

RESULTS

Gene transduction and the effects on cell proliferation

In order to establish the optimal multiplicity of infection (MOI) for high adenoviral gene transfer efficiency, a set of preliminary experiments was performed using various doses of adenovirus. An MOI of 50 plaque forming units (pfu)/cell produced optimal effects in transfer efficiency without excessive cell death in vitro. Three days after transduction with 50 pfu/cell AdLacZ, X-gal staining showed that over 70% BMSCs were stained blue (Figure 1a). Cellular morphology was unchanged (as compared with untransduced control cells) after transduction with AdLacZ, AdNell-1, or AdBMP-2 (Figure 1a-d). Western blot analysis was performed to confirm the expression of Nell-1 protein in AdNell-1-transduced BMSCs. The results revealed Nell-1 over-expression upon transduction with 50 pfu/cell AdNell-1, whereas Nell-1 remained undetectable in AdLacZ- and AdBMP-2-transduced cells (Figure 1e; data not shown). Three to five days after seeding, a growth curve demonstrated a significant decrease in cell proliferation in AdNell-1-transduced cells, which was not seen in AdLacZ transduced cells; P < 0.05. No significant difference in cellular proliferation was detected between AdBMP-2- and AdLacZ-transduced cells; P > 0.05 (Figure 1f).

Figure 1. Gene transduction and the effects on bone marrow stromal cell (BMSC) proliferation.

Figure 1

Open in a new tab

(a) A multiplicity of infection of 50 pfu/cell achieved high transfer efficiency above 70% 3 days after AdLacZ transduction of goat BMSCs. Positive areas with X-gal staining are in blue. (b-d) Cellular morphology after transduction with AdNell-1, AdBMP-2, or untransduced control cells. Original magnification ×200. (e) Western blot probed with antibodies against Nell-1 and β-actin for confirmation of Nell-1 protein expression 3 days after gene transduction. (f) Effects of AdNell-1, AdBMP-2, and AdLacZ infection on BMSC proliferation. A growth curve represents cellular proliferation after transduction with AdNell-1, AdBMP-2, or AdLacZ. Data points represent the mean ± SD. A significant decrease in cell number after transduction with AdNell-1 compared to AdLacZ at days 3, 4, and 5; *P < 0.05. No significant difference was detected between AdBMP-2 and AdLacZ groups; P > 0.05.

Osteoblastic differentiation of BMSCs in vitro after gene transduction

Multipotent BMSCs have the ability to differentiate into many cell types including osteoblasts, chondrocytes, myoblasts, fibroblasts, adipocytes, and neurons, when provided with relevant stimuli in appropriate differentiation media.13,14 In this study, we sought to evaluate the osteoinductive effect of Nell-1, compared with BMP-2 as a positive control, on goat BMSCs in osteoblastic differentiation media containing ascorbic acid, β-glycerophosphate, and dexamethasone.32,33 Twelve days after gene transfer, alkaline phosphatase staining was greater in BMSCs transduced with AdNell-1 and AdBMP-2 than in those with AdLacZ (Figure 2a). In addition, von Kossa staining 2 weeks after gene transfer revealed a significant increase in calcium nodules in AdNell-1- and AdBMP-2-transduced BMSCs as compared to the AdLacZ group (Figure 2b and c; P < 0.05). There was no significant difference between the AdNell-1 and AdBMP-2 groups (P > 0.05). These results suggest accelerated osteoblastic differentiation in AdNell-1- and AdBMP-2-transduced goat BMSCs.

Figure 2. In vitro analysis of osteoblastic differentiation after bone marrow stromal cell transduction with AdNell-1, AdBMP-2, and AdLacZ.

Figure 2

Open in a new tab

(a) alkaline phosphatase expression 12 days after gene transduction. (b) von Kossa assay comparing calcium nodules between AdNell-1-, AdBMP-2-, and AdLacZ-transduced cells 14 days after gene transfer. (c) Quantitative analysis of calcium nodules present in the three groups. A significant increase is seen in the AdNell-1 and AdBMP-2 groups when compared with the AdLacZ group; *P < 0.05, but no significant difference is seen between the AdNell-1 and AdBMP-2 groups; P > 0.05.

Analysis of intramuscular bone formation

We utilized plain radiographic, 3D microCT, and histologic analyses to evaluate the bone formation in vivo with AdNell-1-, AdBMP-2-, and AdLacZ-transduced goat BMSCs. In a pilot study, preliminary experiments with non-osteogenic media and media supplemented without dexamethasone were evaluated, resulting in undetectable radio-opaque areas 4 weeks after in vivo injection in all the groups except in the AdBMP-2-transduced primary goat BMSCs. In this current work, the same osteogenic pre-conditioned culture was utilized for 3 days for every group including the AdNell-1, AdBMP-2, AdLacZ and untransduced BMSCs to yield comparable results. Except in one BMP-2 sample, no radio-opaque areas were detected on plain radiography 2 weeks after in vivo injection. However, histologic examination of the 2-week AdNell-1 samples demonstrated a mixture of cartilaginous/osseous-like tissue along with condensation of osteochondrogenic cells. Areas of bone matrix were seen surrounding marrow cavities. This was also confirmed with Alcian blue staining of the cartilaginous matrix in samples from the muscle injection sites relating to AdNell-1 treated BMSCs (Figure 3a and d). AdBMP-2 samples demonstrated a larger amount of relatively mature bone formation along with fatty marrow and less cartilaginous tissue (Figure 3b and e). In contrast, only a cluster of fibroblast-like stromal cells were detected in AdLacZ-transduced BMSC samples, with the additional presence of a small area of cartilage-like tissue in one of two AdLacZ samples (Figures 3c and f).

Figure 3. Histologic analysis of 2-week tissues in vivo.

Figure 3

Open in a new tab

(a) AdNell-1-transduced bone marrow stromal cell (BMSC) injection sites showing cartilage (arrows) and osseous tissue (arrowheads) with the presence of chondroid matrix (CM) and bone marrow; hematoxylin and eosin (H&E) staining. (b) AdBMP-2 transduced BMSC injection sites showing osseous tissue intramuscularly (arrowheads) with cartilaginous tissue (arrows); H&E staining. (c) AdLacZ-transduced BMSC injection sites showing a small locus of cartilaginous (arrows) and fibroblastic (arrowheads) tissue; H&E staining. (d-f) Alcian blue staining on corresponding tissue sections of a-c to confirm the presence of cartilaginous tissue (blue) Original magnification for all figures: ×100.

