Bone Engineering of the Rabbit Ulna

J Oral Maxillofac Surg 65:1495-1502, 2007

Bone Engineering of the Rabbit Ulna Ahmed El-Ghannam, PhD,* Larry Cunningham, Jr, DDS, MD,† David Pienkowski, PhD,‡ and Amanda Hart, MS§

Purpose: The purpose of the present preliminary study is to show that a novel 3-dimensional porous

silica-calcium phosphate nanocomposite (SCPC) can provide a controlled release of rhBMP-2 and regenerate bone in a load-bearing segmental defect. Materials and Methods: A bone replica of the rabbit ulna was created from SCPC powder using rapid prototyping technology. The ceramic bone replica was coated with rhBMP-2 and then implanted into a 10-mm segmental defect created in a rabbit ulna and fixated with a 1-mm titanium adaptation plate. Bone healing was evaluated using computed tomography (CT) scan, histomorphometry, and biomechanical techniques. The release kinetics of rhBMP-2 and the dissolution kinetics were also determined in vitro. Statistical analysis was performed to compare the biomechanical strength of the grafted bone with the contralateral unoperated ulna. Results: After 4 weeks, CT scans showed that the critical size defect had been replaced by newly formed bone. Torsional testing of the ulna after 12 weeks showed restoration of maximum torque and angle at failure. Histological evaluation showed that the regenerated bone had the morphological characteristics of mature bone. SCPC provided a sustained release profile of an effective dose of rhBMP-2 for 14 days. Conclusions: The SCPC-rhBMP-2 hybrid enhanced bone regeneration in a load-bearing segmental defect in a rabbit ulna. The regenerated bone acquired morphology and mechanical strength typical for natural bone. The enhanced bone formation correlates well with the surface bioactivity and effective release profile of rhBMP-2. The present preliminary study shows the proof of principles that porous, resorbable, bioactive SCPC-rhBMP-2 tissue engineering hybrid can serve as a substitute for autologous bone in load-bearing applications. This is a US government work. There are no restrictions on its use. Published by Elsevier, Inc on behalf of the American Association of Oral and Maxillofacial Surgeons. J Oral Maxillofac Surg 65:1495-1502, 2007 The repair of large segmental bone defects is a challenging problem in orthopedic and maxillofacial surgeries. Current treatment options include bone grafting (auto- or allogenic), vascularized bone grafts, or

Received from the University of Kentucky, Lexington, KY. *Associate Professor of Biomedical Engineering, Center for Biomedical Engineering and Center for Oral Health Research, College of Dentistry. †Associate Professor of Oral and Maxillofacial Surgery, Department of Oral and Maxillofacial Surgery, College of Dentistry. ‡Associate Professor of Biomechanics, Orthopaedic Biomechanics Laboratory, Center for Biomedical Engineering. §Master of Biomedical Engineering, Center for Biomedical Engineering. Supported by the Kentucky Science and Engineering Foundation (Grant KSEF-414-RDE-004) and NIH Grant P30-AR46031. Address correspondence and reprint requests to Dr El-Ghannam: Center for Biomedical Engineering, University of Kentucky, Rose Street, Lexington, KY 40506; e-mail: [email protected] This is a US government work. There are no restrictions on its use. Published by Elsevier, Inc on behalf of the American Association of Oral and Maxillofacial Surgeons. 0278-2391/07/6508-0010$32.00/0 doi:10.1016/j.joms.2006.10.031

distraction osteogenesis. However, these techniques often involve multistage surgical procedures, inhibit early limb function, and require several revision procedures to maintain acceptable alignment and achieve osseous healing. An alternative approach involves the design of biomaterials with specific osteoconductive and osteoinductive properties, geometry to provide the correct anatomy, and internal structure conducive to autogenous bone cells invasion and vascularization. These characteristics in principle represent the prerequisites for ideal graft integration and subsequent optimal mechanical performance. Bone morphogenetic proteins (BMPs) are biologically active osteoinductive cytokines with significant clinical potential, but the lack of a delivery system enabling full osteoinduction has precluded their wider implementation in clinical therapeutics. Incorporation of BMP in hydroxyapatite (HA) ceramic,1-3 HA/collagen,4 and HA/tricalcium phosphate (TCP) composites5-7 accelerate bone formation but require a high BMP dose (100 mg) because of their inability to retain BMPs.8 Other issues that must be considered for the effective application of HA ceramic are the limited resorbability and poor mechanical strength. Moreover, like the collagen composite, there were problems associated with inflammatory and immuno-

