| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Orthopedic Surgery, University of Tokyo Graduate School of Medicine (H.K., K.N.), Hongo 7-3-1, Bunkyo 113-8655, Tokyo; Institute for Frontier in Medical Science, Kyoto University (Y.T.), Kyoto; Faculty of Medical Engineering, Suzuka University of Medical Science (Y.I.), Mie; and Kaken Pharmaceutical Co., Ltd. (I.A., J.A., T.N., Y.H., M.T.), Minamikawara-machi, Kyoto, Japan
Address all correspondence and requests for reprints to: Hiroshi Kawaguchi, M.D., Ph.D., Department of Orthopedic Surgery, University of Tokyo Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: kawaguchi-ort{at}h.u-tokyo.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Among many growth factors intrinsic to the skeletal tissues, fibroblast growth factor-2 (FGF-2 or basic FGF) is recognized as a potent mitogen for a variety of mesenchymal cells (1, 2, 3). In skeletal tissues, FGF-2 is produced by cells of the osteoblastic lineage, accumulates in bone matrix, and acts as an autocrine/paracrine factor (4, 5, 6, 7). Previous studies have shown that FGF-2 is a more potent mitogen for immature mesenchymal cells, prechondrocytes, and preosteoblasts than for differentiated osteoblasts (8, 9, 10). It has also been reported to inhibit the differentiation of osteoblastic and chondroblastic cells and matrix synthesis, such as collagen I (8, 9, 10, 11, 12, 13, 14, 15). Several genetic diseases with abnormalities in bone and cartilage formation, such as achondroplasia and thanatophoric dysplasia type II, have been shown to be due to mutations of genes for FGFs or their receptors (16, 17, 18, 19), suggesting the importance of FGFs in bone and cartilage formation.
Aiming at a clinical application, we reported the anabolic effect of local and systemic administrations of FGF-2 on bone formation using several animal models (20, 21, 22, 23, 24, 25, 26, 27, 28). A single local application of FGF-2 facilitated the healing of bone fracture and segmental bone defect in normal and diabetic rats, rabbits, and dogs (20, 21, 22); stimulated bone formation in callotasis bone lengthening in rabbits (23); and increased bone mass intraosseously in normal and ovariectomized rats and rabbits (24, 25, 26, 27). In addition, daily systemic administration of FGF-2 facilitated endosteal bone formation (28). Numerous reports from other groups have confirmed the anabolic action of FGF-2 on bone formation (29, 30, 31, 32, 33, 34, 35, 36).
In preparatory experiments performed to develop an actual delivery system for clinical use, we compared the anabolic effects of FGF-2 in several carriers using the rat fracture model (20) and the rabbit bone defect model (21). FGF-2 exerted the most potent anabolic activity when a synthetic bioabsorbable hydrogel prepared through glutaraldehyde cross-linking of gelatin was used as the carrier (35, 37, 38). Because studies using nonhuman primates are essential before the clinical application of FGF-2 as an agent to prevent delayed union or nonunion, we intended to make a nonhuman primate model whose fracture healing is rather difficult if left untreated. This study investigated the action of FGF-2 on fracture healing in nonhuman primates by comparing the anabolic effects of recombinant human FGF-2 in the gelatin hydrogel with those of the gelatin hydrogel alone.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant human FGF-2 was purchased from Scios, Inc. (Mountain View, CA). Biodegradable gelatin hydrogel, the carrier of FGF-2, was prepared through the glutaraldehyde cross-linking of acidic gelatin with an isoelectric point of 5.0 as reported previously (37, 38).