Four weeks after in vivo injection, obvious areas of radioopacity were detected in AdNell-1-transduced BMSC injection sites (Figure 4a). Furthermore, microCT and histologic analyses demonstrated a mass of intramuscular bone formation with a high density and morphology similar to native bone (Figure 4c, e and g). In the AdBMP-2 group, initial plain radiography showed a much less radio-opaque shadow (Figure 4b). Under microCT imaging, the mass appeared to be well defined and significantly larger than AdNell-1 (Figure 4d). But when bisected on microCT, it appeared as a hollow cavity with only a shell of calcified tissue on the outer surface (Figure 4f). Further, histology showed a thin bony rim around fatty tissue and marrow spaces with some cyst-like cavities and bony trabeculae interspersed throughout the sample (Figure 4h). Overall, five of the six AdNell-1 transduced BMSC injection sites showed localized bone masses and revealed similar characteristics on microCT and histological analysis, and all the seven AdBMP-2 sites showed bone masses and more marrow cavities filled with fatty tissue on histology. In contrast, only one small opaque area was detectable radiographically out of six AdLacZ samples. Statistical analysis showed a significant increase in radio-opaque sites in the AdNell-1 and AdBMP-2 groups when compared with AdLacZ (analysis of variance (ANOVA); P < 0.05). Further quantitative analysis of bone volume measured from the microCT analysis revealed that the AdBMP-2 group had a significantly larger average volume of 28.10 ± 5.98 mm3, AdNell-1 were the second largest at 3.19 ± 1.59 mm3, while that of the AdLacZ group was almost negligible compared to the other two with only 2.31 × 10-4 ± 5.31 × 10-4 mm3.

Figure 4. Radiographic, microCT, and histologic evidence of bone formation after 4 weeks in vivo.

Figure 4

Open in a new tab

(a) Plain radiography shows a large, defined radio-opaque mass (white arrow) with density similar to iliac bone, representing bone formation in the AdNell-1-transduced bone marrow stromal cell (BMSC) injection site on the left side as compared to the AdLacZ-treated right side. (b) An even larger radio-opaque mass with less dense bone compared to iliac bone is seen on the AdBMP-2-transduced BMSC injection site (white arrow). (c) 3D microCT analysis better demonstrates the bone mass seen in the AdNell-1 treated site (white arrow) similar to native bone, that could not be detected in AdLacZ sites. (d) 3D microCT image of a large bony nodule in the AdBMP-2-treated site (white arrow). (e) When the microCT image is bisected, a radio-dense mass is seen in the AdNell-1 treated site, (f) compared to a hollow cavity with only an outer bony shell with interspersed small bone trabeculae in the AdBMP-2-treated site. (g) Hematoxylin and eosin (H&E) histology shows the typical bony morphology in the AdNell-1 specimen compared to (h) a fatty marrow cavity with an outer bony surface and few bone trabeculae in the AdBMP-2 specimen (arrows). (i, l, o) H&E histological analysis (i) and immunostaining for Nell-1 (l) and bone morphogenetic protein-2 (BMP-2) (o) of in vivo bone formation in AdNell-1-transduced BMSC injection sites. Mature bone formation (arrows) with Haversian systems and marrow cavities with some fat cells are seen. Areas of chondroid bone matrix are also seen (CM). Nell-1 immunohistochemistry demonstrating positive brown staining in osteocytes (arrows) and some bone marrow cells (arrowheads). BMP-2 immunohistochemistry demonstrates low level staining only in new bone forming areas (arrows) and bone marrow cells (arrowheads). (j, m, p) H&E histology (j) and immunostaining for Nell-1 (m) and BMP-2 (p) of AdBMP-2-transduced BMSC injection sites demonstrates more mature but thinner trabecular bone scattered within a large amount of bone marrow in a big cavity filled with mostly fatty tissue, without obvious evidence of cartilage or chondroid matrix. Nell-1 immunohistochemistry shows no obvious staining, while BMP-2 immunohistochemistry demonstrates a strong BMP-2 expression in both the bone (arrows) and surrounding fibroblastic tissue (arrowheads). (k) Most of the AdLacZ-transduced BMSC injection sites show only fibroblastic tissue (arrows); (n) Frozen sections stained with X-gal after 4 weeks in AdLacZ-transduced BMSC injection sites (arrows). Original magnification for figure g and h: ×40; i, j, and k: ×100; l, m, n, o, and p: ×200).

Histologic analysis showed mixtures of cartilage and bone tissue in the site of injection with AdNell-1-transduced BMSCs (Figure 4i). The mature cortical bone areas containing Haversian systems and bone marrow cavities with fat formation were general characteristics of bone tissue in the peripheral part of the new bone mass (Figure 4i). Some parts of the bone mass also displayed immature bone with a chondroid matrix morphology, which stained positive with Alcian blue and contained large osteoblastic cells resembling chondrocytes (Figures 4i, 5a and d). In the AdBMP-2 samples, areas of mature cortical bone formation were seen interspersed among fatty marrow tissue, with much less cartilaginous tissue as seen with Alcian blue (Figure 4j, 5b and e). However, in the AdLacZ-transduced BMSC samples, five out of the six samples showed only fibroblastic tissue (Figure 4k). The remaining sample contained cartilage, with a small amount of lamellar bone forming at the periphery (Figure 5c and f). In addition, three untransduced BMSC injection sites showed fibroblastic tissue and only one site showed a few radio-opaque spots and minimal bone formation (data not shown).

Figure 5. Immunohistochemical analysis of intramuscular bone and cartilage formed in AdNell-1-, AdBMP-2-, or AdLacZ-transduced bone marrow stromal cell (BMSC) injection sites after 4 weeks in vivo.

Figure 5

Open in a new tab

(a) The AdNell-1 group demonstrates significant bone matrix formation (BM) and bone marrow cavities (M) with some areas of immature bone containing large cells resembling chondrocytes (black arrows). (b) The AdBMP-2 group shows more mature bone matrix formation (BM) with marrow cavities (M) mainly filled with fatty tissue. (c) The AdLacZ group shows primarily cartilage (C) with some bone formation at the periphery (arrows and arrow heads) and marrow cavities (M). (d, e, f) Alcian blue better demonstrates the cartilage present in the corresponding tissue sections of figure a, b, and c. (g) Sox9 immunohistochemistry shows minimal staining in the AdNell-1 sites (arrows), (h) light staining in the AdBMP-2 sites, (i) but abundant positive brown staining localized in the nuclei in the AdLacZ sites (arrows). (j) Type X collagen (ColX) immunohistochemistry shows positive staining throughout the extracellular matrix and the hypertrophic chondrocytes (arrows) in the AdNell-1 sites. (k) In the AdBMP-2 sites, positive staining appears only in the extracellular matrix. (l) Isolated areas of positive staining for ColX were observed in the periphery of the cartilage where bone formation occurred in AdLacZ sites (arrows). Original magnification for all figures: ×200.