1495

1496 logic reactions when calcium phosphate ceramics were used as carriers for osteoinductive factors.9,10 Although biodegradable polymers are promising as a temporary tissue replacement and drug carrier, several aspects need to be envisaged for their hard tissue applications. One of the concerns is the acidic environment driven by the degradation of polymers, thereby raising concerns of inflammatory problems in vivo.11,12 Moreover, defect size limits the ability of synthetic polymers to promote bone formation. When 4-cm tibial defects in sheep were treated with only poly-L-lactide acid (PLLA) membranes, no osseous repair was noticed. In contrast, defects treated with PLLA-cancellous bone implant showed significant bony repair.13 The limitations of traditional bioactive ceramics and commonly used metallic and polymeric scaffold materials expose the need for a new generation of bioceramics that can specifically serve as bone grafts and tissue-engineering scaffolds for drug and cell delivery.14 The need to match anatomic shape of tissue and organs has prompted the use of rapid prototyping techniques (RPT) in implant processing. For bone tissue engineering, the bone graft substitute needs to have adequate porosity for bone repair so as to accelerate bone regeneration. The scaffold must also replicate the size and shape of the defect it is going to fill. Both of these characteristics synergistically play an integral role in the success of the bone substitute. An accurate repair of bone defects is not only important in obtaining restoration of bone function but is also important for cosmetic reasons as well. This is especially true for maxillofacial bone reconstructions. Traditional processing or machining of porous implants may result in blocking of the pores by the debris of the material, which hinders cell invasion and tissue ingrowth. The RPT provides a better choice for scaffold processing, due to the ability to engineer the scaffold architecture, material composition, and porosity, which are all critical factors in the development and future success of tissue engineering implant. Through this type of processing technology, it is possible to predefine and reproduce the internal morphology that replicates that of natural bone.15 The use of rapid prototyping techniques has been suggested for producing scaffolds with defined architectures, although typically these have focused on polymeric materials.15-17 Recently, we have been pursuing the idea of using a novel bioactive resorbable silica-calcium phosphate nanocomposite (SCPC) for bone engineering.18-23 Transmission electron microscopy indicated that SCPC is composed of nanocrystals in the size range of 50 to 300 nm and has homogenous distribution of nanopores in the size range of 50 to 100 nm.19 SCPC particulates, without BMP, enhanced bone regeneration and graft material resorp-

BONE ENGINEERING OF THE RABBIT ULNA

tion when implanted in a critical size bone defect in a rabbit femur.18 In the same study, control bioactive glass particulates enhanced bone formation, however, it did not resorb. When the SCPC material was loaded with rhBMP-2, superior stimulatory effect on bone cell function was observed compared with SCPC without BMP-2 (unpublished data). In addition, bone marrow mesenchymal stem cells attached to the SCPC-rhBMP-2 hybrid produced significantly higher alkaline phosphatase activity and mineralized bonelike tissue than cells attached to HA-rhBMP-2 or control SCPC.19 In the present study, we used the RPT technique to construct a 3-dimensional (3D) prototype of a segment of a rabbit ulnar bone using SCPC powder. The 3D porous SCPC implant was loaded with rhBMP-2 and implanted in a load-bearing segmental bone defect in a rabbit ulna. The histology and biomechanical properties of the regenerated bone were evaluated and correlated to material characteristics and rhBMP-2 release.

Materials and Methods SYNTHESIS OF SCPC IMPLANT USING RPT

SCPC powder containing 32.9 Na2O, 32.9 SiO2, 22.8 CaO, and 11.4 P2O5 in mol percentages was prepared as described previously.19 An exact replica of the ulnar bone segment was synthesized using the rapid prototyping technique and SCPC powder. Figure 1 is a summary of the experimental design that shows various implant processing steps using RPT. The rabbits (n ⫽ 9) were anesthetized, and the right forearm was CT (computed tomography) scanned (step 1, Fig 1). The CT data were converted into stereolithography (STL) format and imported into Amira computer-aided design (CAD) software (step 2, Fig 1). Using the CAD software, the ulna was isolated from the radius. Next, a 10-mm bone segment, which will be surgically removed, was selected in the image (20 mm proximal to the styloid process [step 3, Fig 1]). The removed bone segment in the CAD image was converted back into an STL format and then into a slice file that the Z-Corp machine (Z402 System; Z Corporation, Burlington, MA) used to replicate layer by layer using SCPC powder and a sugar-based binder (step 4, Fig 1). The implants processed by the Z-Corp were heat treated at 450°C for 18 hours to burn off the binder and then at 900°C for 3 hours to sinter the SCPC implants (step 5, Fig 1). Prior to surgery, the implants were sterilized in 70% ethanol (step 6, Fig 1), dried at 25°C, and then loaded with 10 ␮g rhBMP-2 (R&D Systems, Inc, Minneapolis, MN) (step 7, Fig 1) following the procedures described earlier.19