Experimental design
Male cynomolgus monkeys (Macaca fascicularis, 4–5-yr old) imported from China National Scientific Instruments and Materials Import/Export Corp. (Beijing, China) were used, and all experiments except histological and biomechanical analyses were carried out at the experimental monkey center of Shin Nippon Biomedical Laboratories (Kagoshima, Japan). After 25 animals were allowed to adapt to vivarium conditions for 1 week, 20 animals in good health were selected and randomly divided into 2 groups (n = 10 each). There was no significant difference in body weight between the groups (range, 3.21–3.92 kg; average, 3.59 ± 0.07 kg in the vehicle group; 3.31–3.92 kg, average 3.62 ± 0.06 kg in the FGF-2 group; mean ± SEM). Radiographs were made to verify skeletal maturity and lack of bone abnormalities. Experiments were performed according to the guidelines of the International Association for the Study of Pain (39), and the experimental work was reviewed by the committees of Tokyo University and Shin Nippon Biomedical Laboratories charged with confirming ethics. The animals were housed individually under the following conditions: temperature, 26 ± 2 C; humidity, 50 ± 10%; air changes, 15 times/h; and 12-h light cycle. The animals were daily fed 108 g Harlan Teklad monkey diet (Harlan Sprague Dawley, Inc., Madison, WI).
Under general anesthesia with ketamine hydrochloride (10 mg/kg; Sigma, St. Louis, MO) the right upper limb was readied and draped in sterile fashion. A 3- to 4-cm lateral incision was made longitudinally in the forearm, and blunt dissection of the muscle was used to expose the ulna. The middle point of the ulna was marked with a surgical marker, then the periosteum was stripped gently from the bone surface for 3 cm along its long axis at the center of the marked point to avoid injury during the osteotomy. The length of the area to be stripped was measured by Vernier calipers and marked on the bone surface. A transverse osteotomy at the marked midpoint of the ulna was produced sharply using a sagittal blade (Striker, Kalamazoo, MI), and the full length of the bone marrow cavity was internally fixed with an intramedullary Kirschner wire, 1.2–1.8 mm in diameter depending on the size of the cavity. After repositioning and copious irrigation with saline to remove bone debris and spilled marrow cells, the gelatin hydrogel alone (250 µL/site; vehicle group; n = 10) or the gelatin containing FGF-2 (200 µg/250 µL; FGF-2 group; n = 10) was injected into the fracture site. The dose of FGF-2 was fixed as 200 µg/site based on the results of a preparatory dose-response study on the bone mineral content (BMC) at the fracture site using the same model. The operators were blinded as to the type of gel. The periosteum, muscle, and skin were then meticulously closed in layers to contain the hydrogel at the fracture site. No external fixation was used, and the animals were allowed unrestricted activity as well as the diet and water ad libitum. At 0, 2, 4, 6, 8, and 10 weeks after the operation, radiographs were taken, and the BMC and bone mineral density (BMD) were measured. Blood and urine samples were taken 1–3 days before these time points for 6 weeks. Animals were killed at 10 weeks by exsanguination under anesthesia with sodium pentobarbital (25 mg/kg; Tokyo Kasei, Inc., Tokyo, Japan), and bilateral ulnae were excised. After extracting the intramedullary nail, the soft tissues surrounding the bilateral ulnae, except for the soft callus around the fracture site, were gently removed for the biomechanical and histological analyses.
Blood and urine chemistries
Blood samples were taken from the femoral vein on the day before the operation and at 2, 4, and 6 weeks after the operation. The serum bone alkaline phosphatase (bone ALP) level was measured using an enzyme immunoassay kit (Mitsubishi Chemical BCL, Tokyo, Japan). Urine samples were collected for 16 h/day for 3 days before the operation and at each time point mentioned above, and the total urine volume was measured. The urinary deoxypyridinoline and creatinine levels were measured by high pressure liquid chromatography (Mitsubishi Chemical BCL) and by Jaffe’s method using an automatic analyzer (Clinalyzer RX-10, Nihon Denshi Co., Ltd., Tokyo, Japan), respectively. The deoxypyridinoline/creatinine value was calculated for each animal, and the data were expressed as the mean ± SEM in each group.