In order to determine the origin of the newly formed intramuscular bone in AdNell-1-, AdBMP-2, and AdLacZ-transduced BMSC injection sites, immunohistochemical analysis for Nell-1 and BMP-2-, and β-galactosidase staining for LacZ were performed. After 4 weeks in vivo, large numbers of osteocytes and osteoblasts in or along the newly formed bone stained positive for Nell-1 in the AdNell-1 group (Figure 4l), but not in the AdBMP-2 group (Figure 4m). Immunohistochemistry testing displayed very intense positive BMP-2 staining for samples injected with AdBMP-2-transduced BMSCs (Figure 4p), whereas in AdNell-1-transduced BMSCs endogenous BMP-2 staining was present, but weak (Figure 4o). In addition, we also verified by X-gal staining that AdLacZ-transduced BMSCs were present intramuscularly and maintained expression of LacZ 4 weeks after injection (Figure 4n). Taken together, these data suggest that AdNell-1 transduction greatly increases the bone-forming ability of goat BMSCs in this intramuscular nude mouse model as does BMP-2. The formation of a chondroid bone intermediate was observed in the AdNell-1 group through the whole course of this experiment. In contrast, control BMSCs, whether transduced with AdLacZ or untransduced, rarely formed bone in this model.

Immunohistochemical analysis of bone formation

In view of the fact that AdNell-1 transduction of BMSCs led to a mixture of both cartilage and bone formation, AdBMP-2 transduction of BMSCs led to more mature bone formation and large amounts of fatty marrow, and AdLacZ-transduced BMSCs led to fibrous tissue or cartilaginous formation at the endpoint of this experiment, we attempted to determine the nature of their cartilage induction and whether the bone formation was mediated through a typical endochondral ossification pathway in the AdNell-1 and AdBMP-2 groups. The expression patterns of Sox9, an early marker of chondrogenesis and of type X collagen (ColX), a later marker, were evaluated in the cartilaginous and bony tissue with immunohistochemistry. Sox9 expression was localized in the nuclei of the chondrocytes, and the percentage of positive cells was much higher in the cartilaginous tissue from AdLacZ when compared with that in AdNell-1- and AdBMP-2-transduced BMSCs (Figure 5g-i). In contrast, the expression of ColX, a marker of hypertrophic chondrocytes, was detected not only in hypertrophic chondrocytes in AdNell-1 and AdLacZ groups as expected, but also produced intense staining in the extracellular matrix of trabecular bones from AdNell-1-transduced BMSCs (Figure 5j and l). The AdBMP-2 group showed positive ColX staining only in the extracellular matrix, as no obvious hypertrophic chondrocytes were seen (Figure 5k). These results demonstrated the subtle differences in the nature of the cartilaginous tissues in the two groups in terms of the maturation stage of chondrocytes and components of the extracellular matrix. The cartilage from AdLacZ-transduced BMSCs was in an earlier stage of maturation than the one from AdNell-1-transduced BMSCs, as the majority of chondrocytes were Sox9 positive, an indicator of proliferative prehypertrophic chondrocytes. Lower levels of Sox9 and more ColX expression in the limited amount of cartilaginous tissue from AdNell-1-transduced BMSCs were also indicative of its advanced maturity, approaching ossification. This was even more evident in the staining with ColX throughout the extracellular matrix in the AdBMP-2 group where almost no cartilage was still present after 4 weeks. This may indicate that in the AdBMP-2 group ossification progressed much further than in the AdNell-1 group. This may also suggest that Nell-1 promotes new bone formation in goat BMSCs in vivo by enhancing not only its osteoblastic differentiation, but also the chondro-blastic differentiation and chondrocyte hypertrophy. In addition, AdNell-1 sites displayed both bone-like matrix and areas of chondroid matrix, suggesting the process of chondroid bone formation, in which the early chondroid matrix becomes an intermediary for new bone formation and disappears as remodeling occus and more mature bone is formed.34 However, this chondroid matrix was not seen in the AdBMP-2 group at the endpoint of the experiment.

DISCUSSION

The objective of this study was to investigate the feasibility of using Nell-1 for bone regeneration through regional ex vivo gene therapy with BMSCs. The data have confirmed the well established osteoinductive potential of BMP-2 gene therapy and have shown that the osteogenic activity of goat BMSCs was also significantly enhanced by adenoviral gene transfer of Nell-1, thereby indicating that Nell-1 is a osteoinductive molecule and a potential candidate for use in ex vivo gene therapy for regenerating new bone.

Few studies have utilized higher mammal BMSCs as models for ex vivo regional gene therapy to regenerate skeletal tissue,11,22 and it is known that the osteogenicity of BMSCs declines from rodents to higher mammals.35 In fact, goats and sheep are regarded as higher mammals, and have been used as feasibility models for testing BMPs for cervical and lumbar intervertebral spinal fusions.36-38 There is evidence that the concentration of cells receptive to growth factors is lower in higher mammals, and osteoinductive agents may demonstrate desirable effects in lower animals but not in higher animals.39 Therefore, bone tissue engineering with ex vivo gene therapy appears to be much more challenging with higher mammal BMSCs, and yet necessary to validate the potential clinical applications.40 Previous studies showed that BMSCs are less efficient for adenoviral ex vivo gene transfer than other cell types. AdRunx2 at an MOI of 250-500 pfu/cell transduced 30-40% BMSCs, and expressed Runx2 protein for only 6 days before gradually decreasing to undetectable levels by 15 days in vitro. Given this short in vitro expression, the in vivo expression is also not expected to be long.23,24 In the current study, goat BMSCs were successfully transduced at a relatively high efficiency of approximately 70% for AdLacZ, AdNell-1, and AdBMP-2 at an MOI of only 50 pfu/cell. In addition, the transduced BMSCs survived at least 4 weeks after in vivo injection, verified by either X-gal staining or specific Nell-1 or BMP-2 immunohistochemistry.

AdNell-1-transduced cells showed a significantly lower cell number at days 3, 4, and 5 after seeding compared to those of AdLacZ. In general, a decrease in cell number could be the result of either an increase in differentiation or profound cell death. However, no increase in cell death was detected morphologically at those early time points, making an increase in cell death an unlikely cause for the decreased cell number. Since the Nell-1 was previously shown to promote rodent calvarial cell differentiation toward osteoblasts,27 the osteogenic differentiation and mineralization of transduced goat BMSCs were assayed in vitro and compared to BMP-2, a molecule known to induce osteogenic differentiation. An increase in alkaline phosphatase staining as well as an increase in mineralized nodules were found in AdNell-1- and AdBMP-2-transduced cells when compared to AdLacZ, with no significant difference between AdNell-1 and AdBMP-2. These findings indicated that the goat BMSCs transduced with AdNell-1 had been directed toward and specifically enhanced the osteoblastic differentiation to levels comparable with AdBMP-2, while their proliferation was remarkably inhibited in vitro. On the other hand, the goat BMSCs transduced with AdLacZ or untreated maintained a relatively high proliferative ability, but revealed only a minimal capacity to differentiate into osteoblasts within this 2-week time frame, even when cultured in the same osteoblastic differentiation media. The data obtained from the in vitro experiments led us to explore Nell-1’s bone formation potential in vivo in comparison to BMP-2.

