Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2018 Dec 6;21(2):110–117. doi: 10.1001/jamafacial.2018.1508

Assessment of Hedgehog Signaling Pathway Activation for Craniofacial Bone Regeneration in a Critical-Sized Rat Mandibular Defect

Matthew Q Miller 1,2, Logan F McColl 1,2, Michael R Arul 3, Jonathan Nip 3,4,5, Vedavathi Madhu 1, Gina Beck 1, Kishan Mathur 6, Vashaana Sahadeo 1, Jason R Kerrigan 1,6, Stephen S Park 2, J Jared Christophel 2, Abhijit S Dighe 1, Sangamesh G Kumbar 3,4,5,✉, Quanjun Cui 1,✉
PMCID: PMC6439804  NIHMSID: NIHMS1009152  PMID: 30520953

Key Points

Question

Does activation of the hedgehog signaling pathway increase bone regeneration compared with delivering growth factors alone?

Findings

In this study of 33 female Lewis rats, activation of the hedgehog signaling pathway through the delivery of smoothened agonist to a critical-sized rat mandibular defect was associated with increases in bone regeneration compared with delivering only growth factors. The most bone regeneration occurred when the defect was treated with bone morphogenetic protein 6, vascular endothelial growth factor, and smoothened agonist tethered to an osteogenic polysaccharide-based scaffold.

Meaning

This finding suggests that targeting the hedgehog signaling pathway may offer a new reconstructive option for bony craniofacial defects as well as nonunion and delayed healing fractures.

Abstract

Importance

Osseous craniofacial defects are currently reconstructed with bone grafting, rigid fixation, free tissue transfer, and/or recombinant human bone morphogenetic protein 2. Although these treatment options often have good outcomes, they are associated with substantial morbidity, and many patients are not candidates for free tissue transfer.

Objective

To assess whether polysaccharide-based scaffold (PS) constructs that are cross-linked with smoothened agonist (SAG), vascular endothelial growth factor (VEGF), and bone morphogenetic protein 6 (BMP-6) would substantially increase bone regeneration.

Design, Setting, and Participants

This animal model study was conducted at the University of Virginia School of Medicine Cui Laboratory from March 1, 2017, to June 30, 2017. Thirty-three 10-week-old female Lewis rats were acquired for the study. Bilateral nonsegmental critical-sized defects were created in the angle of rat mandibles. The defects were either left untreated or filled with 1 of the 9 PSs. The rats were killed after 8 weeks, and bone regeneration was evaluated using microcomputed tomographic imaging and mechanical testing. Analysis of variance testing was used to compare the treatment groups.

Main Outcomes and Measures

Blinded analysis and computer analysis of the microcomputed tomographic images were used to assess bone regeneration.

Results

In the 33 female Lewis rats, minimal healing was observed in the untreated mandibles. Addition of SAG was associated with increases in bone regeneration and bone density in all treatment groups, and maximum bone healing was seen in the group with BMP-6, VEGF, and SAG cross-linked to PS. For each of the 5 no scaffold group vs BMP-6, VEGF, and SAG cross-linked to PS group comparisons, mean defect bone regeneration was 4.14% (95% CI, 0.94%-7.33%) vs 66.19% (95% CI, 54.47%-77.90%); mean bone volume, 14.52 mm3 (95% CI, 13.07-15.97 mm3) vs 20.87 mm3 (95% CI, 14.73- 27.01 mm3); mean bone surface, 68.97 mm2 (95% CI, 60.08-77.85 mm2) vs 96.77 mm2 (95% CI, 76.11-117.43 mm2); mean ratio of bone volume to total volume, 0.11 (95% CI, 0.10-0.11) vs 0.15 (95% CI, 0.10-0.19); and mean connectivity density 0.03 (95% CI, 0.02-0.05) vs 0.32 (95% CI, 0.25-0.38). On mechanical testing, mandibles with untreated defects broke with less force than control mandibles in which no defect was made, although this force did not reach statistical significance. No significant difference in force to fracture was observed among the treatment groups.

Conclusions and Relevance

In this rat model study, activation of the hedgehog signaling pathway using smoothened agonist was associated with increased craniofacial bone regeneration compared with growth factors alone, including US Food and Drug Administration–approved recombinant human bone morphogenetic protein 2. Pharmaceuticals that target this pathway may offer a new reconstructive option for bony craniofacial defects as well as nonunion and delayed healing fractures.

Level of Evidence

NA.

This animal model study investigates whether a biosynthetic bone graft cross-linked with smoothened agonist activates the hedgehog signaling pathway in the reconstruction of craniofacial bony defects in female rats and if this is associated with bone regeneration.

Introduction

Bony craniofacial defects arise from ablative surgical procedures for both benign and malignant tumors as well as complex trauma and osteonecrosis. Current reconstructive options include free tissue transfer, bone grafting, rigid fixation, and synthetic implants. The outcomes are typically good with free tissue transfer, but these are complex procedures that require prolonged hospital admissions and are associated with donor-site morbidities and perioperative complications. Rates of perioperative complications associated with free tissue transfer range from 30% to 40% and can be as high as 60% in older patients with multiple comorbidities.1,2

Development of an osteogenic biosynthetic implant for craniofacial reconstruction could augment and, in some instances, obviate the need for these more traditional reconstructive methods. A US Food and Drug Administration (FDA)–approved bone graft (Infuse Bone Graft; Medtronic) uses recombinant human bone morphogenetic protein 2 (rhBMP-2) to treat complex long bone fractures and is frequently used in spinal fusion procedures.3 However, rhBMP-2 has limited indications for use in craniofacial reconstruction and is associated with substantial adverse effects when used off-label, necessitating the search for better alternatives to enhance craniofacial bone regeneration.4

