Downloaded by Guangxi University for Nationalities from online.liebertpub.com at 02/20/18. For personal use only. ABSTRACT

Transcription

1 TISSUE ENGINEERING Volume 6, Number 4, 2000 Mary Ann Liebert, Inc. Promoted Bone Healing at a Rabbit Skull Gap Between Autologous Bone Fragment and the Surrounding Intact Bone with Biodegradable Microspheres Containing Transforming Growth Factor-b 1 LIU HONG, M.D., 1 YASUHIKO TABATA, Ph.D., 2 SUSUMU MIYAMOTO, M.D., D.Med.Sci., 1 KEISUKE YAMADA, M.D., D.Med.Sci., 1 IKUO AOYAMA, M.S., 1 MAKOTO TAMURA, Ph.D., 3 NOBUO HASHIMOTO, M.D., D.Med.Sci., 1 and YOSHITO IKADA, Ph.D., D.Med.Sci. 2 ABSTRACT This study is a trial to promote repairing of the rabbit skull bone gap between an autologous bone flap and the intact bone with biodegradable gelatin microspheres containing transforming growth factor-b 1 (TGF-b 1). A 10-mm diameter bone defect was prepared in rabbit skulls by drilling out a bone flap of 6 mm in diameter. After a surrounding gap defect of 2 mm was created and treated with 0.5 m g of free TGF-b 1 and gelatin microspheres containing 0.5 m g of free TGF-b1, the circular autologous bone flap was placed in the center. Significant bone healing at the gap defect was observed 3 weeks after implantation of the TGF-b 1-containing gelatin microspheres. The bone mineral density (BMD) was significantly higher than that of other experimental groups. On the contrary, when applied with free TGF-b 1, a fibrous tissue initially infiltrated into the gap defect, resulting in impairing bone healing. The tissue response was similar to that at the defect implanted with empty gelatin microspheres and TGF-b 1-free phosphate-buffered saline solution alone. There was more space in the gap-filling bone in the 16-week view than the 3-week view. It is possible that this was an intermediate step along the way toward normal healing and formation of cancellous bone. We conclude that gelatin microspheres containing TGF-b 1 show promise as an agent to promote bone regeneration of subcritical size defects between surgically positioned autologous bone flaps and surrounding host bone. INTRODUCTION SEVERAL W ORKERS have suggested that the craniofacial skeleton has poor self-regenerative powers beyond the early growing period. This becomes a clinical issue in cases of large defects caused by trauma or tumor resection. 1 Currently, autologous bone grafts have been clinically employed to reconstruct craniofacial 1 Department of Neurosurgery, Medical School, and 2 Institute for Frontier Medical Sciences, Kyoto University, Kyoto, Japan. 3 Kaken Pharmaceutica l Co. Ltd., Kyoto, Japan. 331