Because the potential for adult BMSCs to regenerate bone in vivo is highly correlated with the type, size, and shape of the carrier,41 a simple model without the use of a carrier or scaffold is usually chosen for an initial study. Direct intramuscular injection of adenoviral transduced BMSCs into nude mice is a widely accepted model for initial evaluation of osteoinductive properties of growth factors or stimulants.9,21 In the current study, we first directly injected AdNell-1-, AdBMP-2-, or AdLacZ-transduced BMSCs intramuscularly into nude mice to evaluate the direct osteogenic potential of Nell-1 in vivo within 4 weeks, using BMP-2 as a positive control for comparison, before pursuing further tissue engineering applications of Nell-1. The results demonstrated that AdNell-1-transduced BMSCs gained the ability to form new bone in vivo without any type of physical scaffold support. Importantly, this model was also validated with the results from a positive control of AdBMP-2 and a negative control of AdLacZ-transduced goat BMSCs. This is the first report to evaluate Nell-1’s osteoinductive properties in non-bone forming tissues using an ex vivo gene therapy technique. However, previous studies had shown its ability to stimulate secondary cartilage and bone formation in a rat palatal suture distraction model29 and rat calvarial defect regeneration model.31 In addition, the results of this study are also consistent with previous reports wherein AdLacZ-transduced or untransduced goat BMSCs formed only fibroblastic tissue at 3 weeks, with calcification seen at 6 weeks on histologic examination.9 Other similar studies have been performed with BMP ex vivo gene therapy techniques with varied numbers and concentrations of BMSCs obtained from different animal species, yielding varied results.9,21 In the current study, the optimized number of goat BMSCs per site, the cell culture conditions, adenoviral transfer protocol, and harvesting conditions all contributed to the high rate (five out of six cases) of new bone formation in a short period of time, comparable with BMP induced bone formation which took place in all seven animals.9 Nevertheless, the formation of a small amount of cartilage or minimal ossification with control goat BMSCs is believed to be at least partially caused by the existence of dexamethasone in the current culture system.32,33 Media supplemented with ascorbic acid, beta-glycerophosphate, and dexamethasone were utilized because BMSCs can be stimulated to differentiate to preosteoblasts in cell culture with these factors, thereby facilitating Nell-1’s action on those committed cells. Still, these in vitro manipulations with osteogenic media alone are not likely to be sustained after in vivo implantation.23 Some untransduced and AdLacZ-transduced goat BMSCs did reveal a minimal ability to form ectopic bone at the injection sites, but the goat BMSCs were significantly stimulated to form new bone in this animal model once they were transduced with AdNell-1 or AdBMP-2.

To our surprise, the gross and microscopic morphology of new bone formation in the AdNell-1 samples differed from that of AdBMP-2 samples as revealed by microCT and histology. Although the AdBMP-2 samples showed larger average volumes of bone mass by radiography and microCT, the bisected microCT image showed a hollow cavity with only a shell of calcified tissue on the outer surface. Histologic appearance of the BMP-2 samples was characterized by islands of mature lamellar bone formation among extensive amounts of fatty marrow tissue, consistent with other studies of BMP-2 ex vivo gene therapy21,42 and tissue engineering applications.43 In contrast, as shown by microCT imaging, the bone masses induced by Nell-1 were more localized and solid at the BMSC injection sites. The histologic analysis showed both areas of mature lamellar bone with Haversian systems and some bone marrow cavities adjacent to areas of both mature and immature bone trabeculae with a chondroid-like appearance. The mechanisms behind these differences of bone formation between BMP-2 and Nell-1 may be partially explained by their hierarchical positions in the regulatory pathway of osteoblastic differentiation. BMP-2 is a growth factor with multiple roles including an osteo-inductive capacity on a variety of cells including myoblasts and undifferentiated pluripotent mesenchymal stem cells,44,45 which may account for the larger volume of bone mass and cyst-like structure filled with fatty marrow tissue. To date, Nell-1 has been identified as promoting osteoblastic differentiation only on the cells committed to osteochondrogenic lineage25,27 and functioning downstream of master osteoblast differentiation transcriptional factor, Runx2/Cbfa-1.30 Therefore it is not unexpected that a lesser bone volume with a more dense and localized quality was induced by Nell-1 in comparison with BMP-2 in this specific model using pluripotent primary bone morrow stromal cells. In the clinical setting, BMP-2 and -7 are among the most potent osteoinductive factors and usually require a large dose for therapeutic applications.6 This has the potential to cause ectopic bone formation and cyst formation.8,9 Thus, future clinical applications with BMP ex vivo gene therapy require significant improvement. Our previous experiments and current study have clearly shown and confirmed that Nell-1 is a novel osteoinductive factor with a more restrictive and specific targeting of cells of the osteoblast lineage.25,35 Under certain circumstances, Nell-1 can induce bone formation on its own.29,31 However, it is postulated that Nell-1 may work synergistically in combination with BMP-2 for most applicable conditions demanding bone regeneration, given the complementary nature of their properties in promoting cell proliferation and osteoblastic differentiation. Our ongoing research with the C2C12 mouse myoblast cell line has shown that Nell-1 and BMP-2 do work synergistically to enhance C2C12 osteoblastic differentiation and new bone formation.45 Nell-1 likely does not signal through the same osteogenic cascade as BMP-2, and possibly acts as a molecule to transition preosteoblasts/osteoblasts out of a proliferative holding position and into a differentiated phase.31

The histologic analysis suggests that chondroid bone formation, wherein mesenchymal cells differentiate first into chondro-blasts and then hypertrophy to form a chondroid matrix, was involved in the intramuscular bone formation induced by AdNell-1 through the whole course of experiment. On the other hand, a more mature bone matrix along with a large amount of fatty marrow dominated the bone formation process through typical endochondral bone formation in the AdBMP-2 group.21,42 The histologic pattern of chondroid bone has been well described in the literature, wherein osteoinductive bone formation occurs rapidly over a period of several weeks and is an intermediate between cartilage and bone.34 Chondroid bone formation is a bone formation process similar to that occurring in chondrocytic lacunae during fracture healing through endochondral ossification of the callus.46 However, it differs from endochondral ossification in that there is a lack of vascular invasion and a mixed cellular morphology containing osteocytes and hypertrophic chondrocyte-like cells.34,46 This chondroid matrix then mineralizes and serves as a temporary scaffold for new bone formation and eventually disappears through bone remodeling.34