Osteoconduction, osteoinduction, and osteogenesis are all required for new bone formation to occur.5 Historically, poly(lactic-co-glycolic acid) (PLGA) constructs were used as osteoconductive scaffolds because of their strength and absorptive characteristics. However, concerns over inflammation induced by PLGA’s acidic breakdown products as well as the absence of osseous integration have generated renewed research into scaffolds derived from natural polymers known to be more bioactive and biocompatible.6,7,8,9,10,11 Other research has developed and reported on a novel polysaccharide-based scaffold (PS) with superior strength compared with PLGA, an improved ability to support osteogenesis, and no foreign body reaction.12 Historically, natural polymer scaffolds lacked the mechanical stability to support healing and osteogenesis in bone regeneration. However, nanoengineering allows for the creation of natural polymer–based scaffolds to have the necessary strength to support bone regeneration. Furthermore, unlike PLGA, which is a synthetic polymer, neither a foreign body reaction nor acidic breakdown products were seen after the implant of PSs.12

Growth factors that augment both osteogenesis and angiogenesis are paramount to osteoinduction. The FDA approved rhBMP-2 for use in maxillary sinus floor and alveolar ridge augmentation. However, rhBMP-2 is delivered at supraphysiologic doses, which have been associated with adipogenic rather than osteogenic differentiation as well as edema, ectopic bone formation, and bone resorption.13,14,15 Furthermore, when used for off-label purposes, rhBMP-2 has a high rate of graft failure and other complications that often necessitate a second surgical procedure.8

We now know that bone morphogenetic protein (BMP) 6 and 9 are the most osteogenic BMPs, and studies have shown that vascular endothelial growth factor (VEGF) and BMP-6 work synergistically to produce additive effects on osteogenesis.16,17,18,19 However, when we delivered VEGF and BMP-6 to a critical-sized rat mandibular defect, substantial bone regeneration was not achieved and the new bone was poor quality.20

Previous findings suggest that delivering exogenous stem cells in conjunction with growth factors would be the ideal strategy for bone regeneration.19,21,22,23 Other groups have tried this strategy using bone marrow–derived mesenchymal stem cells and adipose-derived mesenchymal stem cells, but they achieved limited success.24,25,26 More recently, a new stem cell population, Gli1+ craniofacial-specific mesenchymal stem cells, was discovered in craniofacial sutures; these cells give rise to all craniofacial bones and are indispensable for bone growth and repair. Furthermore, these cells continue to reside in the suture mesenchyme in adult mice throughout the craniofacial skeleton, contributing to bone turnover and repair.27 Prior to this description of glioma-associated oncogene (Gli) protein 1 (Gli1) + craniofacial-specific mesenchymal stem cells, Chung et al28 described a similar craniofacial-specific stem cell population (cranial neural crest mesenchymal stem cells) with greater osteogenic potential in the craniofacial skeleton, compared with bone marrow–derived mesenchymal stem cells. The existence of a specific craniofacial mesenchymal stem cell may explain why bone marrow–derived mesenchymal stem cells and adipose-derived mesenchymal stem cells have had limited success in craniofacial bone regeneration.

The Gli1 is the transcriptional activator of Hedgehog (Hh) signaling, a signaling pathway critical to skeletal development, including the activation and osteogenic differentiation of Gli1+ craniofacial-specific mesenchymal stem cells.29,30 The Hh pathway relies on activation of the smoothened receptor, which is normally suppressed by the Patched receptor.30 Activation of the smoothened receptor is associated with activation of the Gli2 or Gli3 complex, which subsequently is associated with Gli1 expression. Although its mechanism is not well understood, Hh signaling encourages pluripotent mesenchymal stem cells to differentiate into osteoblasts while suppressing adipogenesis. This osteoblastogenesis is enhanced further in the presence of BMP signaling.30 Thus, smoothened agonist (SAG) functions as a downstream activator of the Hh signaling pathway, and delivery of exogenous SAG induces osteoblastic markers in mesenchymal stem cells (Figure 1).31,32

Figure 1. Hedgehog-Induced Osteogenesis .

Figure 1.

Open in a new tab

Hedgehog signaling encourages pluripotent mesenchymal stem cells to differentiate into osteoblasts while suppressing adipogenesis. Gli1, Gli2, and Gli3 indicate glioma-associated oncogene protein 1, 2, and 3; MSC, mesenchymal stem cells; PTCH, patched receptor; SAG, smoothened agonist; and SMO, smoothened receptor.

Lee et al33 used SAG to induce bone regeneration in a calvarial defect in mice in a dose-dependent manner, but healing was not complete. Treating bony defects with SAG in combination with BMPs should result in increased bone regeneration because Hh-induced osteogenesis requires BMP signaling, and Hh signaling suppresses the proadipogenic effects of BMP treatment.34,35 We hypothesized that in a rat mandibular critical-sized defect, PS constructs cross-linked with SAG, VEGF, and BMP-6 would significantly increase bone regeneration compared with the FDA-approved rhBMP-2.

Methods

This study was conducted at the University of Virginia School of Medicine Cui Laboratory, Charlottesville, from March 1, 2017, to June 30, 2017. Thirty-three 10-week-old female Lewis rats were acquired for the study. All surgical procedures on animals were performed in accordance with the protocols of the University of Virginia Animal Care and Use Committee.