1497

EL-GHANNAM ET AL

FIGURE 1. A cartoon describing the various processing techniques and implantation of SCPC-rhBMP-2 tissue engineering scaffold (10-mm height ⫻ 5 to 7-mm diameter) in a segmental bone defect in a rabbit ulna. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

POROSITY MEASUREMENTS

The porosity percentage, pore size range, and specific surface area of SCPC discs processed using the Z-Corp 3D printer were determined with the use of mercury intrusion technique (Micromeretics Co, Norcros, GA). The minimum mercury filling pressure was kept at 0.2 pounds per square inch (psi). Nitrogen gas was used to generate pressure points up to 30 psi, after which the sample and the penetrometer were transferred to the high-pressure chamber where pressure points were generated up to 60,000 psi. SURFACE TRANSFORMATION

SCPC discs (10-mm diameter ⫻ 2-mm height) processed using the Z-Corp 3D printer were placed in 6-well plates and immersed in 6-mL simulated body fluid (SBF)24 at 37°C for 12 hours, 1, 3, or 5 days. At the end of each time period, the discs were rinsed and analyzed by Fourier Transform Infrared Spectroscopy (FTIR) (Thermo Nicolet Nexus 670 FT-IR, Madison, WI) in the diffuse reflectance mode. RELEASE KINETICS OF rhBMP-2

The SCPC samples (10-mm height ⫻ 5-mm width) (n ⫽ 5) processed using the Z-Corp 3D printer as well as porous HA discs (PreOsteon HA-200) were placed separately in a 24-well plate and loaded each with 30 ␮L containing 1 ␮g of reconstituted rhBMP-2. The rhBMP-2 was evenly distributed on the ceramic surface and allowed to dry for 1 hour. The dose selection of BMP was based on data in the literature that show

a stimulatory effect of similar doses on bone cell function.25,26 SCPC and HA samples loaded with and without rhBMP-2 were placed in a snap cap vial and immersed in 10 mL of phosphate-buffered saline (PBS; pH 7.4) containing 1% BSA at 37°C. After various time periods, 1, 2, 4, and 12 hours, and 1, 2, 4, 6, 8, 10, 12, and 14 days, 2 mL of the immersion solution was exchanged with 2 mL of fresh PBS. The concentration of BMP-2 released from the SCPC and HA implants into the immersing solution was quantified with a Quantikine BMP-2 ELISA kit from R&D Systems, Inc. SURGERY

Guidelines for the ethical and humane treatment of the care and use of laboratory animals following established practices of the University of Kentucky were followed for this study. New Zealand white, male, 3- to 4-month-old (3 to 3.5-kg) rabbits were anesthetized with an intravenous injection of 10 mg/kg xylazine followed by 50 mg/kg ketamine. Two rabbits died immediately after injection of anesthesia and before the surgery. Following aseptic procedures, a skin incision was made along the lateral side of the front right or left limb, superficial to the ulnar bone. After this, blunt dissection of the muscles and sharp dissection of the periosteum was completed to expose a 3 to 4-cm segment of the bone. After exposure, a 10-mm segmental defect was cut free from the center section of the ulnar bone using a surgical bur and copious irrigation with physiologic saline. The SCPC-rhBMP-2 implant was placed and secured with a

1498 1.0-mm titanium plating system. A bone plate of 12 holes (approximately 35 mm) and 8 screws was used. The wound was irrigated with sterile normal saline and closed in layers: the fascia was closed with 4-0 Vicryl, and the skin was closed with a 4-0 running subcuticular Monocryl. The contralateral side ulna was left intact in each animal. The rabbits received 0.05 mg/kg buprenorphine 30 minutes prior to the end of surgery and then every 8 to 12 hours as needed for routine postoperative pain control. The animals were allowed to function immediately on the operated limbs. RADIOGRAPHIC EVALUATION

CT scans were performed postoperatively at 4, 8, and 12 weeks and were performed using Siemens 10 Slice (Siemens AG; Munich, Germany) at 64 slices per rotation with an isotrope CT resolution of 0.4 mm. The high-resolution CT data were converted into STL data format, remapping the surface data into a series of triangles and then imported into a CAD file, which allowed for a noninvasive 3D view of the remodeled defect. HISTOLOGY AND HISTOMORPHOMETRY