Radiological analysis
X-Ray pictures of the bilateral ulnae were taken at 0 (immediately after the operation), 2, 4, 6, 8 (under general anesthesia with 5–10 mg/kg ketamine hydrochloride), and 10 weeks (after death) using an x-ray apparatus (DOP-82, Hitachi Medico, Tochigi, Japan). To determine bone union, the bony bridging on radiographs was judged by individuals who were blinded with regard to the group.
Measurements of BMC and BMD
BMC and BMD of the fracture site were measured by dual energy x-ray absorptiometry using a bone mineral analyzer (DCS-600, Aloka Co., Tokyo, Japan) at 0 (immediately after the operation), 2, 4, 6, 8, and 10 weeks after the operation. The region of interest was the entire area of the callus and the midportion of the ulna that was 4 cm in length longitudinally at the center of the fracture site. A preliminary experiment revealed that the intramedullary nail did not affect the BMC and BMD values. The operators were blinded as to the status of the animals. The percent gain in BMC and BMD at 10 weeks compared with that at 0 week was also calculated in each animal. The raw value and the percent gain in BMC and BMD were expressed as the mean ± SEM in each group.
Biomechanical analysis
A three-point bending test was performed on the fracture site of the harvested right ulna and the identical site (midportion) of the left ulna using a load torsion tester (MZ-500S, Maruto, Tokyo, Japan). The bending force was applied with the cross-head at a speed of 2 cm/min until fracture occurred. The maximum load (N), the breaking energy (N·m), and the stiffness (N/mm) were interpreted and calculated from the load deflection curve, which was recorded continually in the computerized monitor linked to the tester as described previously (40). The percentage of each parameter of the fractured (right) ulna compared with that of the unfractured (left) ulna was calculated in each animal, and the data were expressed as the mean ± SEM in each group.
Histological analysis
After the mechanical test described above, the tibiae were fixed in phosphate-buffered 10% formalin, demineralized in 10% formid acid for 2 weeks, embedded in paraffin, and cut into 4-µm-thick sections. The sections were stained with Masson-Trichrome and examined under a light microscope.
Statistical analyses
Statistical analysis was performed using Dunnett’s t test at each time point between the vehicle and FGF-2 groups after ANOVA.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
No significant difference in body weight gain was seen between the vehicle and FGF-2 groups after 10 weeks of the experimental period (0.23 ± 0.02 kg in the vehicle group and 0.28 ± 0.03 kg in the FGF-2 group; mean ± SEM). Neither the serum bone ALP level nor the urinary deoxypyridinoline/creatinine level changed during the 6-week observation period after the operation, nor was there a significant difference between the two groups at any time point. The bone ALP levels preoperatively and at 2, 4, and 6 weeks postoperatively were 247.5 ± 23.7, 242.0 ± 28.8, 256.8 ± 31.2, and 234.3 ± 28.6 U/L, respectively, in the vehicle group and 292.0 ± 20.7, 266.8 ± 23.5, 305.7 ± 28.1, and 275.5 ± 32.6 U/L at each time point in the FGF-2 group. The deoxypyridinoline/creatinine levels were 31.0 ± 5.8, 27.0 ± 2.7, 23.8 ± 2.6, and 33.0 ± 4.5 pmol/µmol at each time point in the vehicle group and 23.0 ± 2.2, 23.2 ± 2.4, 22.5 ± 2.3, and 25.7 ± 4.2 pmol/µmol in the FGF-2 group. These results indicate that FGF-2 did not affect the general condition of the subject, including bone turnover.