Preparation of PSs

The PSs used to deliver the growth factors were composed of interconnected cellulose acetate microspheres, as previously described in another study.17 The pore diameter varied from 120 to 165 μm. The growth factors (VEGF: 2.5 μg; BMP-6: 2.5 μg; BMP-2: 11 μg) were tethered to the scaffolds using the EDC (1-ethyl-3-[3-dimethylaminopropyl] carbodiimide)-NHS (N-hydroxysuccinimide)-MES (4-morpholinoethanesulfonic acid) method, and 1 mg of SAG was adsorbed onto the PS. The BMP-6 and VEGF dose was chosen on the basis of the results of a previous study of the association between increasing BMP-2 dose and bone regeneration, the lack of regeneration seen in past work that used a lower dose of the growth factors, and the 1 to 1 BMP-6:VEGF ratio we have shown to generate cross-talk between the growth factors, enhancing osteogenesis.19,20,21,36 The higher dose of BMP-2 was used to imitate the supraphysiologic dose of BMP-2 delivered by the rhBMP-2 bone graft. We administered 1 mg of SAG as this dose is substantially greater than the EC50 (half-maximal effective concentration) of the drug, and previous groups have used a similar dose to induce bone regeneration in vivo.32,33,37

Creation of Mandibular Defects and Implant of Scaffolds

Bilateral mandibular defects were created in 33 ten-week-old female Lewis rats. General anesthesia was induced via intraperitoneal injection of ketamine hydrochloride and xylazine, according to a weight-based formula. The angle of mandible was identified and a 4-mm circular drill was used to create a circular noncontinuity defect in the angle of the mandible. The appropriate scaffold was then implanted and secured in the pterygomasseteric sling.

Bilateral defects were made in all rats, and scaffolds were implanted according to the treatment groups, as described in Table 1. The right and left hemimandible defects were identical; bilateral defects allowed us to test a greater number of treatment groups (Figure 2). After surgery, 10 mL of normal saline was injected subcutaneously to prevent dehydration, and buprenorphine hydrochloride was administered according to a weight-based formula. The rats were given antibiotic-containing water (Baytril; Bayer) for 1 week after the surgical procedure and maintained on a soft diet. No rats had to be killed because of failure to thrive, and no postoperative infections occurred.

Table 1. Treatment Grouping of 33 Female Lewis Rats.

Rat Label Mandible
Right Left
Control NS NS
Control NS NS
Control NS NS
Control NS NS
Control NS NS
Nude scaffold PS NS
Nude scaffold PS NS
Nude scaffold PS NS
Nude scaffold PS NS
VEGF VEGF cross-linked to PS BMP-2 cross-linked to PS
VEGF VEGF cross-linked to PS BMP-2 cross-linked to PS
VEGF VEGF cross-linked to PS BMP-2 cross-linked to PS
VEGF VEGF cross-linked to PS PS
BMP-6 BMP-6 cross-linked to PS BMP-2 cross-linked to PS
BMP-6 BMP-6 cross-linked to PS BMP-2 cross-linked to PS
BMP-6 BMP-6 cross-linked to PS PS
BMP-6 BMP-6 cross-linked to PS PS
BMP-6 and VEGF cross-linked to PS BMP-6 and VEGF cross-linked to PS BMP-2 cross-linked to PS
BMP-6 and VEGF cross-linked to PS BMP-6 and VEGF cross-linked to PS BMP-2 cross-linked to PS
BMP-6 and VEGF cross-linked to PS BMP-6 and VEGF cross-linked to PS BMP-2 cross-linked to PS
BMP-6 and VEGF cross-linked to PS BMP-6 and VEGF cross-linked to PS PS
VEGF and SAG cross-linked to PS VEGF and SAG cross-linked to PS BMP-2 and SAG crossed with PS
VEGF and SAG cross-linked to PS VEGF and SAG cross-linked to PS BMP-2 and SAG crossed with PS
VEGF and SAG cross-linked to PS VEGF and SAG cross-linked to PS BMP-2 and SAG crossed with PS
VEGF and SAG cross-linked to PS VEGF and SAG cross-linked to PS PS
BMP-6 and SAG cross-linked to PS BMP-6 and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6 and SAG cross-linked to PS BMP-6 and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6 and SAG cross-linked to PS BMP-6 and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6 and SAG cross-linked to PS BMP-6 and SAG cross-linked to PS PS
BMP-6, VEGF, and SAG cross-linked to PS BMP-6, VEGF, and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6, VEGF, and SAG cross-linked to PS BMP-6, VEGF, and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6, VEGF, and SAG cross-linked to PS BMP-6, VEGF, and SAG cross-linked to PS BMP-2 and SAG crossed with PS
BMP-6, VEGF, and SAG cross-linked to PS BMP-6, VEGF, and SAG cross-linked to PS PS
Open in a new tab

Abbreviations: BMP, bone morphogenetic protein; NS, no scaffold; PS, polysaccharide-based scaffold; SAG, smoothened agonist; VEGF, vascular endothelial growth factor.

Figure 2. Rat Mandible Critical-Sized Defect.

Figure 2.

Open in a new tab

Example of circular critical-sized defect made in rat hemimandible.

Microcomputed Tomographic Measurements for Evaluating New Bone Formation

At 8 weeks after the procedure, the rats were killed and the mandibles were explanted. Four mandibles from each of the 10 treatment groups underwent microcomputed tomographic (micro-CT) scanning and analysis (Scanco vivaCT 40 scanner; Scanco Medical AG). In the groups with more than 4 mandibles (control, PS, BMP-2 cross-linked to PS, and BMP-2 and SAG cross-linked to PS), baseline and 8-week x-ray imaging results were used to determine the 4 mandibles with the most bone regeneration, which were then selected for micro-CT imaging. The settings for the scanner were as follows: voxel size: 38 μm3; x-ray tube potential: 55 kV; and integration time: 145 milliseconds. Bone volume analysis was used to quantify the amount and density of bone formed; thresholds for bone detection for defined volumes of interest were set at a range of 158 to 1000 H. These settings captured new bone formation only, differentiating it from residual scaffold that had not undergone degradation by the experimental end point. The micro-CT images were also blindly analyzed to determine the area of bone regeneration.