2 HONG ET AL. bone defects because of their higher osteoconductive ability and fewer side effects than currently available alloplastic materials. 2 However, the number of autograft sites and the amount of graftable bone found there is limited. It has been demonstrated that fibrous healing doses occur in craniofacial bones if the gap between bone ends exceeds several millimeters, especially if the bone is thin or if soft tissue is interposed. If a material with the ability of enhancing bone formation is obtained, it will achieve bone healing between the bone autografts or the autograft and the surrounding intact bone. As a trial to promote the bone healing, it will be efficient to use osteogenetic growth factors at the bone gap. It is well recognized that there are one or more signaling molecules like growth factors and cytokines that can induce or accelerate bone healing through the desired stages, from woven unorganized through compact lamellar bone. 3 However, it will be impossible to control the timing, concentration, and interaction of the growth factors artificially. Transforming growth factor-b 1 (TGF-b 1) is a multifunctional regulator possessing widespread effects on cellular proliferation, angiogenesis, migration, and immune response. It has been demonstrated that it can cause proliferation of osteoblasts and preosteoblastic cells and exerts an effect on fracture healing and bone regeneration. 4,5 However, the local application of the free TGF-b 1 solution showed the less osteogenic effect in vivo. A promising delivery system is necessary to promote its osteogenetic ability by providing sufficient quantities over the desired time period. We have developed a biodegradable hydrogel system prepared through chemical crosslinking of acidic gelatin with glutaraldehyde and succeeded in releasing biologically active TGF-b 1 based on hydrogel degradation. 6,7 This gelatin hydrogel effectively promoted TGFb 1-induced bone regeneration at a rabbit skull defect at an effective TGF-b 1 dose, in marked contrast to TGF-b 1 in solution. A long-term evaluation study revealed that the regenerated bone had the tissue structure similar to the compact bone that surrounds the diploic space periostally and endostially. 8 In the present study, we prepared microsphere gelatin hydrogel for application to a 2-mm bone defect, and investigated bone formation at the gap defect after application of TGF-b 1 containing gelatin microspheres. Materials MATERIALS AND METHODS Gelatin with an isoelectric point (IEP) of 5.0, prepared through alkaline processing of bovine bone, was kindly supplied from Nitta Gelatin Co. (Osaka, Japan). It is named acidic gelatin on the basis of its low isoelectric point. Human recombinant TGF-b 1 with an IEP of 9.5 was purchased from Sigma Chemical Co., MI. Other chemicals were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan) and used without further purification. Preparation of gelatin microspheres containing TGF-b 1 Gelatin microspheres were prepared through chemical crosslinking of gelatin aqueous solution in an emulsion state, as described previously. 9 Briefly, 10 ml of the mixed gelatin (10 wt%) and glutaradehyde ( wt %) aqueous solution was stirred (420 rpm) (Heidon 600G, Sinto Scientific Co. Tokyo, Japan) in 375 ml of olive oil at 40 C for 10 min to obtain a water-in-oil emulsion. The emulsion temperature was decreased to below 20 C and stirring was continued for 24 h to allow chemical crosslinking of gelatin to proceed. After acetone addition, the crosslinked microspheres were collected by centrifugation (4 C, 3,000 rpm, 5 min) and washed five times by acetone. The washed microspheres were placed in 100 ml of 100 mm glycine aqueous solution containing 0.1 wt% Tween 80, followed by agitating at 37 C for 1 hr to block the residual aldehyde groups of unreacted glutaraldehyde. Then, the resulting microspheres were washed three times with double-distill ed water by centrifugation, freeze-dried, and sterilized with ethylene oxide gas. When calculated from the microsphere volume before and after swelling in 0.05 M phosphate-buffered saline solution (PBS, ph 7.4) for 24 h at 37 C, the water content of microspheres used here was 95 wt%, being the volum e ratio of water in the microspheres to the wet microspheres. At least 100 microspheres were viewed with a light microscope to calculate their volum e. The average diameter or wet microspheres was 100 m m. Next, 8 mg of gelatin microspheres was applied for each skull gap defect, this is the largest amount of 332

3 PROMOTED BONE HEALING BY TGF-b 1-MICROSPHERES this material that can fit in the defect. TGF-b 1 was impregnated into gelatin microspheres by dropping 12.5 m g/ml of TGF-b 1 solution in PBS (40 m L) onto 8 mg of freeze-dried gelatin microspheres, and the mixture was left at 4 C for 12 h to allow the solution to absorb into the microspheres. TGF-b 1-free empty microspheres were prepared similarly, except that PBS was added. The PBS solution with or without TGFb 1 was completely absorbed into these microspheres because it was less than the amount required for equilibration of the internal versus external microsphere environments. Animal experiments Fifty Japanese white rabbits weighing from 2.0 to 2.5 kg were anesthetized by intravenous administration of sodium pentobarbita l. In addition, lidocaine solution containing 1 wt% epinephrine was injected subcutaneously at the rabbit head for local anesthesia. The 4-cm-long scalp skin cut was prepared at midline head of rabbit. Then the pericranium was cut along the sagittal suture and retracted to expose the bilateral parietal bone of rabbit. Two bone defects of 10 mm in diameter were prepared per rabbit skull, one on each side of the sagittal suture symmetrically by taking out a skull bone flap of 6 mm in diameter with a microdrill. All the procedures were performed with the aid of an operating microscope so as not to injure the underlying dura mater. After the autologous bone flap obtained was placed at the center of skull defect, the 2-mm-width gap defect between the bone flap and the surrounding skull bone was filled with gelatin microspheres containing TGF-b 1 (Fig. 1). There were four kinds of agents applied: group A, gelatin microsphere containing TGF-b 1; group B, PBS solution with TGF-b 1; group C, gelatin microsphere containing without TGF-b 1; and group D, PBS solution without TGF-b 1. The type of agent was randomly selected from either the right or left bone defect of different rabbits. Following implantation, the pericranium and scalp were carefully sutured with 4-0 nylon suture. The rabbits were sacrificed by intravenous administration of an overdose of sodium pentobarbital 3, 6, 9, 12, and 16 weeks after implantation (n 5 10). Five defects were obtained for each group at each time point. The skull bone specimen containing two defects was taken out from each rabbit and fixed in 10 wt% neutralized aqueous solution of formaldehyde for bone assessments. FIG. 1. A photograph of rabbit skull defects (10 mm in diameter) applied with combination of an autologous bone flap (6 mm in diameter) and PBS (right) or gelatin microspheres containing 0.5 m g of TGF-b 1 (left). The bone flap was placed in the center of the skull defect and the 2 mm gap between the bone flap and the surrounding skull bone was implanted with PBS and the gelatin microspheres. 333