In order to further explore the potential of an intermediate chondroid bone later mineralizing to mature bone as the possible mechanism involved in Nell-1 induced bone formation, we compared the expressions of both Sox9 and ColX, the key markers of chondrocyte differentiation in bone and cartilaginous tissues, from AdNell-1-, AdBMP-2-, and AdLacZ-transduced BMSC injection sites. In the process of endochondral ossification, chondrocytes first enter a proliferative phase followed by differentiation, where Sox9 inhibits this transition.47 ColX is a marker of hypertrophic chondrocytes directly preceding cartilage calcification.48 Upon immunohistochemical analysis, the percentage of Sox9 positive cells was much less in AdNell-1- and AdBMP-2-transduced BMSC samples than in AdLacZ, which indicates that the maturity of cartilage tissue in the AdNell-1 and AdBMP-2 samples was more advanced than in AdLacZ. This was further verified with Alcian blue showing the cartilaginous matrix. The AdLacZ samples showed the most intense staining in the widest area of tissue, followed by AdNell-1 in the areas of chondroid matrix. The AdBMP-2 samples did not show any apparent cartilaginous staining, indicating its more advanced stage of osteogenic differentiation. On the other hand, ColX expression was much more pronounced in the extracellular matrix of the AdNell-1 and AdBMP-2 samples, in addition to the expected staining of hypertrophic chondrocytes in the AdNell-1 samples. This finding may be an indicator of intermediate chondroid bone formation occurring in the early stage of the endochondral bone-forming process because ColX is generally present in hypertrophic chondrocytes, and not abundant in mature bone matrix. We therefore assume that AdNell-1 may promote new bone formation continuously by inducing chondrocytic hypertrophy, and in turn contribute to unique endochondral ossification through the intermediate of chondroid bone. This is quite different from classic endochondral ossification as occurs in the growth plate, in that the ectopic bone via chondroid bone formation progresses rapidly over a 3-4 week period.34

In summary, the adult goat BMSCs can be efficiently transduced with AdNell-1 as well as with AdBMP-2 and AdLacZ using the current protocol. The BMSCs transduced with AdNell-1 were capable of fully differentiating into osteoblasts in vitro and forming new bone after intramuscular injection in nude mice within 4 weeks, comparable to the results produced using AdBMP-2. The bone masses induced by Nell-1 were more dense and localized at the BMSC injection sites. Nell-1 has the potential advantage of achieving the required precision and delicacy of bone regeneration. The current results further confirm our previous findings and suggest that Nell-1 is a novel osteoinductive protein with a potential clinical role in skeletal tissue engineering or local bone regeneration via ex vivo regional gene therapy. Unlike the well-established molecule BMP-2, which has been intensively investigated for decades, the osteoinductive mechanism of Nell-1 still remains largely unknown, and will be the main focus of our future studies.

MATERIALS AND METHODS

Culture of goat BMSCs

Adult male goats were obtained from the Ninth People’s Hospital Animal Center (Shanghai, China), and all procedures were approved by the Animal Research Committee of the Shanghai Ninth People’s Hospital. Approximately 5 cc bone marrow was obtained from the goat iliac crest by needle aspiration under general anesthesia. The cells were centrifuged and suspended in Dulbecco’s modified Eagles’s medium (Gibco BRL, Grand Island, NY) containing 10% fetal bovine serum (Hyclone, Logan, UT), 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mmol/l l-glutamine (Sigma, St. Louis, MO). Hematopoietic cells, which did not adhere to flasks, were discarded, and the cells which adhered represented a population of primary BMSCs. After the first passage, cells were cultured in Dulbecco’s modified Eagles’s medium supplemented with 50 μg/ml ascorbic acid, 10 mmol/l β-glycerophosphate, and 10-8 mol/l dexamethasone for 3 days. Experiments were performed with cells from the second passage.

Gene transduction of BMSCs

An adenoviral vector encoding rat Nell-1 was generated as described previously.27 BMSCs were cultured for 24 hours to reach 80% confluence and transduced with a MOI of 10, 20, and 50 pfu/cell AdNell-1 and AdLacZ, and 50 pfu/cell AdBMP-2. Cell morphology was evaluated microscopically (Leica DM 1RB, Germany), and gene transfer efficiency was determined by X-gal staining 3 days after transduction with AdLacZ by calculating the number of blue stained cells among all the cells observed.49 Fifty plaque forming units per cell of AdNell-1 and AdLacZ were chosen for in vitro and in vivo experiments based on transduction efficiency and level of Nell-1 protein expression.

Cell proliferation analysis

AdNell-1-, AdBMP-2-, and AdLacZ-transduced BMSC proliferation was determined by cell counting utilizing a hemocytometer. A parallel set of 20,000 cells/well were cultured in 24-well plates (in quadruplicate). Gene transfer was performed on the second day, and cell numbers were counted for the next 4 days. At each time point, the numbers of AdNell-1- and AdBMP-2-treated cells were compared with AdLacZ-treated BMSCs by a single factor ANOVA with Student Newman Keuls method.

Western blot analysis

In order to determine the expression of Nell-1 protein, whole cell extracts were prepared from transduced BMSCs at 72 hours after transduction. After washing with ice-cold phosphate-buffered saline, the cells were lysed using a protein extraction regent (Kangchen Bio-tech, Shanghai, China). Proteins were fractionated by electrophoresis on 6% polyacrylamide gels and transferred to polyvinylidene fluoride membranes (Amersham Biosciences, Piscataway, NJ). The membranes were exposed to anti-Nell-1 (1:850 dilution) and anti-β-actin antibodies (1:10,000 dilution; Sigma, St. Louis, MO). Blots were exposed to secondary goat anti-rabbit for Nell-1 and anti-mouse for β-actin immunoglobulin G antiserum conjugated to horse radish peroxidase, and developed with ECL plus chemiluminescence reagent (Amersham Biosciences, Piscataway, NJ).

Alkaline phosphatase staining

BMSCs transduced with AdNell-1, AdBMP-2, and AdLacZ were evaluated for alkaline phosphatase activity 12 days after transduction as per the manufacturer’s instructions (Jianchen, China), modifying Kaplow’s method.50 Briefly, cells were fixed for 10 minutes at 4 °C and incubated with a mixture of substrates A, B, and C. Areas stained brown were designated as positive.

von Kossa assay

BMSCs plated in triplicate in 6-well plates were fixed in 70% ethanol 2 weeks after gene transduction. Cells were stained with von Kossa silver and placed under ultraviolet light for 10 minutes. Cells were then treated with 5% NaS2O3 for 2 minutes, and washed with distilled water. Calcium nodules with a diameter greater than 1 mm were counted and analyzed. Comparison between AdNell-1, AdBMP-2, and AdLacZ was performed using a single factor ANOVA with the Student Newman Keuls method.

