Platelet-Derived Growth Factor BB Treated Osteoprogenitors Inhibit Bone Regeneration

Various materials have been used to promote bone regeneration around dental implants, including autografts, allografts, xenografts, and alloplasts. Of these materials, autogenous bone grafts remain the gold standard for bone regeneration.1 Autografts contain living cells and osteoinductive factors. However, the scarce locations available for autogenous bone harvesting as well as donor site morbidity require the search for other alternatives. These include molecular and cellular materials that tissue engineer the bone.2^,3 Bone engineering involves the delivery of cells or biologic molecules such as growth factors to a defect site for tissue regeneration. Cells used in bone engineering are the osteoblasts, osteoprogenitor cells, periosteal cells, and stem cells. These cells may be seeded onto scaffolds to be transplanted into osseous defects. It is often necessary to use a scaffold for transplanting these cells to enhance the regeneration of the affected tissue or organ.4 

Vinyl styrene microbeads (VSM) are a new alloplastic material made of microspheres of approximately 500 μm in diameter. The microspheres are radiolucent and neutrally charged, but they pick up a charge when placed in a charged field.5 It was found that this alloplast is osteoconductive and has no toxic effect on the vital organs of rats.6 It has been shown that these microbeads are capable of delivering transforming growth factor-β (TGF-β) to rat calvarial critical-size defects.⁷ VSM are neither osteoinductive nor osteogenic per se; however, we hypothesize that the VSM can be used to deliver osteoblastic progenitors into bone defects. The aim of the present study is to evaluate the efficacy of this alloplast to carry and deliver osteoprogenitor cells to rat calvarial critical-size defects.

Platelet-derived growth factor BB (PDGF-BB) is the most potent mitogenic and chemotactic member of PDGFs.8 It stimulates DNA synthesis of rat calvarial cells and increases the production of osteopontin but decreases bone sialoprotein, osteocalcin, alkaline phosphatase activity, and mineralization in bone cells.9^,10 Nevertheless, the role of PDGF in bone regeneration is controversial. The pretreatment of osteoblastic cells with PDGF-BB before transplantation into osseous defects is investigated in the present study to further investigate the role of this growth factor in bone regeneration.

This research was approved by the committee of the Animal Laboratory Services at the Medical College of Georgia.

Calvarial fragments were collected from 18-day-old-embryos. Fragments were repeatedly digested using 2 mg/mL collagenase. Released cells were suspended in 4 mL of osteoblast basal medium (Canbrex BioScience, Walkersville, Md) with 10% fetal bovine serum, and 1% antibiotics (10 000 IU/mL penicillin and 10 000 μg/mL streptomycin). Cells were seeded at 10³ cell/mL concentration in 25 cm² cell culture flasks and kept in a humidified atmosphere of 5% CO₂ and 95% air. Cells from the second passage were used for both in vitro and animal (in vivo) experiments.

Vinyl styrene microbeads were fabricated into discs and fixed to the bottom of 48 tissue culture wells. Cells were seeded on top of 10 discs at a 5 × 10⁴/well cell density. Cells on 5 discs were incubated with 40 ng/mL PDGF-BB added to the culture medium, while the rest were incubated without this growth factor. After 3 days, discs were removed, without disturbing their surfaces, and processed for scanning electron microscopy. First, they were dehydrated with graded concentrations of ethanol (70%, 80%, 90%, 95%, and 100%) for 2 hours in each concentration. Discs were sputter coated with silver in a vacuum chamber and viewed with a scanning electron microscope for evaluation of the shape and topography of attached cells.

Twenty-five Long-Evans male rats of the same age were randomly divided into 5 groups of 5 animals each. After anesthetizing the animal, a midsagittal incision was made and a full-thickness flap reflected to expose the calvarial bone. An 8-mm diameter trephine was used to create a critical-size surgical defect in the midline of calvarium. A first Millipore membrane (Millipore Corporation, Bedford, Mass) was placed on top of the dura to separate the cranium from the defect. Defects were managed as follows. Group I (GPI) rats were used as negative controls with no grafting material used to fill the defects. Group II (GPII) rats were treated with approximately 8-mm diameter VSM discs (Hayes Inc, Panadora, Tex). Defects of group III (GPIII) were filled with 100-μL solution of PDGF at 40 ng/mL concentration. Group IV defects were reconstructed with VSM carrying osteoblasts (GPIV) at a concentration of 19 × 10⁴ cells/disc. Finally, group V was treated similar to GPIV except that the cells were pretreated with 40 ng recombinant human PDGF-BB for 3 days in culture before transplantation. Cells used for transplantation were from the second passage.

A second Millipore membrane was placed to cover the defect and isolate it from the overlying periosteum that was sutured to membrane peripheries. Skin was sutured with resorbable sutures and stapled with metallic staples for secure wound closure.

Animals were euthanized 16 weeks post surgery, at which time block sections of calvariae were removed and fixed in 10% formalin. Samples were sectioned midsagittally to yield 2 samples of approximately equal size. Half of each sample was demineralized and stained with hematoxylin and eosin while the other half was dehydrated, plastic embedded, sectioned, and stained with modified Masson (trichrome) stain. Qualitative analysis of the histology sections was evaluated by light microscopic examination at different magnifications. For histomorphometric (quantitative) analysis, an image analysis software (Bioquant Nova Prime, BIOQUANT Image Analysis Corporation, Nashville, Tenn) was used.

Data collection and statistical comparison of the regenerated bone volume (histomorphometry) were compared statistically via paired 2-sample t test. Significant differences between the groups was determined based on an alpha level of .05.