Mechanical Testing

The harvested mandibles were loaded to failure in 3-point bending to assess their mechanical strength. Five groups were tested, with 3 mandibles in each group: 10-week-old female Lewis rat mandibles that had not undergone the procedure, untreated mandibles, PS mandibles, BMP-2 cross-linked to PS group, and BMP-2 and SAG cross-linked to PS group. These groups were selected because adequate numbers were available for both mechanical testing and micro-CT analysis. Further, these groups allowed the comparison of the PS alone, PS with FDA-approved BMP-2, and PS with BMP-2 and SAG to activate the Hh pathway. The mandibles were positioned lateral side down, and the central load was applied just posterior to the molars at 0.5 millimeter per second until gross failure occurred. The force-displacement values were recorded, and the force required to fracture was calculated.

Statistical Analysis

Statistical analysis was performed using SPSS statistics, version 24 (IBM). Analysis of variance testing was used to compare the treatment groups, and post hoc analysis using Tukey test was performed to minimize type I error and account for the unequal sample sizes. Significance was asserted at α = .05. One-way analysis of variance was used to calculate P values, and significance was asserted at P < .05. The analysis of variance model worked well for all comparisons studied (mean percent defect bone regeneration, F9,27 = 7.68, P < .001; mean bone volume, F9,27 = 7.26, P < .001; mean bone surface, F9,27 = 10.60, P < .001; mean ratio of bone volume to total volume, F9,27 = 9.00, P < .001; mean connective density, F9,27 = 10.82, P < .001). All groups were compared with one another using post hoc analysis. All comparisons were reported against the control group and nude scaffold group to assess how growth factors and SAG affect endogenous bone regeneration and regeneration induced by the osteogenic scaffold alone.

Results

Analysis of the quantity and quality of regenerated bone seen on micro-CT imaging was performed (Figure 3). No statistically significant bone regeneration in the control group was observed. No statistically significant differences in mean defect bone regeneration were seen in groups treated with PS, although all of these groups had statistically significantly more bone regeneration than did the control group. The addition of SAG to the treatment groups is associated with statistically significant increases in bone regeneration, and the most regeneration was seen in the BMP-6, VEGF, and SAG cross-linked to PS group. For each of the 5 no scaffold group vs BMP-6, VEGF, and SAG cross-linked to PS group comparisons, mean defect bone regeneration was 4.14% (95% CI, 0.94%-7.33%) vs 66.19% (95% CI, 54.47%-77.90%); mean bone volume, 14.52 mm3 (95% CI, 13.07-15.97 mm3) vs 20.87 mm3 (95% CI, 14.73- 27.01 mm3); mean bone surface, 68.97 mm2 (95% CI, 60.08-77.85 mm2) vs 96.77 mm2 (95% CI, 76.11-117.43 mm2); mean ratio of bone volume to total volume, 0.11 (95% CI, 0.10-0.11) vs 0.15 (95% CI, 0.10-0.19); and mean connectivity density 0.03 (95% CI, 0.02-0.05) vs 0.32 (95% CI, 0.25-0.38). The pro-osteogenic effect of SAG was also demonstrated by the statistically significantly increased bone volume, bone surface area, and bone density seen in mandibles treated with SAG (Table 2). For example, for each of the 3 no scaffold group vs treatment including SAG (VEGF and SAG cross-linked to PS) group, mean bone volume was 14.52 mm3 (95% CI, 13.07-15.97 mm3) vs 21.34 mm3 (95% CI, 19.61-23.08 mm3); mean bone surface, 68.97 mm2 (95% CI, 60.08-77.85 mm2) vs 95.44 mm2 (95% CI, 82.23-108.65 mm2); and mean connectivity density, 0.03 (95% CI, 0.02-0.05) vs 0.30 (95% CI, 0.02-0.58).

Figure 3. Microcomputed Tomographic Imaging.

Figure 3.

Open in a new tab

A-J, Representative images of the treatment groups. Red outlines represent blinded analysis of images. Black dashed circles represent standardized 4.5 mm circle. BMP-2 indicates bone morphogenetic protein 2; BMP-6, bone morphogenetic protein 6; PS, polysaccharide-based scaffold; SAG, smoothened agonist; and VEGF, vascular endothelial growth factor.

Table 2. Bone Regeneration Values for Different Treatment and Control Groups.