Animal experiments

BMSCs 3 days after gene transduction were used for in vivo experiments. Five- to six-week-old BALB/c nude mice were obtained (Charles River Laboratories, Wilmington, MA), and given access to food and water ad libitum. Five million AdNell-1- or AdLacZ-transduced BMSCs suspended in 100 μl phosphate-buffered saline were injected into the left and right thigh muscles respectively of eight mice under general anesthesia (number of injection sites: AdNell-1 = 8; AdLacZ = 8). Two mice were sacrificed after 2 weeks, and the remaining six were sacrificed after 4 weeks. AdBMP-2-transduced BMSCs were used as a positive control of osteoinductive bone formation, and were injected unilaterally into the left thigh muscle in nine nude mice. Two of these mice were sacrificed after 2 weeks, and seven after 4 weeks. Four additional mice served as controls, receiving injections with the same numbers of untransduced BMSCs bilaterally (Control sites = 8). They were sacrificed at weeks 2 and 4, two mice at each time point.

Radiography, microCT and histologic analysis of ectopic bone formation

Radiographic analysis was performed on all collected samples (Faxitron MX-20). The radio-opacity in the injection sites of the AdNell-1- and AdBMP-2-transduced BMSCs was compared with those of the AdLacZ sites after 4 weeks. ANOVA was utilized to determine a significant difference between the groups. The samples fixed in 10% formalin were then scanned using high resolution microCT with 9-20 μm resolution, utilizing technology from μCT40 (Scanco Medical, Basserdorf, Switzerland) as previously published.27 Visualization and reconstruction of the data were carried out using the μCT Ray T3.3 and μCT Evaluation Program V5.0 provided by Scanco Medical. Bone volume was compared among the groups using single factor ANOVA with the Student Newman Keuls method. For histologic analysis, the paraffin embedded decalcified samples were sectioned at 4 μm and stained with hematoxylin and eosin. Alcian blue staining was performed to visualize cartilage formation at weeks 2 and 4. X-gal staining was performed on cryosections to determine the presence of AdLacZ BMSCs in vivo.

Immunohistochemical analysis

Sections embedded in paraffin were deparaffinized and incubated with primary antibodies including anti-Nell-1 (1:850 dilution),27 anti-BMP-2 (1:200 dilution), anti-Sox9 (1:100 dilution) (Santa Cruz Biotechnology, Santa Cruz, CA), and anti-ColX (1:80 dilution) (Research Diagnostics, Flanders, NJ). ABC complex (Vector Laboratories, Burlingame, CA) was applied to the sections following the incubation with biotinylated secondary antibody (Dako Corporation, Carpinteria, CA). AEC plus substrate in red color (Dako, Carpinteria, CA) was used as a chromagen, and the sections were counterstained with light Hematoxylin. Phosphate-buffered saline substituted for the primary antibody was utilized as a negative control.

ACKNOWLEDGMENTS

We thank Renny T Franceschi for kindly providing us with the AdBMP-2 seed adenovirus. This work was supported by National Natural Science Foundation of China 30400502, Science and Technology Commission of Shanghai Municipality 04dz05601, 05DJ14006, 055407034, Shanghai Rising-star Program 05QMX1426. Shanghai Education Committee 03BC39, 04YQHB081, Y0203, National High Technology and Development Program of China 2002AA205011, and by grants from the Wunderman Family Foundation, March of Dimes Birth Defect Foundation (6-FY02-163), NIH/NIDCR RO3 DE 014649-01, NIH/NIDCR K23DE00422, NIH DE016107, and the Thomas R. Bales Endowed Chair.

Footnotes

The authors report a conflict of interest as Xinli Zhang, K.T., B.W., and C.S. are co-founders of Bone Biologics, Inc.