4 HONG ET AL. Assessment of bone formation of gap defects Bone formation at the skull gap defect between the bone flap and the surrounding skull bone was assessed first on the basis of plain film X-ray to evaluate qualitatively mineral deposition at the gap defect. The plain film X-ray radiograph of each bone specimen was taken at 46 kv and 2 ma for 15 s. Then, bone mineral density (BMD) at every gap defect was quantitated by dual energy X-ray absorptom etry (DEXA) (model DCS 600, Aloka Co., Tokyo, Japan). The BMD value for each gap defect was obtained by average of BMD values measured at eight sites evenly selected from the gap defect around the bone flap. The region of interest in the absorptometry measurement was adjusted at mm 2. Following the BMD measurement, the skull bone specimens were demineralized in 10 wt% ethylenediamine tetraacetic acid aqueous solution, embedded in paraffin, and cut into 2-m m-thick sections at the center of the skull defect or the nearest site possible. The sections were stained with tartrate-resistant acid phosphatase and eosin to highlight osteoclasts, histologically viewing throughout the trabecular bone newly formed at the gap defect. From the 2-mm-width area located at the gap defect on histological sections, the perimeter length of total trabecular bone surface, the bone surface where active osteoblasts characterized by basophilic cuboided cytoplasm located were recruited (active formation surface), and the bone surface where osteoclasts characterized by red staining irregular multinucleated cell located were recruited (active resorption surface) was measured by use of an image analyzer with a light microscope. The measurement FIG. 2. Soft X-ray photographs of rabbit skull defects 3 weeks after application with an autologous bone flap combined with PBS (A), TGF-b 1-free gelatin microspheres (B), 0.5 m g of free TGF-b 1 (C), and gelatin microspheres containing 0.5 m g of TGF-b 1 (D). Bar 5 1 mm. 334

5 PROMOTED BONE HEALING BY TGF-b 1-MICROSPHERES was performed by a researcher blinded to the individual experiment groups. The fractional active formation surface (FrAFS) and fractional active resorption surface (FrARS) were defined as the perimeter percentage of the active formation surface and the active resorption surface to the total bone surface, respectively. The value of FrAFS and FrARS for each specimen was averaged by two gap defects at one skull defect. Statistical analysis Each defect was considered independent. All the data were statistically analyzed by Scheffe s F test for multiple comparison and statistical significance was accepted at p, Experimental results were expressed as the mean 6 the standard deviation (SD) about the mean. RESULTS Bone formation at the skull gap defect by gelatin microspheres containing TGF-b 1 Figures 2 and 3 show plain film X-ray photographs of skull defects 3 and 16 weeks after implantation with TGF-b 1-containing gelatin microspheres or other agents. The gap defect became radiopaque 3 weeks FIG. 3. Soft X-ray photographs of rabbit skull defects 16 weeks after application with an autologous bone flap combined with PBS (A), TGF-b 1-free gelatin microspheres (B), 0.5 m g of free TGF-b 1 (C), and gelatin microspheres containing 0.5 m g of TGF-b 1 (D). Bar 5 1 mm. 335

