Enhanced Bone Morphogenetic Protein-2-Induced Ectopic and Orthotopic Bone Formation by Intermittent Parathyroid Hormone (1–34) Administration

Experimental design

A total of 54 rats were used for the experiment. The rats were divided into six groups of nine animals for each treatment (Table 1). The three different surgical treatments consisted of (1) no implant, (2) implantation of empty composite, and (3) implantation of BMP-2-loaded composite. Each of the surgical treatments was studied alone or in combination with PTH injections administered in the subcutis of the neck. The treatments were studied in both ectopic (subcutaneous) and orthotopic (critical-sized femoral defect) sites. Whereas the ectopic site shows the osteoinductive potential of the construct without interference of osteoconduction or periosteal bone formation as disturbing mechanisms in the BMP-2-induced bone formation, the femoral defect represents a clinically applicable site. The empty defect with no implant was only studied orthotopically. The implants were removed after 8 weeks of implantation for evaluation of bone formation by microcomputed tomography (μCT), dual energy X-ray absorptiometry (DEXA), and histology.

Materials

PLGA (Medisorb^®; Lakeshore Biomaterials) with a 50:50 lactic-to-glycolic acid ratio and a weight-average molecular weight (M_w) of 23 kDa was used for microsphere preparation. PPF with a M_w of 3100 and a polydispersity index of 2.7 was synthesized by using a two-step reaction process as previously described. ¹³ Recombinant human BMP-2 (Medtronic Sofamor Danek) was concentrated by centrifuging at 5000 g in a Centricon-10 filter unit (Amicon) and reconstituted to the appropriate concentrations in an aqueous buffer consisting of 5 mM glutamate, 5 NaCl, 0.5% sucrose, 2.5% glycine, and 0.01% polysorbate 80; pH 4.5 (all Sigma-Aldrich). The receptor binding 34-amino acid N-terminal fragment of the PTH molecule [PTH (1–34); Bachem] was diluted in an aqueous buffer for injection consisting of 0.9% NaCl, 2% heat-inactivated serum (obtained from Sprague-Dawley rats), and 1 mM HCl. N-vinylpyrrolidinone (NVP; Acros), bis(2,4,6-trimethylbenzoyl) phenylphosphine oxide (BAPO; Ciba Specialty Chemicals), gelatin (300 Bloom number, derived from acid-cured porcine skin; Sigma-Aldrich), glutaraldehyde (Sigma-Aldrich), and glycine (Sigma-Aldrich) were used as received.

BMP-2 delivery vehicle

The implant consisted of PLGA microspheres incorporated into a solid PPF rod that was surrounded by a cylindrical gelatin hydrogel. The fabrication procedure of the PLGA microsphere/PPF/gelatin composites was described previously.^11,12 The BMP-2-loaded or unloaded PLGA microspheres were fabricated using a water-in-oil-in-water (W1-O-W2) double-emulsion solvent extraction technique. ¹⁴ On the basis of an starting amount of 1.3 μg BMP-2/mg PLGA and a previously achieved entrapment efficiency of 85%, the loading of the microspheres was estimated at 1.1 μg BMP-2/mg PLGA. ¹²

The cylindrical microsphere/PPF scaffolds were then created by photocrosslinking PPF with NVP under UV light using BAPO as a photoinitiator. ¹⁵ PLGA microspheres (55 wt%) were combined with a PPF/NVP paste (45 wt%) and injected into a glass mold to create a solid cylindrical composite with a diameter of 1.6 mm and a length of 5.5 mm.^11,12 On the basis of an average weight of 10.3 ± 0.8 mg/cylinder, BMP-2 loading in the microsphere/PPF rods was estimated at 6.5 ± 0.5 μg/cylinder. The gelatin hydrogels were fabricated by chemically crosslinking a 10 wt% aqueous gelatin solution with 0.004 wt% glutaraldehyde in tubular molds (1.6 mm inner diameter and 3.5 mm outer diameter) for a period of 1 h.^11,15–17 The crosslinked hydrogels were placed in a 100 mM glycine solution for 1 h to block the residual aldehyde groups. The microsphere/PPF rods and gelatin hydrogels were sterilized separately in 70% alcohol, washed in phosphate-buffered saline, and stored at −20°C. The microsphere/PPF rods were inserted into the tubular gelatin hydrogels just before implantation.