Treatment Group Mean DBR, % (95% CI)a P Valuea Mean BV, mm3 (95% CI)a P Valuea Mean BS, mm2 (95% CI)a P Valuea Mean BV/TV, % (95% CI)a P Valuea Mean CD, % (95% CI)a P Valuea
NS 4.14 (0.94-7.33) <.001 14.52 (13.07-15.97) .001 68.97 (60.08-77.85) .001 0.11 (0.10-0.11) .002 0.03 (0.02-0.05) .008
PS 42.97 (33.24-52.71) .28 16.59 (14.33-18.85) .04 67.59 (59.33-75.86) .001 0.12 (0.11-0.14) .01 0.06 (0.04-0.14) .002
VEGF cross-linked to PS 39.55 (19.86-59.24) .14 17.52 (15.08-19.97) .22 75.26 (65.65-84.87) .02 0.13 (0.12-0.14) .04 0.08 (−0.02 to 0.18) .04
BMP-2 cross-linked to PS 34.11 (−0.12 to 68.38) .06 16.16 (13.76-18.56) .04 68.15 (58.20-78.10) .002 0.12 (0.10-0.13) .03 0.06 (−0.07 to 0.20) .04
BMP-6 cross-linked to PS 34.25 (24.32-44.18) .04 16.42 (15.57-17.27) .03 70.37 (65.58-75.16) .002 0.12 (0.11-0.12) .02 0.13 (−0.08 to 0.35) .22
BMP-6 and VEGF cross-linked to PS 38.20 (22.31-54.09) .10 15.59 (12.70-18.48) .01 68.11 (58.26-77.98) .001 0.11 (0.09-0.13) .004 0.04 (0.00-0.08) .01
VEGF and SAG cross-linked to PS 51.41 (35.92-66.91) .82 21.34 (19.61-23.08) >.99 95.44 (82.23-108.65) >.99 0.16 (0.15-0.17) .97 0.30 (0.02-0.58) >.99
BMP-2 and SAG cross-linked to PS 38.89 (−38.46 to 116.24) .18 17.75 (11.76-23.74) .38 87.99 (62.93-113.04) .89 0.13 (0.08-0.17) .48 0.42 (0.16-0.68) .93
BMP-6 and SAG cross-linked to PS 60.34 (50.23-70.45) >.99 19.31 (15.80-22.82) .95 92.41 (73.29-111.53) >.99 0.14 (0.12-0.16) >.99 0.40 (0.27-0.53) .97
BMP-6, VEGF, and SAG cross-linked to PS 66.19 (54.47-77.90) NA 20.87 (14.73-27.01) NA 96.77 (76.11-117.43) NA 0.15 (0.10-0.19) NA 0.32 (0.25-0.38) NA
Open in a new tab

Abbreviations: BMP, bone morphogenetic protein; BS, bone surface area; BV, bone volume; BV/TV, ratio of bone volume to total volume; CD, connectivity density (trabeculae/mm3); DBR, defect bone regeneration; NA, not applicable; NS, no scaffold; PS, polysaccharide-based scaffold; SAG, smoothened agonist; VEGF, vascular endothelial growth factor.

a

P values and CIs were calculated using 1-way analysis of variance testing with Tukey post hoc analysis. All P values and CIs represent a comparison with the BMP-6 and VEGF and SAG cross-linked to PS group, the treatment resulting in the greatest percentage of defect bone regeneration on blinded analysis, and BV/TV and bone surface area on software analysis.

When the no procedure mandibles, untreated mandibles, PS mandibles, BMP-2 cross-linked to PS group, and BMP-2 and SAG cross-linked to PS group underwent the 3-point bending assay test, no statistically significant difference was observed in force to fracture between the treatment groups and the no procedure mandibles. However, the group in which a defect was created but not treated fractured with less force: The mean (SD) peak load for the no procedure mandibles was 86.30 (9.00); untreated mandibles, 77.60 (20.10; P = .40); PS mandibles, 106.50 (14.60; P = .10); BMP-2 and SAG cross-linked to PS group, 88.70 (9.60; P = .80); and BMP-2 and SAG cross-linked to PS group, 97.60 (4.70; P = .30).

Discussion

Developing an osteogenic biosynthetic implant for craniofacial reconstruction could alter the standards of care in trauma management, osteonecrosis treatment, and repair of oncologic surgical defects. Such an implant could decrease perioperative morbidities associated with current reconstructive methods and could provide a reconstructive option for patients unable to undergo free tissue transfer.

Other research has attempted to use growth factors in isolation and in combination with stem cells to achieve craniofacial bone regeneration with limited success. To our knowledge, the in vivo bone regeneration potential of recently discovered craniofacial stem cells has not been studied, but the invasiveness of such harvest and regulatory issues will likely preclude these interventions from being used in clinical therapies for quite some time. However, as these stem cells differentiate into osteoblastic lineages via Hh signaling and as SAG is a downstream activator of the Hh pathway, treatment with this small-molecule therapy may enhance endogenous stem cell differentiation and craniofacial osteogenesis.

Hence, we hypothesized that delivering SAG to a critical-sized rat mandibular defect in combination with BMP-6 and VEGF on a PS would lead to more bone regeneration than would delivery of the FDA-approved rhBMP-2 alone. Our results demonstrate the advantages in total bone regeneration and bone density seen by treating critical-sized defects with SAG in addition to growth factors. The most bone regeneration was seen in the BMP-6 and SAG cross-linked to the PS group and the BMP-6, VEGF, and SAG cross-linked to PS group (Figure 1, Figure 2, and Figure 3; Table 2). This reflects the synergistic relationship seen between BMPs and Hh signaling–induced osteogenesis seen in vitro.33,38,39 When comparing the BMP-6, VEGF, and SAG cross-linked to the PS group with the BMP-2 cross-linked to the PS group (the group most reflective of the rhBMP-2 bone graft), the defects treated with BMP-6, VEGF, and SAG cross-linked to PS had significantly greater bone regeneration and density of regenerated bone (Table 2).

With regard to mechanical testing, no significant difference in force to fracture of the treatment groups was observed, compared with the hemimandibles in which a surgical procedure was not performed. However, the hemimandibles where a defect was made but no scaffold was placed broke with the least force, and the mandibles in which a nude scaffold was placed required the greatest force to fracture. All groups in which a scaffold was placed required more force to fracture compared with the hemimandibles in which no defect was created. These results likely illustrate the mechanical strength of the PS used to deliver the growth factors; the PS has been shown to be superior to PLGA and equal to trabecular bone.12 Because of the intrinsic strength of the PS, however, drawing conclusions is difficult regarding the association of Hh activation with growth factor delivery using our mechanical testing model; we believe the micro-CT data are of more clinical relevance.