Radiological and histological findings
X-Ray features of the fracture site of all animals at 10 weeks are
shown in Fig. 1
(vehicle, animals
1–10; FGF-2, animals 11–20). As neither external fixation nor any
restriction of activity was imposed on the animals after the operation,
we regard this as a model for delayed union or nonunion. In the vehicle
group, in fact, fracture healing was poor, and four animals (no. 1, 4,
5, and 8) showed nonunion even at 10 weeks, whereas in the FGF-2 group
bone union was complete in all animals. In a representative
vehicle-treated animal (no. 1 in Fig. 1), no bony bridging was seen at
the fracture site, and the fracture gap remained for 10 weeks,
indicating nonunion (Fig. 2). In a
representative FGF-2-treated animal (no. 11 in Fig. 1), on the
contrary, the calcified callus was observed at as early as 2 weeks, and
bony bridging was seen at the fracture site 4 weeks after the
operation. Histological features of the fracture site at 10 weeks
demonstrated that the fibrous and cartilaginous tissues remained
unmineralized in the vehicle animal, whereas the entire callus was
calcified in the FGF-2 animal. The time-course study of the number of
animals with fracture union determined by bony bridging on x-ray
revealed that all of those in the FGF-2 group showed fracture union,
whereas only four animals in the vehicle group did so at 6 weeks (Fig. 3A). Four animals in the vehicle group
that showed no bone union at 8 weeks remained in a nonunion state even
at 10 weeks. Even in animals with bone union within 10 weeks (n =
6 for vehicle and n = 10 for FGF-2), the average time for such
union was significantly reduced by FGF-2 (P < 0.05;
6.67 ± 0.42 and 5.60 ± 0.27 weeks in the vehicle and FGF-2
groups, respectively; mean ± SEM).
|
|
|
New bone formation at the fracture site of the right (fractured)
ulna was further examined by measuring the BMC and BMD of the area that
was 4 cm in length longitudinally at the center of the fracture site
(Fig. 3
, B and C). Significant differences between the vehicle (n
= 10) and FGF-2 (n = 10) groups were seen at 6 weeks in both BMC
and BMD, and these differences increased with time. To examine the
systemic effect of FGF-2 that has been reported in rodent models
(28, 30, 31), BMC and BMD at the identical region of the
contralateral (left, unfractured) ulna were measured at 10 weeks. These
features were not different between the vehicle and FGF-2 groups (BMC,
661.7 ± 25.9 vs. 648.6 ± 15.5 mg; BMD,
262.3 ± 8.3 vs. 258.4 ± 6.3
mg/cm2; mean ± SEM).
These results indicate that the local effect of FGF-2 is much stronger
than its systemic effect in this model.
Mechanical property
Ten weeks after the operation we harvested bilateral ulnae for the
biomechanical analysis. The FGF-2 group showed significantly higher
parameter levels of the mechanical property of the fracture site than
the vehicle group. Levels of both maximum load and breaking energy in
the FGF-2 group were almost double those in the vehicle group, whereas
stiffness in the two groups was not significantly different (Fig. 3D).
Comparison of bone mass and mechanical property in animals with bone union
To examine whether the significant differences in bone mass and
mechanical property between the vehicle and FGF-2 groups are dependent
on the presence or absence of nonunion animals, these parameters at 10
weeks were compared among three groups: vehicle without bone union
(n = 4), vehicle with bone union (n = 6), and FGF-2 (n =
10) groups (Table 1). Not only the FGF-2
group but also the vehicle with union group showed significantly higher
values than those of the vehicle without union group in all parameters
except breaking energy. The FGF-2 group still exhibited higher values
than the vehicle with union group in all parameters except stiffness,
although a significant difference was seen only in the percent gain in