Animals and procedures

The bone-forming capacity of the scaffolds was tested in both ectopic and orthotopic models in 12-week-old Harlan Sprague Dawley rats (weight 324 ± 12 g). The experiments were performed according to established protocols approved by the Institutional Animal Care and Use Committee at the Mayo Clinic. The rats were anesthetized with an intramuscular injection of a ketamine/xylazine mixture (45/10 mg/kg) and the surgical sites were shaved and disinfected with a 30% betadine solution. The right femur was exposed through a lateral approach and was circumferentially stripped from soft tissue and periosteum. A predrilled, high-density polyethylene plate (23 × 4 × 4 mm, custom-made) was fixed along the anterior cortex of the femur using four threaded Kirschner wires (1 mm diameter; Zimmer). After plate fixation, a 5-mm mid-diaphyseal, full-thickness defect was created using a surgical burr and irrigated with physiologic saline to remove bone debris and bone marrow. After insertion of the implant in the defect, the wound was closed in layers using resorbable sutures (Vicryl 1/0; Ethicon, Inc.) and nonresorbable nylon skin sutures. In addition to the orthotopic implant, each rat received an ectopic implant in a subcutaneous pocket at the lower left limb.

Postoperative analgesia was provided by intramuscular injections of buprenorphine (0.1 mg/kg) for the duration of 72 h and acetaminophen (880 mg/L added to the water bottle) for the duration of 1 week.

PTH (1–34) treatment was started on the third postoperative day. PTH injections were administered in the subcutis of the neck at a dosage of 10 μg/kg/day for 5 days a week till the end of the 8 week follow-up period. In addition to PTH injections, the rats received fluorochrome markers calceine green (10 mg/kg; Sigma-Aldrich) and alizarin red (10 mg/kg; Sigma-Aldrich) after 4 and 6 weeks, respectively. After 8 weeks, the animals were sacrificed by an overdose of pentobarbital and the vessels were perfused with Microfil^® (Flow Tech) to study vessel formation at the implantation site using previously described methods. ¹¹ Following a midline laparatomy, the aorta and vena cava were canulated to clear the lower extremities from blood with heparinized saline. The vessels were subsequently infused until the toes were clearly colored blue by the Microfil compound. Since there were no significant differences between the treatment groups, these data are not shown. After polymerization of the compound, the implants were removed and fixed in phosphate-buffered saline containing 1.5% glutaraldehyde. The vertebral column and humerus were also harvested to analyze the effects of PTH treatment on overall bone density. Before μCT and DEXA scans, the two K-wires closest to the defect were removed from the femur.

BMD measurements

BMD measurements were performed by DEXA using a PIXImus densitometer (software version 1.44.005; Lunar Corporation). Before the measurements, the PIXImus densitometer was calibrated using a hydroxyapatite phantom that was provided by the manufacturer. The excised implants, humerus, and vertebral column were scanned by the machine and the regions of interest were identified for the analysis of the BMD.

μCT analysis

The implants were scanned on a custom-build μCT system at 0.49° angular increments (providing 721 views around 360°) using 18 keV. ¹⁸ The projections were reconstructed using a modified Feldkamp cone beam tomographic reconstruction algorithm into a three-dimensional image consisting of 20 μm cubic voxels with a radiopacity represented by a 16-bit grayscale value. The bone formation in the implants was analyzed using image analysis software Analyze 6.0 (Biomedical Imaging Resource; Mayo Clinic, Rochester, MN). Due to the difficulty to distinguish the newly formed bone from the original femur, a 6-mm section of the femur was used for bone quantifications. The reconstructions and volume quantification of the ectopic and orthotopic bone were obtained using standardized thresholds.