6 HONG ET AL. after implantation with the TGF-b 1-containing gelatin microspheres (Fig. 2D). Little gap defect radio-opacity was found in the specimens receiving 0.5 m g of free TGF-b 1. This radiographic appearance was similar to that of gap defect treated with TGF-b 1-free gelatin microspheres and PBS (Fig. 2). The free TGFb 1-induced radio-opaque area tended to increase 16 weeks after implantation, although a radiolucent line remained around the bone flap. The similar time change was observed for defects after application with groups C and D. For the gelatin microspheres containing TGF-b 1, the gap defect remained radio-opaque at 16-week after implantation (Fig. 3). Figure 4 shows the time profile of BMD values at the gap defect of rabbit skulls after application of various agents. The BMD value of gap defects after application with group A was significantly higher than that of other agents 3 weeks after implantation, and reached a maximum level at 6 weeks after implantation, followed by the subsequent gradual decrease. There was little change in the BMD value at the gap defect after application with free TGF-b 1 over the time range studied, which was similar to the time profile of BMD value for empty microspheres and PBS. No significant difference of BMD values at gap 16 weeks after application with all agents. Histological evaluation of bone formation at the gap defect enhanced by gelatin microspheres containing TGF-b 1 Figures 5 and 6 show the histological sections of gap defects 3 and 16 weeks after application of various agents. Significantly, bone formation occurred at the gap defect 3 weeks after receiving the TGF-b 1- containing gelatin microsphere implantation, and the bone flap was connected to the surrounding skull bone. Bone formation enhanced by free TGF-b 1 at the gap defect was hardly noticeable and instead a fibrous tissue was observed at the defect, which was histologically similar to that occurring in groups C and D (Fig. 5). Gelatin microspheres were completely degraded and did not remain at the gap defect 3 weeks after implantation, irrespective of the TGF-b 1 incorporation. The bone tissue regenerated by TGF-b 1-containing gelatin microspheres 16 weeks after implantation turned loose compared with bone tissues 3 weeks after implantation (Fig. 6D). The amount of bone formed 16 weeks after application with groups B, C, FIG. 4. The time course of BMD values at the gap defect of rabbits between the bone flap and the surrounding skull bone after treatment with gelatin microspheres containing 0.5 m g of TGF-b 1 and other agents. (p ) p, 0.05 against the BMD value of defects applied with PBS. ( ) p, 0.05 against the BMD value of defects applied with TGF-b 1-free gelatin microspheres. ( ) p, 0.05 against the BMD value of defects applied with free TGF-b

7 PROMOTED BONE HEALING BY TGF-b 1-MICROSPHERES and D seemed to increase when compared with that 3 weeks after treatment. However, the thickness of the newly formed bone seems thinner than that formed after application with group A, and, the newly formed bone was located near to dura mater. Fibrous tissues were still observed to interpose at the gap defect (Fig. 6A C). Figure 7 shows the time profile of FrAFS and FrARS values for bone newly formed at the gap defect after application with various agents. For all experimental agents, the FrAFS and FrARS values of bone regenerated initially increased to attain a maximum level and thereafter decreased to a similar level. However, the implantation period to attain the maximum levels was shortened by application with the Group A reagent. After an implantation period of 12 weeks, both the FrAFS and FrARS levels became similar, irrespective of the agent of experimental groups. There was no significant difference in the BMD value and the bone area for each experimental group during the whole observation period. DISCUSSION It has been reported that several growth factors such as bone morphogenetic protein, basic fibroblast growth factor, and TGF-b 1 possessed osteogenetic ability, and they were applied to skull defect models to accelerate the bone regeneration TGF-b 1 exists mainly in bone and has been demonstrated to possess an effect on the proliferation of preosteoblastic and osteoblastic cells, including human osteoblasts. A single therapeutic dose of TGF-b 1 also has been shown to promote bone formation at a critical-sized defect. 18,19 We have demonstrated that TGF-b 1 was ionically complexed with acidic gelatin and, as a result, was released in vivo from the gelatin hydrogel during hydrogel biodegradation. In our previous study, application of TGF-b 1 could effectively recruit osteoblasts at defect margin adjacent to dipole and dura mater. FIG. 5. Histological sections of the gap defect of rabbits between the bone flap and the surrounding skull bone 3 weeks after application with PBS (A), TGF-b 1-free gelatin microspheres (B), 0.5 m g of free TGF-b 1 (C), and gelatin microspheres containing 0.5 m g of TGF-b 1 (D). Arrows indicate the edge of gap defect. Bar 5 1 mm. BF, Bone flap; C, connective tissue; DM, dura mater; NB, new bone; HB, host bone. (TRAP and eosin staining; original magnification 3 40.) 337