Sixteen weeks postoperatively, the rabbits were euthanized by an overdose of sodium pentobarbital. The grafted and the contralateral unoperated ulnas were harvested from the rabbits, fixed in 10% buffered formalin and embedded in polyethlymethacrylate (PMMA). Nondecalcified sections (40 ␮m) were ground and prepared using an Exakt machine (Exakt Technologies, Inc, Oklahoma City, OK). The sections were made along the long axis of the ulna and were stained with Toluidine Blue and examined using an optical microscope. Using the Bioquant Image Analysis Software (R&M Biometrics, Nashville, TN), 10 random regions of interest were selected, and within these areas, total tissue area, total bone area, bone perimeter (BS), osteoblast perimeter (ObS), ObS/BS, and number of osteoblast per unit surface (N.Ob/BS) were measured. TORSIONAL TESTING

For the grafted limbs, due to partial fibrous fusion between the radius and ulna, both bones were used for mechanical testing 12 weeks postoperatively. To eliminate the contribution of the radius to the mechanical strength, the values of maximum torsional load to failure were normalized to unit cross-sectional area and compared with that of the contralateral side. The fixation devices on those ulnae were carefully removed without applying excessive mechanical loads during the harvesting process. Contralateral ulna of each rabbit was used as a control. Each bone sample was immediately wrapped in gauze moistened

BONE ENGINEERING OF THE RABBIT ULNA

with Ringer’s solution, sealed in a labeled storage bag, and then fresh frozen at ⫺20°C until testing. The fixation plates in 3 ulnae were found broken; therefore, these samples were excluded from the mechanical test. Prior to testing, all samples were defrosted at 2°C for 24 hours and trimmed with a low-speed diamond saw. Both ends of the bone were inserted into a custom-made alignment fixture, and 1 cm of each end of the ulna was potted in JB Weld for 24 hours. After hardening, the potted specimens were mounted in a custom-built torsional testing fixture that was secured to a computer-controlled, biaxial servohydraulic material testing system (Instron 8521, Canton, MA). With the proximal end held fixed, the distal end was externally rotated at a constant rate of 90° min⫺1 until failure. Torque versus angular deformation curves were obtained by using the accompanying Instron System 8800 Controller. Maximum torsional load to failure and angle at failure were measured. After the specimens had catastrophically failed, the cross-sectional areas of the ulnae were obtained by sectioning with a diamond wafer blade subjected to continuous irrigation. These cross-sections were photographed, and the cross-sectional areas were calculated by using image analysis freeware (NIH Image 1.59, Bethesda, MD). The calculated cross-sectional area data along with the gauge length and other measured torsional test data were then used to calculate the values of maximum torsional load to failure normalized to unit cross-sectional area and torsional angle at failure for grafted and control ungrafted ulna bone. STATISTICAL ANALYSIS

Representative values of parameters used in this study were shown as mean ⫾ standard deviation. Each parameter was statistically analyzed by a t test with 2 samples assuming unequal variances. A P value of less than .05 was considered to be statistically significant.

Results POROSITY MEASUREMENTS

Mercury porosimetry showed that the SCPC sample (average weight 0.4 gm) has a surface area of 2.615 m2/g and pore size distribution range of 17.5% (25360 ␮m), 60% (10-25 ␮m), 20% (1-10 ␮m), and 2.5% (1 ␮m-4 nm). SURFACE ANALYSIS

FTIR analysis showed that after 12 hours immersion in SBF, the surface of the SCPC discs developed an HA layer similar to the mineral phase of bone (Fig 2).

1499

EL-GHANNAM ET AL

% Reflectance

11 10 9 8 7 6 5 4 3 2 1 0 1200

Si-O P-O

Before Immersion 12 Hours

P-O

Si-O

3 2.5

P-OH

Si-O...Na

P-O

3.5

2

Si-O

1.5

C-O

1 0.5

P-O

0 1100

1000

900

800

700

600

500

400

-1

Wavenumber (cm )

FIGURE 2. Fourier transform infrared spectroscopy spectra of silicacalcium phosphate nanocomposite discs show the development of a carbonate HA surface layer after 12 hours of immersion in simulated body fluid. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