Histology

After μCT and BMD measurements, the subcutaneous implants and femurs were dehydrated in a graded series of alcohol and embedded in, respectively, glycol-methylmethacrylate and methylmethacrylate for histological evaluation. The subcutaneous samples were longitudinally sectioned using a rotary microtome (Reichert-Jung Supercut 2050 microtome; Leica Microsystems) and the femurs using a sawing microtome (Leica SP1600; Leica Microsystems). The sections were stained with trichrome Goldner, hematoxylin/eosin, and/or methylene blue/basic fuchsine, and evaluated for general tissue response and bone formation. Additional unstained sections were evaluated for incorporation of fluorochrome labels using a fluorescence microscope (E600 Nikon) with a double-filter block (dichroic mirror 505 and 590 nm).

Statistical analysis

All data are given as means ± standard deviations for n = 9. Independent t-tests were performed to analyze the BMD and bone volumes of the ectopic implants. The BMD and bone volumes of the orthotopic implants were statistically compared using one-way analysis of variance with Bonferroni's post hoc tests. The effect of PTH treatment on overall BMD was analyzed using two-way analysis of variance. All tests were performed by SPSS (version 13.0; SPSS, Inc.), and the level of significance was set at p < 0.05.

Animal

Six animals (one in empty defect group, two in empty defect/PTH group, one in empty composite group, and two in the empty composite/PTH group) developed deep infections of their femoral defects during the study. Their BMD and bone volume measurements were excluded from further analysis. The other animals remained in good health and did not show any signs of complications.

Bone mineral density

After 8 weeks of implantation, all implants were easily identified and retrieved for further analysis. At the ectopic site, bone formation was only seen in BMP-2-loaded scaffolds. The ectopic BMP-2 implants in the PTH-treated animals had a significantly (p < 0.05) higher BMD than the BMP-2 implants in the nontreated animals (Fig. 1). At the orthotopic site, the BMP-2-loaded scaffolds had a significantly higher BMD than the unfilled defects or defects filled with empty scaffolds (p < 0.02). Further, BMD of the BMP-2-containing scaffolds in these femoral defects was significantly higher in the animals treated with PTH injections (p < 0.04). PTH administration had no effect on the BMD of empty defects or defects treated with empty scaffolds. The BMD of the right humerus and vertebra L1 of the PTH-treated rats was significantly higher than that of the nontreated rats (p < 0.05).

OPEN IN VIEWER

FIG. 1.

Bone mineral density of the ectopic implants, orthotopic implants, vertebra L1, and humerus after 8 weeks of follow-up. The symbols indicate significant differences relative to (*) ectopic bone morphogenetic proteins (BMP)-2-loaded implants, (#) unfilled and empty scaffold filled orthotopic defects, (+) all other groups, or (x) nontreated vertebrae or humeri (p < 0.05). PTH, parathyroid hormone.

Bone volume

The effects of BMP-2 released from the scaffold and of PTH treatment on the ectopic and orthotopic volumes of the newly formed bone were studied by μCT imaging at 20 μm resolution. No radiological signs of bone formation were seen in the ectopically implanted empty scaffolds. Ectopic bone formation was seen in the μCT reconstructions of BMP-2-containing implants. Part of the ectopic implant surface was covered by a compact shell of bone, and trabecular bone was seen inside the pores of the scaffold (Fig. 2A). Quantification of the bone volume showed that PTH had a significant effect on ectopic bone formation in these scaffolds (p = 0.02, Fig. 2B).

OPEN IN VIEWER

FIG. 2.

Three-dimensional microcomputed tomography reconstructions (A) and bone volumes (B) of subcutaneously formed bone after 8 weeks of implantation in rats. The asterisk (*) indicates significant difference relative to BMP-2-loaded implants (p = 0.02).

All orthotopic sites showed some newly formed bone around the defect edges (Fig. 3A). Analysis of connections between the proximal and distal part of the femur showed that none of the unfilled defects contained an osseous connection between the two femur parts. Trabecular connections between the two parts were observed in 2/8 and 2/7 of the untreated or PTH-treated defects filled with empty scaffolds, respectively, whereas almost full cortical regeneration was seen in all of the defects containing the BMP-2-loaded implants. The defects filled with BMP-2-loaded implants contained significantly more bone than the untreated unfilled defects and defects filled with empty scaffolds (p = 0.04). The bone volume in the defects of animals treated with both BMP-2 scaffolds and PTH injections was significantly higher than all other groups (p = 0.02). No significant differences were found between the untreated and PTH-treated unfilled defects or defects filled with empty scaffolds.