Note that, although significant bone regeneration was seen on micro-CT in the SAG treatment groups, no treatment group demonstrated complete ossification of the critical-sized mandibular defect. This result may be associated with the suboptimal dose, inadequate treatment duration, or limitations of Hh signaling-induced bone formation. The treatment duration of 8 weeks was chosen because previous groups used this duration in their study of SAG to induce craniofacial bone regeneration.32,33 However, the critical-sized defects used in our study were not segmental; thus, the mechanotransductive bone healing processes that occur in fracture healing and segmental bone loss are likely not as robust in our model.40,41

Of importance is that in vitro data demonstrate that BMP-2 expression and exposure of oral cavity squamous cell carcinoma antigen tissues to rhBMP-2 may increase tumor invasiveness.38,39 In one study, 17 patients, in whom rhBMP-2 was used with vascularized bony free tissue transfer to treat osteoradionecrosis of the mandible, demonstrated no recurrences.42 Given the mixed evidence on whether rhBMP-2 is associated with worse oncologic outcomes, it is appropriately contraindicated for use in patients with active malignant neoplasm. Because one of the most common causes of craniofacial bony defects is ablative oncologic surgical procedure, the implications of activation of Hh signaling for oncologic outcomes will require careful study. Smoothened is a human proto-oncogene with activating mutations found in cutaneous basal cell carcinoma and medulloblastoma.43 Current research into how Hh signaling affects oral cavity squamous cell carcinoma antigen is limited. One paper demonstrated that Broders grade and N stage were associated with lower Gli1 expression when oral cavity and oropharyngeal squamous cell carcinoma antigen tissues were exposed to sonic Hh protein and examined in vitro.44 Future research is needed to clarify the association between hedgehog signaling and oral cavity squamous cell carcinoma before any Hh-activating bone graft can be used to reconstruct patients with active cancer.

Limitations

The circular, nonsegmental defect used in our rat model is a limitation of this study in that the defect is not physiologic. Activation of the Hh signaling pathway was associated with significant bone regeneration in this model, but further testing in a segmental defect model as well as in nonunion and delayed healing fracture models will provide greater insight into the potential clinical utility of the PS, our biosynthetic bone graft. The 3-point bending test was successful in testing the strength of the rat hemimandibles, but the intrinsic strength of the PS (stronger than the native rat mandibular bone) makes drawing conclusions difficult regarding how BMP-2 and SAG affect the strength of regenerated bone. However, our results do show that SAG is associated with significantly denser bone than bone regenerated by growth factors alone without SAG. In addition, we do not believe that implanting scaffolds with different growth factors in the contralateral hemimandibles had systemic implications, but there was no way to control for this variable. Future work could perform unilateral implants to control for this possibility. Last, the mechanisms of Hh-signaling bone induction and growth factor bone induction have been well elucidated in vitro, but their interaction with the novel PS we developed has not been studied to our knowledge. Future studies into the in vitro interaction between SAG and growth factors on our PSs are needed, and such research will also investigate the potential of BMP-6 and SAG cross-linked to PS and BMP-6, VEGF, and SAG cross-linked to PS to create bony union in a critical-sized, segmental mandibular defect model.

Conclusions

Activation of the Hh signaling pathway using SAG appears to increase craniofacial bone regeneration compared with growth factors alone, including the FDA-approved rhBMP-2. Pharmaceuticals that target this pathway may offer a new reconstructive option for bony craniofacial defects as well as nonunion and delayed-healing fractures. The PS reported in this study, cross-linked with BMP-6 and SAG with or without VEGF, may prove to be a suitable bone graft substitute that may be used to reconstruct larger defects than what can currently be reconstructed with current FDA-approved rhBMP-2.