RELEASE PROFILE rhBMP-2

Figure 3 shows the mean cumulative concentration of rhBMP-2 released into PBS from SCPC-rhBMP-2 and HA-rhBMP-2 hybrids as a function of immersion time. During the first 4 days of immersion, the rhBMP-2 released from SCPC-rhBMP-2 hybrid into the solution was 27 pg/mL/h, which was significantly higher than that released from HA-rhBMP-2 (12 pg/mL/h) at the same time period (n ⫽ 5, P ⬍ .04). The significant increase in the concentration of rhBMP-2 released from SCPC compared with that released from HA continued for 14 days. The sustained release profile for SCPC is due to the domination of the precipitation reaction that forms a heavy layer of HA on the material surface. PBS incubated with control SCPC and HA

Active rhBMP-2 released (pg/ml)

7000

SCPC HA

6000 5000 4000 3000 2000 1000

FIGURE 4. CT scan of the rabbit ulna grafted with the SCPC-rhBMP-2 hybrid 1 month postoperatively. The dashed lines in the figure refer to the initial edges of the 10-mm defect. The arrows point to the plate that was used to provide initial support. Significant bone healing and full integration of the regenerated bone with the host bone can be observed. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

discs without rhBMP-2 showed minimal (undetectable) absorbance. RADIOGRAPHY

Clinically, the rabbits were showing normal leg loading. CT scans showed that the defect had been almost fully replaced by newly formed bone 4 weeks postoperative (Fig 4). The edges of the initial segmental bone defect totally disappeared, indicating complete integration between the newly formed bone and the host bone. CT scans 8 and 12 weeks postoperatively showed continuous improvement of the continuity of the newly formed bone with the host bone structure. Reconstructed 3D images showed the development of new bone, which grew in perfect continuity with the surrounding host bone and spread homogeneously inside the whole defect volume. HISTOLOGY AND MECHANICAL STRENGTH

0 0.04 0.08 0.17

0.5

1

2

4

6

8

10

12

14

Time (days)

FIGURE 3. The mean cumulative concentration of rhBMP-2 released from silica-calcium phosphate nanocomposite (SCPC) and HA-200 samples as a function of immersion duration. The rate of rhBMP-2 release was dependent on dissolution-precipitation reactions at the material-solution interface. A first-order release is observed in the first 4 days, which was followed by a sustained release profile in the period of 4 to 8 days, probably due to the domination of the precipitation reaction. In the period of 8 to 14 days, a first-order release profile with an average release of 19 pg/mL/h was observed. The concentration of rhBMP-2 released from SCPC was significantly higher than that released from HA after 14 days of immersion. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

Preliminary results of histological analysis (Fig 5) showed bone regeneration with high cellular activity throughout the defect. The arrows point to active osteoblasts at the edge of the newly formed bone. Apparently, new bone formation started at the surface of the SCPC material and grew as the graft material resorbed. Histomorphometric analysis of 10 random fields of the grafted defect of 1 rabbit was analyzed and showed 19.1% graft material and 32.4% mature bone. The remaining 48.5% (of the region of interest) contains primarily immature woven bone and bone marrow. The bone perimeter (BS) and osteoblast pe-

1500

FIGURE 5. Histological analysis of the center of the bone defect grafted with SCPC-rhBMP-2 after 16 weeks of implantation. The SCPC-rhBMP-2 implant had been replaced by regenerated mature bone. The black arrows point to blood vessels containing red blood cells indicating high vascularization of the regenerated bone tissue. In addition, osteocytes (red arrows) can be seen in the mature mineralized tissue. Remnants of the SCPC material (white arrows) are incorporated in the newly formed bone, and osteoblasts (yellow arrows) can be seen actively forming new bone. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

rimeter (ObS) were 10.577484 mm and 3.024876 mm, respectively. The ratio ObS/BS was 28.6 and the N.Ob/BS was 18.8 cell/mm2. Areas with mature bone appeared highly vascularized, as evident by the presence of numerous blood vessels filled with red blood cells. Mechanical testing showed that the maximum torsional load to failure was significantly higher (n ⫽ 3, P ⬍ .04) for the ulnae grafted with SCPC-rhBMP-2 hybrid than the contralateral intact ulna (Fig 6). In comparing the angle at failure, there was not a significant difference between the grafted and ungrafted ulnae. The angle to failure for ungrafted defects was 0.32 ⫾ 0.10 and that for grafted defects was 0.45 ⫾ 0.22.