8 HONG ET AL. FIG. 6. Histological sections of the gap defect of rabbits between the bone flap and the surrounding skull bone 16 weeks after application with PBS (A). TGF-b 1-free gelatin microspheres (B), 0.5 m g of free TGF-b 1 (C), and gelatin microspheres containing 0.5 m g of TGF-b 1 (D). Arrows indicate the edge of gap defect. Bar 5 1 mm. BF, Bone flap; C, connective tissue; DM, dura mater; NB, new bone; HB, host bone. (TRAP and eosin staining; original magnification 3 40.) An application of more than 0.1 m g of TGF-b 1-containing gelatin hydrogels with a suitable biodegradability in a disc type could significantly induce bone formation at a rabbit skull defect in 6 mm diameter 6 weeks after implantation. 8 It appeared that the controlled release of TGF-b 1 not only enhanced osteoblast recruitment into the bone defect but also prolonged osteoblasts in the active state. The present study clearly demonstrated that the TGF-b 1 incorporated in gelatin microspheres could significantly promote bone healing at the skull bone gap between the autograft and the intact bone, which is similar to the result obtained by the TGF-b 1-incorporating gelatin hydrogel of disc type. It enabled bone to connect between the autologous bone flap and the surrounding skull bone within a short time period, whereas free TGF-b 1 was ineffective. This is likely explained in terms of controlled release of TGF-b 1 from gelatin microspheres. It is possible that the controlled release of TGF-b 1 localized at the skull defect enhances recruitment of active osteoblasts, thereby resulting in superior bone formation. Free TGF-b 1 may have been shown to reduce bone regeneration due to shunting of the TGF-b 1 solution out of the site and into the interstitial compartment at large, or to its being rapidly metabolized. The achievement of the balance of resorption and formation at newly formed bone by application of TGF-b 1-containing gelatin microspheres seemed no different with normal bone formation at the defect. It is likely that this microsphere system enables TGF-b 1 to be released over an extended time period at the defect site, resulting in promoting the bone healing at the gap defect between the bone flap and host bone. The self-regenerated bone was observed after 16 weeks, and the BMD values of the gap defect with PBS applied were comparable to those of TGF-b 1-containing gelatin microspheres. However, the thickness of the newly formed bone apparently was thinner than that promoted by TGF-b 1-containing microspheres. The new bone formed near to the dura mater is probably explained by the markedly high-affinity receptors for TGF-b 1 located in the dura mater. 20 It is well known that TGF-b 1 is also associated with the formation of fibrous scar tissue. 21 Thus, the prevention of the scalp fibrous tissue infiltrating into the defect and the formation of fibrous scar tissue are the 338

9 PROMOTED BONE HEALING BY TGF-b 1-MICROSPHERES FIG. 7. The time course of FrAFS (s ) and FrARS (d ) values of the bone newly formed at the gap defect of rabbits between the bone flap and the surrounding skull bone after application with various agents. important problems to be resolved for the practical application of TGF-b 1-containing gelatin microspheres as a promising therapeutic agent for promoting the bone healing. ACKNOWLEDGMENT This research was supported by Research for the Future Program Grant No. JSPS-RFTF from the Japan Society for the Promotion of Science. 339