References

  • 1.Suh JD, Sercarz JA, Abemayor E, et al. Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction. Arch Otolaryngol Head Neck Surg. 2004;130(8):962-966. doi: 10.1001/archotol.130.8.962 [DOI] [PubMed] [Google Scholar]
  • 2.Blackwell KE, Azizzadeh B, Ayala C, Rawnsley JD. Octogenarian free flap reconstruction: complications and cost of therapy. Otolaryngol Head Neck Surg. 2002;126(3):301-306. doi: 10.1067/mhn.2002.122704 [DOI] [PubMed] [Google Scholar]
  • 3.Swiontkowski MF, Aro HT, Donell S, et al. Recombinant human bone morphogenetic protein-2 in open tibial fractures: a subgroup analysis of data combined from two prospective randomized studies. J Bone Joint Surg Am. 2006;88(6):1258-1265. doi: 10.2106/JBJS.E.00499 [DOI] [PubMed] [Google Scholar]
  • 4.Woo EJ. Adverse events reported after the use of recombinant human bone morphogenetic protein 2. J Oral Maxillofac Surg. 2012;70(4):765-767. doi: 10.1016/j.joms.2011.09.008 [DOI] [PubMed] [Google Scholar]
  • 5.Fishero BA, Kohli N, Das A, Christophel JJ, Cui Q. Current concepts of bone tissue engineering for craniofacial bone defect repair. Craniomaxillofac Trauma Reconstr. 2015;8(1):23-30. doi: 10.1055/s-0034-1393724 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Stähelin AC, Weiler A, Rüfenacht H, Hoffmann R, Geissmann A, Feinstein R. Clinical degradation and biocompatibility of different bioabsorbable interference screws: a report of six cases. Arthroscopy. 1997;13(2):238-244. doi: 10.1016/S0749-8063(97)90162-6 [DOI] [PubMed] [Google Scholar]
  • 7.Stähelin AC, Weiler A. All-inside anterior cruciate ligament reconstruction using semitendinosus tendon and soft threaded biodegradable interference screw fixation. Arthroscopy. 1997;13(6):773-779. doi: 10.1016/S0749-8063(97)90019-0 [DOI] [PubMed] [Google Scholar]
  • 8.Böstman OM. Osteoarthritis of the ankle after foreign-body reaction to absorbable pins and screws: a three- to nine-year follow-up study. J Bone Joint Surg Br. 1998;80(2):333-338. doi: 10.1302/0301-620X.80B2.8302 [DOI] [PubMed] [Google Scholar]
  • 9.Weiler A, Scheffler SU, Südkamp NP. Current aspects of anchoring hamstring tendon transplants in cruciate ligament surgery [in German]. Chirurg. 2000;71(9):1034-1044. doi: 10.1007/s001040051179 [DOI] [PubMed] [Google Scholar]
  • 10.Nudelman F, Pieterse K, George A, et al. The role of collagen in bone apatite formation in the presence of hydroxyapatite nucleation inhibitors. Nat Mater. 2010;9(12):1004-1009. doi: 10.1038/nmat2875 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Travan A, Marsich E, Donati I, et al. Polysaccharide-coated thermosets for orthopedic applications: from material characterization to in vivo tests. Biomacromolecules. 2012;13(5):1564-1572. doi: 10.1021/bm3002683 [DOI] [PubMed] [Google Scholar]
  • 12.Aravamudhan A, Ramos DM, Nip J, et al. Cellulose and collagen derived micro-nano structured scaffolds for bone tissue engineering. J Biomed Nanotechnol. 2013;9(4):719-731. doi: 10.1166/jbn.2013.1574 [DOI] [PubMed] [Google Scholar]
  • 13.Kim IS, Lee EN, Cho TH, et al. Promising efficacy of Escherichia coli recombinant human bone morphogenetic protein-2 in collagen sponge for ectopic and orthotopic bone formation and comparison with mammalian cell recombinant human bone morphogenetic protein-2. Tissue Eng Part A. 2011;17(3-4):337-348. doi: 10.1089/ten.tea.2010.0408 [DOI] [PubMed] [Google Scholar]
  • 14.Zhang Q, He QF, Zhang TH, Yu XL, Liu Q, Deng FL. Improvement in the delivery system of bone morphogenetic protein-2: a new approach to promote bone formation. Biomed Mater. 2012;7(4):045002. doi: 10.1088/1748-6041/7/4/045002 [DOI] [PubMed] [Google Scholar]
  • 15.Wu G, Liu Y, Iizuka T, Hunziker EB. The effect of a slow mode of BMP-2 delivery on the inflammatory response provoked by bone-defect-filling polymeric scaffolds. Biomaterials. 2010;31(29):7485-7493. doi: 10.1016/j.biomaterials.2010.06.037 [DOI] [PubMed] [Google Scholar]
  • 16.Kang Q, Sun MH, Cheng H, et al. Characterization of the distinct orthotopic bone-forming activity of 14 BMPs using recombinant adenovirus-mediated gene delivery. Gene Ther. 2004;11(17):1312-1320. doi: 10.1038/sj.gt.3302298 [DOI] [PubMed] [Google Scholar]
  • 17.Wang EA, Rosen V, D’Alessandro JS, et al. Recombinant human bone morphogenetic protein induces bone formation. Proc Natl Acad Sci U S A. 1990;87(6):2220-2224. doi: 10.1073/pnas.87.6.2220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Scharpfenecker M, van Dinther M, Liu Z, et al. BMP-9 signals via ALK1 and inhibits bFGF-induced endothelial cell proliferation and VEGF-stimulated angiogenesis. J Cell Sci. 2007;120(Pt 6):964-972. doi: 10.1242/jcs.002949 [DOI] [PubMed] [Google Scholar]
  • 19.Cui F, Wang X, Liu X, Dighe AS, Balian G, Cui Q. VEGF and BMP-6 enhance bone formation mediated by cloned mouse osteoprogenitor cells. Growth Factors. 2010;28(5):306-317. doi: 10.3109/08977194.2010.484423 [DOI] [PubMed] [Google Scholar]
  • 20.Das A, Fishero BA, Christophel JJ, et al. Poly(lactic-co-glycolide) polymer constructs cross-linked with human BMP-6 and VEGF protein significantly enhance rat mandible defect repair. Cell Tissue Res. 2016;364(1):125-135. doi: 10.1007/s00441-015-2301-x [DOI] [PubMed] [Google Scholar]