Scanning electron microscopic analysis confirmed the ability of the cultured osteoblast cells to attach to the vinyl styrene microbeads. Figure 1A shows osteoblasts incubated for 3 days on the VSM without PDFG-BB. In these samples, cells attained a semispherical topography and rounded outlines. They revealed multiple filopodia attaching the cells to the underlying substrate. Intercellular connections were maintained through cytoplasmic elongations as cells have extended filopodia.

In the PDGF-BB treated samples, cells appeared spreading on the microbeads surface with more flat topography with polygonal and rounded outline (Figure 1B).

Evaluation of the retrieved calvarial specimens revealed that the non-VSM groups (GPI and GPIII) have collapsed membranes reducing the space available for bone regeneration. In contrast, in the VSM groups (GPII, GPIV, and GPV) the microbeads provided a rigid support for the membranes. The newly formed bone was seen encasing some microbeads and infiltrating among them (Figure 2). At high magnification (×200), some slides showed bands of connective tissue between the microbeads and newly formed bone. These bands were composed of spindle-shaped cells aligned parallel to the surface of the microbeads (Figure 3).

Histomorphometric image analysis revealed the highest regenerated percent bone volume in the GPIV defects (43.24 ± 3.21) filled with the osteoblastic progenitors plus the VSM. On the other hand, the lowest mean bone volume was reported in GPV (23.93 ± 0.79) with the VSM plus the PDGF-treated cells (Table 1 and Figure 4). Paired 2-sample statistical comparison revealed significant differences between all tested groups except GPI (negative control) vs GPII (VSM only), and GPIII (PDGF) vs GPIV (VSM/cells) (Table 2 and Figure 4). PDGF treatment (GPIII) resulted in significant increase in bone volume when compared with the other groups except GPIV (VSM/cells) (Tables 1 and 2). Reconstructing the defects with the VSM alone resulted in no increase in bone volume (Tables 1 and 2; Figure 4). When the microbeads were used to carry osteoblastic progenitors, a statistically significant increase in bone volume was found. In contrast, the PDGF-treated cells resulted in significant inhibition of bone regeneration upon comparing GPV with all other groups (Table 2).

The study revealed that the vinyl styrene microbeads are able to provide a hospitable substratum for the attachment of the calvarial cells. This VSM property promoted its use as a carrier of osteoblastic cells to augment bone regeneration. It was interesting to note that the PDGF-BB changed the shape of attached cells. With PDGF, cells attained flat topography indicating firm attachment. None of the PDGF-treated cells revealed round topography, a property indicating less attachment.11 Therefore, PDGF-BB seems to have positive effect on the adhesion of osteoblasts to their surrounding environment. PDGF-BB upregulates the extracellular matrix proteins such as collagen and cell adhesion proteins (eg, fibronectin and vitronectin), which have positive effects on osteoblast adhesion and spreading.12,^–14 

Investigation of the histologic slides from the VSM-free groups (GPI and GPIII) revealed membrane collapse; a feature that might compromise bone healing. On the other hand, in VSM groups (GPII, GPIV, and GPV) the microbeads acted as a scaffold preserving a space for bone formation. Collapse of adjacent soft tissue and rapid migration of fibroblasts into the bony defect have been described as difficulties in bone regeneration.15 Though the microbeads prevented membrane collapse, they did not allow for more expansion of the regenerating bone because the material is nonbiodegradable. However, calculating the space occupied by both the VSM and bone revealed significant increase in defect fill as compared with the VSM-free groups (data not shown).

Osteoblasts were seen close to the VSM surface, which might involve an intrinsic biocompatible nature of the microbeads. In addition, there were no signs of inflammatory reaction to the VSM. These osteoconductive and biocompatibility properties confirm what has been reported earlier.6^,7 On the other hand, fibrous encapsulation of the microbeads seen in some slides has been reported with other graft materials such as hydroxyapatite in periodontal defects of dogs,16 nonresorbable ceramics in humans,17 and porous particulate composite (HTR).¹⁸ This connective tissue encapsulation might be related to graft micromotion19 or different charges acquired during sterilization and handling of the VSM. Further studies are needed to elucidate this phenomena and its effect on the long-term success rate of the graft material.

Although PDGF-BB by itself has potent mitogenic and chemotactic effects,20 it has no definite role in the overall bone regeneration process. In this study, it was found that PDGF-BB affected the function of the delivered cells significantly. This negative effect of PDGF-BB on bone formation is mainly attributed to reduction of the delivered cells' differentiation21^,22 as a result of incubation of the transplanted cells with PDGF-BB for 3 consecutive days. In contrast, when the PDGF-BB was directly applied to the surgical defects (GPIII), the effect was of short duration and resident osteoblastic cells utilized the PDGF-BB to proliferate more during that short application time. Another study has confirmed such enhancing effect of PDGF on short-term application.23 Therefore, the resident osteoblastic cells were not inhibited by the PDGF-BB action because it was not sustained for a long time to affect the differentiation of these cells.

Vinyl styrene microbeads are able to deliver osteoblastic cells to osseous defects. The ability of osteoblastic cells to lay down bone is significantly reduced upon pretreatment with PDGF-BB in vitro. However, short-term application of PDGF-BB to surgical bone defects has a stimulatory effect on bone regeneration.

The authors would like to thank Mrs Vera Larke, Mrs Salwa Kirbah, and Ms Cheryl Mims from the department of Oral Biology/Anatomy at the School of Dentistry, Medical College of Georgia, Augusta, Ga, for their dedication to this work and technical support. This work was partially supported by the Egyptian Cultural and Educational Bureau in Washington, DC.