REFERENCES

  • 1.Bauer TW, Muschler GF. Bone graft materials. An overview of the basic science. Clin Orthop Relat Res. 2000;371:10–27. [PubMed] [Google Scholar]
  • 2.King GN. The importance of drug delivery to optimize the effects of bone morphogenetic proteins during periodontal regeneration. Curr Pharm Biotechnol. 2001;2:131–142. doi: 10.2174/1389201013378716. [DOI] [PubMed] [Google Scholar]
  • 3.Yoneda M, Terai H, Imai Y, Okada T, Nozaki K, Inoue H, et al. Repair of an intercalated long bone defect with a synthetic biodegradable bone-inducing implant. Biomaterials. 2005;26:5145–5152. doi: 10.1016/j.biomaterials.2005.01.054. [DOI] [PubMed] [Google Scholar]
  • 4.Govender S, Csimma C, Genant HK, Valentin-Opran A, Amit Y, Arbel R, et al. Recombinant human bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, controlled, randomized study of four hundred and fifty patients. J Bone Joint Surg Am. 2002;84-A:2123–2134. doi: 10.2106/00004623-200212000-00001. [DOI] [PubMed] [Google Scholar]
  • 5.Johnson EE, Urist MR. Human bone morphogenetic protein allografting for reconstruction of femoral nonunion. Clin Orthop Relat Res. 2000;371:61–74. doi: 10.1097/00003086-200002000-00008. [DOI] [PubMed] [Google Scholar]
  • 6.Valentin-Opran A, Wozney J, Csimma C, Lilly L, Riedel GE. Clinical evaluation of recombinant human bone morphogenetic protein-2. Clin Orthop Relat Res. 2002;395:110–120. doi: 10.1097/00003086-200202000-00011. [DOI] [PubMed] [Google Scholar]
  • 7.Bonadio J, Smiley E, Patil P, Goldstein S. Localized, direct plasmid gene delivery in vivo: prolonged therapy results in reproducible tissue regeneration. Nat Med. 1999;5:753–759. doi: 10.1038/10473. [DOI] [PubMed] [Google Scholar]
  • 8.van den Bergh JP, ten Bruggenkate CM, Groeneveld HH, Burger EH, Tuinzing DB. Recombinant human bone morphogenetic protein-7 in maxillary sinus floor elevation surgery in 3 patients compared to autogenous bone grafts. A clinical pilot study. J Clin Periodontol. 2000;27:627–636. doi: 10.1034/j.1600-051x.2000.027009627.x. [DOI] [PubMed] [Google Scholar]
  • 9.Tang TT, Xu XL, Dai KR, Yu CF, Yue B, Lou JR. Ectopic bone formation of human bone morphogenetic protein-2 gene transfected goat bone marrow-derived mesenchymal stem cells in nude mice. Chin J Traumatol. 2005;8:3–7. [PubMed] [Google Scholar]
  • 10.Lieberman JR, Daluiski A, Stevenson S, Wu L, McAllister P, Lee YP, et al. The effect of regional gene therapy with bone morphogenetic protein-2-producing bone-marrow cells on the repair of segmental femoral defects in rats. J Bone Joint Surg Am. 1999;81:905–917. doi: 10.2106/00004623-199907000-00002. [DOI] [PubMed] [Google Scholar]
  • 11.Chang SC, Wei FC, Chuang H, Chen YR, Chen JK, Lee KC, et al. Ex vivo gene therapy in autologous critical-size craniofacial bone regeneration. Plast Reconstr Surg. 2003;112:1841–1850. doi: 10.1097/01.PRS.0000091168.73462.1A. [DOI] [PubMed] [Google Scholar]
  • 12.Kang R, Ghivizzani SC, Muzzonigro TS, Herndon JH, Robbins PD, Evans CH. The Marshall R. Urist Young Investigator Award. Orthopaedic applications of gene therapy. From concept to clinic. Clin Orthop Relat Res. 2000;375:324–337. [PubMed] [Google Scholar]
  • 13.Owen M, Friedenstein AJ. Stromal stem cells: marrow-derived osteogenic precursors. Ciba Found Symp. 1988;136:42–60. doi: 10.1002/9780470513637.ch4. [DOI] [PubMed] [Google Scholar]
  • 14.Bianco P, Riminucci M, Gronthos S, Robey PG. Bone marrow stromal stem cells: nature, biology and potential applications. Stem Cells. 2001;19:180–192. doi: 10.1634/stemcells.19-3-180. [DOI] [PubMed] [Google Scholar]
  • 15.Quarto R, Mastrogiacomo M, Cancedda R, Kutepov SM, Mukhachev V, Lavroukov A, et al. Repair of large bone defects with the use of autologous bone marrow stromal cells. N Engl J Med. 2001;344:385–386. doi: 10.1056/NEJM200102013440516. [DOI] [PubMed] [Google Scholar]
  • 16.Yoo JU, Barthel TS, Nishimura K, Solchaga L, Caplan AI, Goldberg VM, et al. The chondrogenic potential of human bone-marrow-derived mesenchymal progenitor cells. J Bone Joint Surg Am. 1998;80:1745–1757. doi: 10.2106/00004623-199812000-00004. [DOI] [PubMed] [Google Scholar]
  • 17.Krebsbach PH, Mankani MH, Satomura K, Kuznetsov SA, Robey PG. Repair of craniotomy defects using bone marrow stromal cells. Transplantation. 1998;66:1272–1278. doi: 10.1097/00007890-199811270-00002. [DOI] [PubMed] [Google Scholar]
  • 18.Park J, Ries J, Gelse K, Kloss F, von der Mark K, Wiltfang J, et al. Bone regeneration in critical size defects by cell-mediated BMP-2 gene transfer: a comparison of adenoviral vectors and liposomes. Gene Ther. 2003;10:1089–1098. doi: 10.1038/sj.gt.3301960. [DOI] [PubMed] [Google Scholar]
  • 19.Huang W, Carlsen B, Wulur I, Rudkin G, Ishida K, Wu B, et al. BMP-2 exerts differential effects on differentiation of rabbit bone marrow stromal cells grown in two-dimensional and three-dimensional systems and is required for in vitro bone formation in a PLGA scaffold. Exp Cell Res. 2004;299:325–334. doi: 10.1016/j.yexcr.2004.04.051. [DOI] [PubMed] [Google Scholar]
  • 20.Jiang XQ, Chen JG, Gittens S, Chen CJ, Zhang XL, Zhang ZY. The ectopic study of tissue-engineered bone with hBMP-4 gene modified bone marrow stromal cells in rabbits. Chin Med J (Engl) 2005;118:281–288. [PubMed] [Google Scholar]
  • 21.Sugiyama O, Orimo H, Suzuki S, Yamashita K, Ito H, Shimada T. Bone formation following transplantation of genetically modified primary bone marrow stromal cells. J Orthop Res. 2003;21:630–637. doi: 10.1016/S0736-0266(02)00260-7. [DOI] [PubMed] [Google Scholar]
  • 22.Dai KR, Xu XL, Tang TT, Zhu ZA, Yu CF, Lou JR, et al. Repairing of goat Tibial Bone Defects with BMP-2 Gene-Modified Tissue-Engineered Bone. Calcif Tissue Int. 2005;77:55–61. doi: 10.1007/s00223-004-0095-z. [DOI] [PubMed] [Google Scholar]
  • 23.Zhao Z, Zhao M, Xiao G, Franceschi RT. Gene transfer of the Runx2 transcription factor enhances osteogenic activity of bone marrow stromal cells in vitro and in vivo. Mol Ther. 2005;12:247–253. doi: 10.1016/j.ymthe.2005.03.009. [DOI] [PubMed] [Google Scholar]
  • 24.Franceschi RT, Yang S, Rutherford RB, Krebsbach PH, Zhao M, Wang D. Gene therapy approaches for bone regeneration. Cells Tissues Organs. 2004;176:95–108. doi: 10.1159/000075031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ting K, Vastardis H, Mulliken JB, Soo C, Tieu A, Do H, et al. Human NELL-1 expressed in unilateral coronal synostosis. J Bone Miner Res. 1999;14:80–89. doi: 10.1359/jbmr.1999.14.1.80. [DOI] [PubMed] [Google Scholar]