OPEN IN VIEWER

FIG. 3.

Three-dimensional microcomputed tomography reconstructions (A) and bone volumes (B) of newly formed bone in the 5-mm femoral defects after 8 weeks of follow-up. The symbols indicate significant differences relative to (#) unfilled defects, defects filled with empty scaffolds, or defects filled with empty scaffolds treated with parathyroid hormone (PTH) injections and (+) all defects (p < 0.05).

Histology

Microscopic evaluation of transverse sections showed that the gelatin hydrogel around the microsphere/scaffold composite was completely resorbed and the PLGA microspheres were degrading to create a porous PPF network. In the ectopic site, the gelatin hydrogel was replaced by a well-vascularized fibrous capsule. There was a mild foreign body reaction to the microsphere/PPF rod as indicated by some inflammatory cells inside the pores of the scaffold. The interconnected pores of the ectopically implanted empty scaffolds were filled with connective tissue (Fig. 4A). Ectopic bone formation and osteoid depositions were seen on the surface and inside the pores of BMP-2-containing implants (Fig. 4B–D). Bone formation was more extensive in the ectopic BMP-2 implants of rats treated with PTH compared to the nontreated animals. Fluorochrome analysis showed the presence of calcein green along the surface and alizarin red at both the surface and in the pores of the implant, indicating that mineralization started at the periphery and progressed toward the center.

OPEN IN VIEWER

FIG. 4.

Histological sections of subcutaneously implanted empty (A) or BMP-2 (B–D) scaffolds in rats that were treated with (A, C, D) or without (B) PTH. Hematoxylin/eosin-stained section (A) showing the interconnected porous polymer network and fibrous tissue (f) in the scaffold. Goldner-stained sections (B–D) showing bone formation (b) and osteoid deposition (arrows) at the surface and in the pores of BMP-2 scaffolds (s). Scale bars represent 500 μm (A, D), 200 μm (B), and 50 μm (C). Color images available online at www.liebertonline.com/ten.

In the orthotopic implant, new bone formation extending from the femur edges was seen in all implants (Fig. 5). The empty defects were filled with inter-positioning muscles, fibrous tissue, and a limited amount of bone (Fig. 5A). Implantation of empty scaffolds resulted in bone formation at the surface and in the scaffold pores at the defect edges, whereas the center was filled with connective tissue (Fig. 5B). In contrast to the empty defects and defects filled with empty scaffold, almost full cortical bridging was seen in the BMP-2 composite containing defects (Fig. 5C, D). A narrow area with dense connective tissue containing chondrocyte-like cells showing signs of endochondral ossification could be observed in the center of most of these defects (Fig. 5E). Overall, the rats treated with PTH showed broadening of the preexisting parts of the femur and newly formed cortices with a more compact structure of the bone (Fig. 5C, D). The calcein green and alizarin red were found throughout the newly formed cortical bone, indicating that mineralization was present in the entire cortical area before 4 weeks and continued over the rest of the follow-up period (Fig. 5F).

OPEN IN VIEWER

FIG. 5.

Histological sections of an empty femoral defect (A) and defects filled with empty (B) or BMP-2 scaffolds (C–F) in rats treated without (A, C) or with (B, D–F) PTH. Methylene blue/basic fuchsin-stained sections (A–E) showing interpositioning of muscles (m), fibrous tissue formation (f), and bone formation (b) extending from the original femur cortices (c) into the defect and the pores of the scaffold (s). The vascular network in the bone and fibrous tissue is observed by the blue Microfil^® agent (arrows). Endochondral ossification was seen in the center of the defects of BMP-2-containing implants (E). Fluorochrome analysis of unstained sections (F) showed calceine green and alizarin red deposition in newly formed bone. Scale bars represent 500 μm (A–D), 50 μm (D, E), and 500 μm (F). Color images available online at www.liebertonline.com/ten.