BMC and BMD.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The precise mechanism by which FGF-2 stimulates bone formation remains to be identified. FGF-2 is reported to stimulate the proliferation of immature mesenchymal cells and inhibit the differentiation and matrix synthesis of osteoblastic cells, suggesting that the anabolic action of FGF-2 is not via mature osteoblasts (8, 9, 11, 12, 13, 14). Our preliminary experiments using the rat and rabbit models demonstrated that the local half-life of the injected 125I-labeled FGF-2 in the gelatin hydrogel was 1–2 days. The percentages of 125I-labeled FGF-2 remaining localized were approximately 20%, 3%, and 1% at 1, 2, and 3 weeks, respectively. Hence, FGF-2 appears to have its effect primarily in the earlier stage of fracture healing, probably through its mitogenic action on undifferentiated mesenchymal cells, as we observed histologically in rodent models (22, 25). Although histological findings suggest that the defect in impaired healing seen in vehicle-treated animals is due to a failure of chondrocyte maturation, little FGF-2 remains localized at the stage of chondrocyte maturation. The effect of FGF-2 on chondrocyte differentiation is controversial (10, 12, 15, 41, 42, 43); however, the majority of previous reports have shown that FGF-2 is inhibitory, rather than stimulatory, for chondrocyte differentiation (10, 12, 15). Genetic studies showing that constitutive activation of FGF receptor 3 results in achondroplasia also indicate the inhibitory action of FGF signaling on chondrocyte differentiation (18, 19). Another possible mechanism of FGF-2 may be the induction of other anabolic factors, such as PGs, transforming growth factor-ß, and bone morphogenetic proteins (BMPs), which may compose a serial cascade of bone formation, including chondrocyte differentiation. In fact, we and others have shown that FGF-2 induces transforming growth factor-ß expression and PG production in a rat fracture model and in cultured osteoblastic cells (21, 44, 45). Thus, FGF-2 may act as a factor that initiates the cascade of the bone formation process. The fact that the difference in percent gain in BMC between the FGF group and the vehicle with bone union group was greater than that in BMD indicates a large callus formation by FGF-2. It is therefore speculated that the formation of large callus at the early stage may be the major effect of FGF-2 in the entire fracture-healing process. Because the instability of the fracture site due to an improperly formed callus at the early stage of healing is reported to lead to a definite pathogenesis of nonunion (46), this character of FGF-2 is likely to be ideal in assuring nonunion is avoided.
Earlier reports have shown the anabolic action of FGF-2 on bone formation in nonhuman primates (35, 36). Tabata et al. reported that FGF-2 (100 µg) incorporated in the gelatin hydrogel used in this study induced bone formation at the site of a skull defect (6 mm in diameter) in monkeys (35). This result demonstrates that this formulation stimulates not only endochondral bone formation, as shown here, but also intramembranous bone formation, because the skull bone is formed without chondrogenesis. These findings support our hypothesis on the mechanism of action of FGF-2: that the target cells of FGF-2 are immature mesenchymal cells that have not differentiated into cells of chondroblastic or osteoblastic lineage. The stimulatory effect of FGF-2 on the fracture healing of a long bone of nonhuman primates through endochondral bone formation was also reported by Radomsky et al. (36). They showed that a single local injection of FGF-2 in a hyaluronan gel, an extracellular matrix component, significantly promoted fracture healing of the fibulae of baboons, as evidenced by increased callus formation and mechanical strength. As the fibula is much less weight-bearing than the ulna even without casting or other immobilization, most fractures (40 of 44) healed spontaneously if left untreated. However, the aim of the present study was to examine the effect of FGF-2 on fracture healing that is difficult to achieve spontaneously. We therefore intended to make a model for delayed union or nonunion, and neither external fixation nor any restriction of activity was imposed on the animals after the operation. In fact, a higher frequency of nonunion (4 of 10 fractures) was seen in vehicle-treated animals, and FGF-2 completely prevented the nonunion. In addition, the amounts of FGF-2 applied in the previous study were 400, 1000, and 3000 µg/site, with lack of a doseresponse effect (36). Because application of a lower dose of FGF-2 (200 µg/site) showed a potent anabolic effect in the present study, there may be a threshold amount of FGF-2 that will augment the repair process. Taking these reports together, it is suggested that FGF-2 in an order of magnitude (100–1000 µg/site) can stimulate endochondral and intramembranous bone formation and accelerate fracture healing even under severe conditions in nonhuman primates.