  • 26.Watanabe TK, Katagiri T, Suzuki M, Shimizu F, Fujiwara T, Kanemoto N, et al. Cloning and characterization of two novel human cDNAs (NELL1 and NELL2) encoding proteins with six EGF-like repeats. Genomics. 1996;38:273–276. doi: 10.1006/geno.1996.0628. [DOI] [PubMed] [Google Scholar]
  • 27.Zhang X, Kuroda S, Carpenter D, Nishimura I, Soo C, Moats R, et al. Craniosynostosis in transgenic mice overexpressing Nell-1. J Clin Invest. 2002;110:861–870. doi: 10.1172/JCI15375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Desai J, Shannon ME, Johnson MD, Ruff DW, Hughes LA, Kerley MK, et al. Nell1-deficient mice have reduced expression of extracellular matrix proteins causing cranial and vertebral defects. Hum Mol Genet. 2006;15:1329–1341. doi: 10.1093/hmg/ddl053. [DOI] [PubMed] [Google Scholar]
  • 29.Cowan CM, Cheng S, Ting K, Soo C, Walder B, Wu B, et al. Nell-1 induced bone formation within the distracted intermaxillary suture. Bone. 2006;38:48–58. doi: 10.1016/j.bone.2005.06.023. [DOI] [PubMed] [Google Scholar]
  • 30.Truong T, Zhang X, Miao S, Chiu R, Ting K. Craniosynostosis-associated gene nell-1 is regulated by runx2. J Bone Miner Res. 2006;22:7–18. doi: 10.1359/jbmr.061012. [DOI] [PubMed] [Google Scholar]
  • 31.Aghaloo T, Cowan CM, Chou YF, Zhang X, Lee H, Miao S, et al. Nell-1-induced bone regeneration in calvarial defects. Am J Pathol. 2006;169:903–915. doi: 10.2353/ajpath.2006.051210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Cheng SL, Yang JW, Rifas L, Zhang SF, Avioli LV. Differentiation of human bone marrow osteogenic stromal cells in vitro: induction of the osteoblast phenotype by dexamethasone. Endocrinology. 1994;134:277–286. doi: 10.1210/endo.134.1.8275945. [DOI] [PubMed] [Google Scholar]
  • 33.Ohgushi H, Okumura M. Osteogenic capacity of rat and human marrow cells in porous ceramics. Experiments in athymic (nude) mice. Acta Orthop Scand. 1990;61:431–434. doi: 10.3109/17453679008993556. [DOI] [PubMed] [Google Scholar]
  • 34.Nakagawa T, Sugiyama T, Shimizu K, Murata T, Narita M, Nakamura S, et al. Characterization of the development of ectopic chondroid/bone matrix and chondrogenic/osteogenic cells during osteoinduction by rhBMP-2: a histochemical and ultrastructural study. Oral Dis. 2003;9:255–263. doi: 10.1034/j.1601-0825.2003.02912.x. [DOI] [PubMed] [Google Scholar]
  • 35.Krebsbach PH, Kuznetsov SA, Satomura K, Emmons RV, Rowe DW, Robey PG. Bone formation in vivo: comparison of osteogenesis by transplanted mouse and human marrow stromal fibroblasts. Transplantation. 1997;63:1059–1069. doi: 10.1097/00007890-199704270-00003. [DOI] [PubMed] [Google Scholar]
  • 36.Cunningham BW, Kanayama M, Parker LM, Weis JC, Sefter JC, Fedder IL, et al. Osteogenic protein versus autologous interbody arthrodesis in the sheep thoracic spine. A comparative endoscopic study using the Bagby and Kuslich interbody fusion device. Spine. 1999;24:509–518. doi: 10.1097/00007632-199903150-00002. [DOI] [PubMed] [Google Scholar]
  • 37.Magin MN, Delling G. Improved lumbar vertebral interbody fusion using rhOP-1: a comparison of autogenous bone graft, bovine hydroxylapatite (Bio-Oss) and BMP-7 (rhOP-1) in sheep. Spine. 2001;26:469–478. doi: 10.1097/00007632-200103010-00009. [DOI] [PubMed] [Google Scholar]
  • 38.Khan SN, Lane JM. Spinal fusion surgery: animal models for tissue-engineered bone constructs. Biomaterials. 2004;25:1475–1485. doi: 10.1016/s0142-9612(03)00491-5. [DOI] [PubMed] [Google Scholar]
  • 39.Sandhu HS, Khan SN. Animal models for preclinical assessment of bone morphogenetic proteins in the spine. Spine. 2002;27(16 suppl 1):S32–S38. doi: 10.1097/00007632-200208151-00008. [DOI] [PubMed] [Google Scholar]
  • 40.Kruyt MC, de Bruijn JD, Wilson CE, Oner FC, van Blitterswijik CA, Verbout JJ, et al. Viable osteogenic cells are obligatory for tissue-engineered ectopic bone formation in goats. Tissue Eng. 2003;9:327–336. doi: 10.1089/107632703764664792. [DOI] [PubMed] [Google Scholar]
  • 41.Mankani MH, Kuznetsov SA, Fowler B, Kingman A, Robey PG. In vivo bone formation by human bone marrow stromal cells: effect of carrier particle size and shape. Biotechnol Bioeng. 2001;72:96–107. doi: 10.1002/1097-0290(20010105)72:1<96::aid-bit13>3.0.co;2-a. [DOI] [PubMed] [Google Scholar]
  • 42.Sugiyama O, An DS, Kung SP, Feeley BT, Gamradt S, Liu NQ, et al. Lentivirus-mediated gene transfer induces long-term transgene expression of BMP-2 in vitro and new bone formation in vivo. Mol Ther. 2005;11:390–398. doi: 10.1016/j.ymthe.2004.10.019. [DOI] [PubMed] [Google Scholar]
  • 43.Yue B, Lu B, Dai KR, Zhang XL, Yu CF, Lou JR, et al. BMP2 gene therapy on the repair of bone defects of aged rats. Calcif Tissue Int. 2005;77:395–403. doi: 10.1007/s00223-005-0180-y. [DOI] [PubMed] [Google Scholar]
  • 44.Katagiri T, Yamaguchi A, Ikeda T, Yoshiki S, Wozney JM, Rosen V, et al. The non-osteogenic mouse pluripotent cell line, C3H10T1/2, is induced to differentiate into osteoblastic cells by recombinant human bone morphogenetic protein-2. Biochem Biophys Res Commun. 1990;172:295–299. doi: 10.1016/s0006-291x(05)80208-6. [DOI] [PubMed] [Google Scholar]
  • 45.Cowan CM, Jiang X, Hsu T, Soo C, Zhang B, Wang JZ, et al. Synergistic effects of Nell-1 and BMP-2 on the osteogenic differentiation of myoblasts. J Bone Miner Res. 2007;22:918–930. doi: 10.1359/jbmr.070312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Scammell BE, Roach HI. A new role for the chondrocyte in fracture repair: endochondral ossification includes direct bone formation by former chondrocytes. J Bone Miner Res. 1996;11:737–745. doi: 10.1002/jbmr.5650110604. [DOI] [PubMed] [Google Scholar]
  • 47.Akiyama H, Chaboissier MC, Martin JF, Schedl A, de Crombrugghe B. The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. Genes Dev. 2002;16:2813–2828. doi: 10.1101/gad.1017802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Pacifici M, Golden EB, Oshima O, Shapiro IM, Leboy PS, Adams SL. Hypertrophic chondrocytes. The terminal stage of differentiation in the chondrogenic cell lineage? Ann N Y Acad Sci. 1990;599:45–57. doi: 10.1111/j.1749-6632.1990.tb42363.x. [DOI] [PubMed] [Google Scholar]
  • 49.Zhang X, Carpenter D, Bokui N, Soo C, Miao S, Truong T, et al. Overexpression of Nell-1, a craniosynostosis-associated gene, induces apoptosis in osteoblasts during craniofacial development. J Bone Miner Res. 2003;18:2126–2134. doi: 10.1359/jbmr.2003.18.12.2126. [DOI] [PubMed] [Google Scholar]
  • 50.Kaplow LS. A histochemical procedure for localizing and evaluating leukocyte alkaline phosphatase activity in smears of blood and marrow. Blood. 1955;10:1023–1029. [PubMed] [Google Scholar]

RESOURCES