  • 21.Seamon J, Wang X, Cui F, et al. Adenoviral delivery of the VEGF and BMP-6 genes to rat mesenchymal stem cells potentiates osteogenesis. Bone Marrow Res. 2013;2013:737580. doi: 10.1155/2013/737580 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Madhu V, Li CJ, Dighe AS, Balian G, Cui Q. BMP-non-responsive Sca1+ CD73+ CD44+ mouse bone marrow derived osteoprogenitor cells respond to combination of VEGF and BMP-6 to display enhanced osteoblastic differentiation and ectopic bone formation. PLoS One. 2014;9(7):e103060. doi: 10.1371/journal.pone.0103060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wang X, Cui F, Madhu V, Dighe AS, Balian G, Cui Q. Combined VEGF and LMP-1 delivery enhances osteoprogenitor cell differentiation and ectopic bone formation. Growth Factors. 2011;29(1):36-48. doi: 10.3109/08977194.2010.544656 [DOI] [PubMed] [Google Scholar]
  • 24.Kaigler D, Pagni G, Park CH, et al. Stem cell therapy for craniofacial bone regeneration: a randomized, controlled feasibility trial. Cell Transplant. 2013;22(5):767-777. doi: 10.3727/096368912X652968 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bajestan MN, Rajan A, Edwards SP, et al. Stem cell therapy for reconstruction of alveolar cleft and trauma defects in adults: a randomized controlled, clinical trial. Clin Implant Dent Relat Res. 2017;19(5):793-801. doi: 10.1111/cid.12506 [DOI] [PubMed] [Google Scholar]
  • 26.Lee MK, DeConde AS, Lee M, et al. Biomimetic scaffolds facilitate healing of critical-sized segmental mandibular defects. Am J Otolaryngol. 2015;36(1):1-6. doi: 10.1016/j.amjoto.2014.06.007 [DOI] [PubMed] [Google Scholar]
  • 27.Zhao H, Feng J, Ho TV, Grimes W, Urata M, Chai Y. The suture provides a niche for mesenchymal stem cells of craniofacial bones. Nat Cell Biol. 2015;17(4):386-396. doi: 10.1038/ncb3139 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Chung IH, Yamaza T, Zhao H, Choung PH, Shi S, Chai Y. Stem cell property of postmigratory cranial neural crest cells and their utility in alveolar bone regeneration and tooth development. Stem Cells. 2009;27(4):866-877. doi: 10.1002/stem.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Riddle RD, Johnson RL, Laufer E, Tabin C. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell. 1993;75(7):1401-1416. doi: 10.1016/0092-8674(93)90626-2 [DOI] [PubMed] [Google Scholar]
  • 30.James AW. Review of signaling pathways governing MSC osteogenic and adipogenic differentiation. Scientifica (Cairo). 2013;2013:684736.doi: 10.1155/2013/684736 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hojo H, Ohba S, Yano F, et al. Gli1 protein participates in Hedgehog-mediated specification of osteoblast lineage during endochondral ossification. J Biol Chem. 2012;287(21):17860-17869. doi: 10.1074/jbc.M112.347716 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Lee S, Wang C, Pan HC, et al. Combining Smoothened Agonist and NEL-Like Protein-1 Enhances Bone Healing. Plast Reconstr Surg. 2017;139(6):1385-1396. doi: 10.1097/PRS.0000000000003367 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Lee S, Shen J, Pan HC, et al. Calvarial defect healing induced by small molecule smoothened agonist. Tissue Eng Part A. 2016;22(23-24):1357-1366. doi: 10.1089/ten.tea.2016.0167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Yuasa T, Kataoka H, Kinto N, et al. Sonic hedgehog is involved in osteoblast differentiation by cooperating with BMP-2. J Cell Physiol. 2002;193(2):225-232. doi: 10.1002/jcp.10166 [DOI] [PubMed] [Google Scholar]
  • 35.Zehentner BK, Leser U, Burtscher H. BMP-2 and sonic hedgehog have contrary effects on adipocyte-like differentiation of C3H10T1/2 cells. DNA Cell Biol. 2000;19(5):275-281. doi: 10.1089/10445490050021186 [DOI] [PubMed] [Google Scholar]
  • 36.Cowan CM, Aghaloo T, Chou YF, et al. MicroCT evaluation of three-dimensional mineralization in response to BMP-2 doses in vitro and in critical sized rat calvarial defects. Tissue Eng. 2007;13(3):501-512. doi: 10.1089/ten.2006.0141 [DOI] [PubMed] [Google Scholar]
  • 37.Wang J, Lu J, Bond MC, et al. Identification of select glucocorticoids as smoothened agonists: potential utility for regenerative medicine. Proc Natl Acad Sci U S A. 2010;107(20):9323-9328. doi: 10.1073/pnas.0910712107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zaid KW, Chantiri M, Bassit G. Recombinant human bone morphogenetic protein-2 in development and progression of oral squamous cell carcinoma. Asian Pac J Cancer Prev. 2016;17(3):927-932. doi: 10.7314/APJCP.2016.17.3.927 [DOI] [PubMed] [Google Scholar]
  • 39.Kim MJ, Kim KM, Kim J, Kim KN. BMP-2 promotes oral squamous carcinoma cell invasion by inducing CCL5 release. PLoS One. 2014;9(10):e108170. doi: 10.1371/journal.pone.0108170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Loi F, Córdova LA, Pajarinen J, Lin TH, Yao Z, Goodman SB. Inflammation, fracture and bone repair. Bone. 2016;86:119-130. doi: 10.1016/j.bone.2016.02.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Huang C, Ogawa R. Mechanotransduction in bone repair and regeneration. FASEB J. 2010;24(10):3625-3632. doi: 10.1096/fj.10-157370 [DOI] [PubMed] [Google Scholar]
  • 42.Sweeny L, Lancaster WP, Dean NR, et al. Use of recombinant bone morphogenetic protein 2 in free flap reconstruction for osteonecrosis of the mandible. J Oral Maxillofac Surg. 2012;70(8):1991-1996. doi: 10.1016/j.joms.2011.08.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Rubin LL, de Sauvage FJ. Targeting the Hedgehog pathway in cancer. Nat Rev Drug Discov. 2006;5(12):1026-1033. doi: 10.1038/nrd2086 [DOI] [PubMed] [Google Scholar]
  • 44.Leovic D, Sabol M, Ozretic P, et al. Hh-Gli signaling pathway activity in oral and oropharyngeal squamous cell carcinoma. Head Neck. 2012;34(1):104-112. doi: 10.1002/hed.21696 [DOI] [PubMed] [Google Scholar]

Articles from JAMA Facial Plastic Surgery are provided here courtesy of American Medical Association

RESOURCES