The powerful anabolic agents for bone formation include not only intrinsic peptide growth factors such as FGF-2 and BMPs, but also synthetic agents. Among BMPs, recombinant human BMP-2 and BMP-7 (osteogenic protein-1) have been reported to be useful clinically for the treatment of bone defects and spinal fusion (47, 48). However, the mechanism of action of BMPs appears to be different from that of FGF-2, in that the former act as a differentiation factor for precursor cells. The statins, drugs used for lowering serum cholesterol, have recently been attracting attention as representative anabolic drugs among synthetic agents (49). Because their anabolic action is thought to be mediated by BMP-2, they may also work as differentiation factors. The sequential application of FGF-2 at the earlier stage and of these differentiation factors at the later stage might have potential as the optimal method of inducing bone formation.
Our results demonstrate that a single local application of recombinant human FGF-2 in gelatin hydrogel accelerated fracture healing in nonhuman primates as well as in lower animals, suggesting the clinical usefulness of this agent for the treatment of fracture to prevent delayed union or nonunion. Because this formulation is viscid but can be injected into the fracture site easily, it not only can be applied to the fracture site after open reduction and internal fixation during surgery but can be percutaneously injected under roentgenological guidance at the time of closed reduction. A multicenter clinical study is now underway.
Received July 11, 2000.
Revised August 24, 2000.
Revised October 24, 2000.
Accepted October 26, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. J. Seeherman, K. Azari, S. Bidic, L. Rogers, X. J. Li, J. O. Hollinger, and J. M. Wozney rhBMP-2 Delivered in a Calcium Phosphate Cement Accelerates Bridging of Critical-Sized Defects in Rabbit Radii J. Bone Joint Surg. Am., July 1, 2006; 88(7): 1553 - 1565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. R. W. Simpson, L. Mills, and B. Noble The role of growth factors and related agents in accelerating fracture healing J Bone Joint Surg Br, June 1, 2006; 88-B(6): 701 - 705. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nakae, H. Kamiya, K. Naruse, N. Horio, Y. Ito, R. Mizubayashi, Y. Hamada, E. Nakashima, N. Akiyama, Y. Kobayashi, et al. Effects of basic fibroblast growth factor on experimental diabetic neuropathy in rats. Diabetes, May 1, 2006; 55(5): 1470 - 1477. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Seeherman, M. Bouxsein, H. Kim, R. Li, X. J. Li, M. Aiolova, and J. M. Wozney Recombinant Human Bone Morphogenetic Protein-2 Delivered in an Injectable Calcium Phosphate Paste Accelerates Osteotomy-Site Healing in a Nonhuman Primate Model J. Bone Joint Surg. Am., September 1, 2004; 86(9): 1961 - 1972. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Einhorn, R. J Majeska, A. Mohaideen, E. M. Kagel, M. L. Bouxsein, T. J. Turek, and J. M. Wozney A Single Percutaneous Injection of Recombinant Human Bone Morphogenetic Protein-2 Accelerates Fracture Repair J. Bone Joint Surg. Am., August 1, 2003; 85(8): 1425 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Seeherman, R. Li, and J. Wozney A Review of Preclinical Program Development for Evaluating Injectable Carriers for Osteogenic Factors J. Bone Joint Surg. Am., August 1, 2003; 85(90003): 96 - 108. [Full Text] [PDF]
|
|
|
|
 |
|
 F. Liu, L. Malaval, and J. E. Aubin Global amplification polymerase chain reaction reveals novel transitional stages during osteoprogenitor differentiation J. Cell Sci., May 1, 2003; 116(9): 1787 - 1796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Shimoaka, T. Ogasawara, A. Yonamine, D. Chikazu, H. Kawano, K. Nakamura, N. Itoh, and H. Kawaguchi Regulation of Osteoblast, Chondrocyte, and Osteoclast Functions by Fibroblast Growth Factor (FGF)-18 in Comparison with FGF-2 and FGF-10 J. Biol. Chem., February 22, 2002; 277(9): 7493 - 7500. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