10 HONG ET AL. REFERENCES 1. Schmitz, J.P., and Hollinger, J.O. The critical size defect as an experimental model for craniomandibulo facial nonunions. Clin. Orthop. 205, 299, Habal, M.B. Bone grafting in craniofacial surgery. Clin. Plast. Surg. 21, 349, Lind, M. Growth factor: possible new clinical tools. A review. Acta Orhop. Scand. 67, 407, Bosch, C., Melsen, B., Gibbons, R., and Vargervik, K. Human recombinant transforming growth factor- b 1 in healing of calvarial bone defects. J. Craniofac. Surg. 7, 300, Lind, M., Schumacker, B., Soballe, K., Keller, J., Melsen, F., and Bunger, C. Transforming growth factor-b enhances fracture healing in rabbit tibiae. Acta Orthop. Scand. 64, 553, Yamamoto, M., Tabata, Y., Hong, L., Miyamato, S., Hashimoto, N., and Ikada, Y. Bone regeneration by transforming growth factor b 1 release from a biodegradable hydrogel. J. Cont. Rel. 14, 133, Tabata, Y., and Ikada, Y. Protein release from gelatin matrics. Adv. Drug Deliv. Rev. 31, 287, Hong, L., Tabata, Y., Miyamoto, S., Yamamoto, M., Yamada, K., Hashimoto, N., and Ikada, Y. Bone regeneration at rabbit skull defects treated with transforming growth factor b 1 incorporated in hydrogels with different biodegradabilities. J. Neurosurg. 92, 315, Tabata, Y., Hijikata, S., Muniruzzaman, M.D., and Ikada, Y. Neovascularizati on effect of biodegradable gelatin micropheres incorporating basic fibroblast growth factor. J. Biomater. Sci. Polymer Edn. 10, 79, Linkhart, T.A., Mohan, S., and Baylink, D.J. Growth factors for bone growth and repair: IGF, TGFb and BMP. Bone 19 (Suppl. 1), 1S, Canalis, E. Effect of growth factor on bone cell replication and differentiation. Clin. Orthop. 193, 246, Cook, S.D., Wolfe, M.W., Salkeld, S.L., and Rueger, D.C. Effect of recombinant human osteogenic protein on healing of segmental defects in non-human primates. J. Bone Joint Surg. 77, 734, Takita, H., Tsuruga, E., Ono, I., and Kuboki, Y. Enhancement by bfgf of osteogenesis induced by rhbmp-2 in rats. Eur. J. Oral Sci. 105, 588, Tabata, Y., Yamada, K., Miyamoto, S., Nagata, I., Kikuchi, H., Aoyama, I., Tamura, M., and Ikada, Y. Bone regeneration by basic fibroblast growth factor complexed with biodegradable hydrogels. Biomaterials 19, 807, Yamada, K., Tabata, Y., Yamamoto, K., Miyamoto, S., Nagata, I., Kikuchi, H., and Ikada, Y. Potentiality of bfgf incorporated in biogradable hydrogens for skull bone regeneration. J. Neurosurg. 86, 871, Hong, L., Tabata, Y., Yamamoto, M., Miyamoto, S., Yamada, K., Hashiomoto, N., and Ikada, Y. Comparison of bone regeneration in a rabbit skull defect by recombinant human BMP-2 incorporated in biodegradable hydrogel and in solution. J. Biomater. Sci. Polymer Edn. 9, 1001, Moxham, J.P., Kibblewhite, D.J., Dvorak, M., Perey, B., Tencer, A.F., Bruce, A.G., and Strong, D.M. TGF-b 1 forms functionally normal bone in a segmental sheep tibial diaphyseal defect. J. Otolaryngol. 25, 388, McKinney, L., and Hollinger, J.O. A bone regeneration study: transforming growth factor-b 1 and its delivery. J. Craniofac. Surg. 7, 36, Moxham, J.P., Kibblewhite, D.J., Bruce, A.G., Rigley, T., Gillespy, T 3rd, and Lane, J. Transform ing growth factor-b 1 in a guanidine-extracte d demineralized bone matrix carrier rapidly closes a rabbit critical calvarial defect. J. Otolaryngol. 25, 82, Mehrara, B.J., Steinbrech, D.S., Saadeh, P.B., Gittes, G.K., and Longaker, M.T. Expression of high-affinity receptors for TGF-b during rat cranial suture fusion. Ann. Plast. Surg. 42, 502, Shah, M., Foreman, D.M., and Ferguson, M.W. Neutralisation of TGF-beta 1 and TGF-beta 2 or exogenous addition of TGF-beta 3 to cutaneous rat wounds reduces scarring. J. Cell Sci. 108, 983, Address reprint requests to: Yasuhiko Tabata, Ph.D. Institute for Frontier Medical Sciences Kyoto University 53 Kawara-cho Shogoin Sakyo-ku, Kyoto , Japan yasuhiko@frontier.kyoto- u.ac.jp 340