BONE ENGINEERING OF THE RABBIT ULNA

At 4 months, the defect previously grafted with SCPC-rhBMP-2 had been almost completely refilled with bone, with only a small amount of graft material remaining. The new bone formed could be distinguished from the normal host bone by apparent increased vascularity. The regenerated bone mass was formed in the exact size and shape of the natural bone. There was no histological evidence of ectopic bone formation beyond the implanted sites in all defects treated with SCPC-rhBMP-2, implying that the implant properly regulated the bone formation in the defect area. The recovery of the bone anatomy is enhanced by the material bioactivity and perfect implant geometry produced by the RPT. Other BMP scaffolds processed using traditional techniques did not allow for enhanced integration and regain of the bone anatomy.27 In our study, using the RPT allowed for the synthesis of a 3D porous SCPC bone replica. Upon implantation, the intimate contact between the host tissue and the implant allowed for guided bone regeneration. On the cellular level, previous studies have shown that the SCPC provides guided cell adhesion and tissue growth in vitro.19 Clinical use of rhBMP-2 has been hampered by a lack of suitable delivery systems. Such systems should be capable of maintaining the protein in situ for sufficient time to interact with target cells, release the protein at effective concentrations during bone formation, cause no unnecessary tissue distress, and be resorbed.28,29 Results of our study showed that SCPCrhBMP-2 provided favorable release kinetics of bioactive rhBMP-2 superior to that provided by HA ceramic. The enhanced rhBMP-2 release is attributed to the facilitated solubility of SCPC.18 It is also possible that the high surface area of SCPC due to the presence of nanopores would facilitate the release of rhBMP-2. Preliminary in vivo results indicated that the enhancement effect of SCPC-rhBMP-2 on bone regeneration was superior to that reported for other scaffolds. Seto

Discussion

30 Torsional Load (N-m)

Grafted

The SCPC-rhBMP-2 hybrid enhanced bone regeneration in a load-bearing segmental defect in the rabbit ulna. The regenerated bone acquired morphology and mechanical strength typical for natural bone. The enhanced bone formation correlates well with the surface bioactivity and the effective release profile of rhBMP-2. CT scans provided radiographic evidence that the resorbable SCPC-rhBMP-2 scaffold had induced bone regeneration in the initially 10-mm-long defect. The rapid bone healing shows the high bioactivity of the SCPC and the osteoinductive ability of the released rhBMP-2. These results indicate that the SCPC-rhBMP-2 hybrid can serve as an alternative to autologous bone grafting.

Control

25 20

20.07

15 10

6.61

5 0

FIGURE 6. Torsional testing of the ulna grafted with the SCPCrhBMP-2 hybrid and the ungrafted contralateral ulna showed that the maximum torsional load to failure was significantly greater (P ⬍ .04) for the grafted ulna. However, there was not a significant difference between the angles at failure for the grafted ulna as compared with the ungrafted ulna. El-Ghannam et al. Bone Engineering of the Rabbit Ulna. J Oral Maxillofac Surg 2007.

1501

EL-GHANNAM ET AL

et al grafted rhBMP-2/poly(lactic-co-glycolic acid) (PLGA) in a segmental mandibular bone defect and observed only a small amount of woven bone formation after 16 weeks of implantation.30 The limited bone formation could be attributed to the inadequate ability of PLGA to deliver BMP. To overcome this problem, Kokubo et al used a PLGA-coated gelatin sponge as a carrier for various doses of rhBMP-2 to treat segmental ulnar defects in rabbits. At BMP concentration of 0.1 mg/cm3 (a concentration that is almost 15 times higher than what was used in our study), 5 of 8 ulnae could not achieve radiographic union after 16 weeks of implantation.27 The superior bone-healing effect of SCPC loaded with a significantly lower (7 ␮g/cm3) dose of rhBMP-2 indicates that SCPC provided controlled BMP release, enabling the amount of BMPs required for bone repair to be reduced by ensuring effective BMP retention in reactive cells. This is in agreement with data in the literature that indicated that the nanopores of SCPC serve as protective pockets for biological molecules.19 The release of rhBMP-2 from the scaffold into tissue fluids is controlled by the SCPC dissolution as well as the back precipitation of the HA layer. The histological evaluation of bone defects grafted with SCPC-rhBMP-2 hybrids showed no signs of an adverse inflammatory reaction. There were no obvious abnormalities in the surrounding soft tissue, and muscle tissue was completely regenerated and attached normally to regenerated bone at 12 and 16 weeks. These findings suggest that implantation of SCPC-rhBMP-2 did not result in local irritation to the surrounding soft tissue or adverse effects on soft tissue healing. Moreover, histological analyses, 12 weeks after implantation, showed concentric lamellar structures (osteon) in the bony tissue induced by SCPC-rhBMP-2. However, islands of woven bone in addition to activated osteoblasts were observed in regenerated bony tissue at this time, suggesting that the mechanism of bone regeneration is similar to that of the natural bone remodeling process. Of prime importance is the observation that the regenerated bone was highly vascularized (Fig 5), which indicates that the SCPC scaffold is characterized by interconnected pathways to allow for cell and blood vessel invasion. In conjunction with the abundant bone regeneration, a major resorption of the graft material was observed, indicating that the rate of bone regeneration matches the rate of SCPC resorption. SCPC remnants were seen incorporated in the newly formed bone, with direct bone apposition onto its surface. As the SCPC resorbs, it leaves a space for attached bone cells to deposit new bone. To regain complete functionality of bone, the regenerated bone needs to have the same mechanical

strength as natural bone. Results of our study showed the restoration of bone strength in a significantly short time compared with data in the literature.27 Segmental bone defects grafted with PLGA-BMP showed restoration of maximum torque and angle at failure to the intact levels after 32 weeks of implantation.27 Another study, using resorbable calciumphosphate particles in a segmental defect, found that after 12 weeks, the bone formed in the defect had relatively poor mechanical strength due to the resorption of the particles taking place before complete bone remodeling could occur.31 The restoration of maximum torque and angle at failure of the regenerated bone induced by SCPC-rhBMP-2 was seen after 12 weeks of implantation. The rapid recovery of the mechanical strength correlates well with the mature bone morphology observed throughout the entire thickness of the defect. Bone marrow osteoprogenitor cells infiltrated the interconnected pores of SCPC, and differentiated and deposited bone. The dissolution products of SCPC and the released BMP-2 created a favorable environment inside the implantation bed, which facilitated ossification and maturation of the regenerated bone. The present preliminary study shows the proof of principles that porous, resorbable, bioactive SCPCrhBMP-2 tissue engineering hybrid can serve as a substitute for autologous bone implant in load-bearing applications. In vivo testing of the SCPC-rhBMP-2, prepared by a rapid prototyping technique, enhanced bone regeneration in a segmental, load-bearing bone defect in the rabbit ulna. Torsional testing of the ulna showed the restoration of complete functionality after just 12 weeks of implantation. Histological evaluation showed regeneration of mature vascularized bone with differentiated osteoblasts and osteocytes. The enhancement of complete bone regeneration after 16 weeks was attributed to the high bioactivity, resorbability, and interconnected porosity of the material, as well as to rhBMP-2 release. Acknowledgment The authors would like to thank Dr Leslie Sherwood, Dr Kenneth Dickey, and the staff at the DLAR; University of Kentucky, Bill Gregory and the staff at UK Center for Manufacturing, and University of Alabama at Birmingham, Center for Metabolic Bone Disease—Histomorphometry and Molecular Analysis Core Laboratory.

References 1. Koempel JA, Patt BS, O’Grady K: The effect of recombinant human bone morphogenetic protein-2 on the integration of porous hydroxyapatite implants with bone. J Biomed Mater Res 41:359, 1998 2. Takahashi T, Tominaga T, Watabe N: Use of porous hydroxyapatite graft containing recombinant human bone morphogenetic protein-2 for cervical fusion in a caprine model. J Neurosurg 90:224, 1999

1502 3. Tsuruga E, Takita H, Itoh H: Pore size of porous hydroxyapatite as the cell-substratum controls BMP-induced osteogenesis. J Biochem 121:317, 1997 4. Asahina I, Watanabe M, Sakurai N: Repair of bone defect in primate mandible using a bone morphogenetic protein (BMP)– hydroxyapatite-collagen composite. J Med Dent Sci 44:63, 1997 5. Ono I, Gunji H, Kaneko F: Efficacy of hydroxyapatite ceramics as a carrier of recombinant human bone morphogenetic protein. J Craniofac Surg 6:238, 1995 6. Ripamonti U, Ma SS, Heever B, et al: Osteogenin, a bone morphogenic protein, absorbed on porous hydroxyapatite substrata, induces rapid bone differentiation in calvarial defects of adult primates. Plast Reconstr Surg 90:382, 1992 7. Urist MR, Lietze A, Dawson E: Beta-tricalcium phosphate delivery system for bone morphogenetic protein. Clin Orthop 157: 259, 1981 8. Gao TJ, Lindholm TS, Ragni P: Enhanced healing of segmental tibial defects in sheep by a composite bone substitute composed of tricalcium phosphate cylinder, bone morphogenetic protein, and type IV collagen. J Biomed Mater Res 32:505, 1996 9. Doll BA, Towle HJ, Hollinger JO: The osteogenic potential of two composite graft systems using osteogenin. J Periodontol 61:745, 1990 10. Horisaka Y, Okamoto Y, Matsumoto N: Subperiosteal implantation of bone morphogenetic protein adsorbed to hydroxyapatite. Clin Orthop 268:303, 1991 11. Kimura Y: Biomedical Applications of Polymeric Materials. Boca Raton, FL, CRC Press, 1993, p 163 12. Taylor MS, Daniels AU, Andriano KP, et al: Six bioabsorbable polymers: In vitro acute toxicity of accumulated degradation products. J Appl Biomater 5:151, 1994 13. Gugala Z, Gogolewski S: Regeneration of segmental diaphyseal defect in sheep tibia using resorbable polymeric memberanes: A preliminary study. J Orthop Trauma 11:551, 1997 14. El-Ghannam A: Bone reconstruction: From bioceramics to tissue engineering (review article). Expert Rev Med Devices 2:87, 2005 15. Hollister S, Taboas J, Schek R: Bone Tissue Engineering. Boca Raton, FL, CRC Press, 2004, p 167 16. Hutmacher DW: Scaffold design and fabrication technologies for engineering tissues—State of the art and future perspectives. J Biomater Sci Polym Ed 12:107, 2001 17. Yang S, Leong KF, Du Z, et al: The design of scaffolds for use in tissue engineering. II. Rapid prototyping techniques. Tissue Eng 8:1, 2002 18. El-Ghannam A: Advanced bioceramic composite for bone tissue engineering: Design principles and structure-bioactivity relationship. J Biomed Mater Res 69:490, 2004

BONE ENGINEERING OF THE RABBIT ULNA 19. El-Ghannam A, Ning CQ, Mehta J: Cyclosilicate nanocomposite: A novel resorbable bioactive tissue engineering scaffold for BMP and bone-marrow cell delivery. J Biomed Mater Res 71: 377, 2004 20. El-Ghannam A, Karima A, Omran M: Nanoporous delivery system to treat osteomyelitis and regenerate bone: Gentamicin release kinetics and bactericidal effect. J Biomed Mater Res B Appl Biomater 73:277, 2005 21. El-Ghannam A, Ning CQ: Effect of bioactive ceramic dissolution on the mechanism of bone mineralization and guided tissue growth in vitro. J Biomed Mater Res A 76:386, 2005 22. Ning CQ, Greish Y, El-Ghannam A: Crystallization behavior of silica-calcium phosphate biocomposites: XRD and FTIR studies. J Mater Sci Mater Med 15:1227, 2004 23. Ning CQ, Mehta J, El-Ghannam A: Effects of silica on the bioactivity of calcium phosphate composite in vitro. J Mater Sci Mater Med 16:1, 2005 24. Kokubo T, Kushitani H, Sakka S: Solutions able to reproduce in vivo surface-structure changes in bioactive glass-ceramic A-W. J Biomed Mater Res 24:721, 1990 25. Nicoll SB, Radin S, Santos EM: In vitro release kinetics of biologically active transforming growth factor-␤1 from a novel porous glass carrier. Biomaterials 18:853, 1997 26. Santos EM, Radin S, Shenker BJ: Si-Ca-P xerogels and bone morphogenetic protein act synergetically on rat stromal marrow cell differentiation in vitro. J Biomed Mater Res 41:87, 1998 27. Kokubo S, Mochizuki M, Fukushima S: Long-term stability of bone tissues induced by an osteoinductive biomaterial, recombinant human bone morphogenetic protein-2 and a biodegradable carrier. Biomaterials 25:1795, 2004 28. Kaito T, Myoui A, Takaoka K: Potentiation of the activity of bone morphogenetic protein-2 in bone regeneration by a PLAPEG/hydroxyapatite composite. Biomaterials 26:73, 2005 29. Lucas PA, Syftestad GT, Goldberg VM, et al: Ectopic induction of cartilage and bone by water-soluble proteins from bovine bone using a collagenous delivery vehicle. J Biomed Mater Res 23:23, 1989 30. Seto I, Asahina I, Oda M, et al: Reconstruction of the primate mandible with a combination graft of recombinant human bone morphogenetic protein-2 and bone marrow. Int J Oral Maxillofac Surg 59:53, 2001 31. Bloemers FW, Blokhuis TJ, Patka P: Autologous bone versus calcium-phosphate ceramics in treatment of experimental bone defects. J Biomed Mater Res B Appl Biomater 66:526, 2003