- Email: [email protected]
Sustained expression of transforming growth factor-β1 by distraction during distraction osteogenesis
Life Sciences 71 (2002) 67 – 79 www.elsevier.com/locate/lifescie
Sustained expression of transforming growth factor-h1 by distraction during distraction osteogenesis Hiu-Yan Yeung a, Kwong-Man Lee b, Kwok-Pui Fung c, Kwok-Sui Leung a,* a
Department of Orthopaedics and Traumatology, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China b Lee Hysan Clinical Research Laboratory, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China c Department of Biochemistry, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China Received 29 March 2001; accepted 17 December 2001
Abstract Distraction osteogenesis is a well-established clinical treatment for limb length discrepancy and skeletal deformities. In our previous studies, we have shown that the tension at the distraction gap correlated with the plasma bone specific alkaline phosphatase activity during distraction. Transforming growth factor-h1 (TGF-h1) has been shown to have a regulatory role in alkaline phosphatase activity during fracture healing. This study is to investigate the expression of TGF-h1 during distraction as a biological response to mechanically stimulated osteoblastic activity by immunohistochemistry. The expression of TGF-h1 in the distraction callus was compared with that in the fracture callus. During the distraction phase, the osteoblasts and osteocytes expressed a high level of TGF-h1. Moderate expression of TGF-h1 was observed in fibroblast-like cells in the fibrous zone of the distraction callus. After the distraction stopped, the expression of TGF-h1 in different cell types decreased. In fracture healing, the strong expression of TGF-h1 declined after the first week. Our results showed that the mechanical force induced and sustained TGF-h1 expression in osteoblasts and fibroblasts-like cells of the distraction callus. Transforming growth factor-h1 may play a role in transducing mechanical stimulation to biological tissue during in distraction osteogenesis. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Distraction Osteogenesis; Mechanical Stimulation; Transforming Growth Factor-h1
*
Corresponding author. Tel.: +852-2632-2724; fax: +852-2637-7889. E-mail address: [email protected] (K.-S. Leung). 0024-3205/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 4 - 3 2 0 5 ( 0 2 ) 0 1 5 7 5 - 8
68
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Introduction Distraction osteogenesis is a well-established clinical treatment for limb length discrepancy and skeletal deformities. This technique induces new bone formation at the osteotomy site by continuous distraction. In our previous studies [1–3], we have demonstrated that the tension in the newly formed osseous tissue increased gradually with distraction. The increase in tension correlated with the increase in the plasma activity of bone specific alkaline phosphatase in distraction osteogenesis. However, very little is known about the molecular basis of the increased bone formation activity by mechanical stimulation in the distraction model. We speculate that the mechanical stimulation is translated to a biological stimulation that activates the osteoblastic activity. Transforming growth factor-h1 is present in bone and platelets almost 100 times more than in any other tissues. Osteoblasts were shown to bear the highest amount of TGF-h1 receptors [4,5]. The initial in vivo studies demonstrating stimulatory effects of TGF-h1 on bone formation used injections of TGFh1 into rat and mouse calvaria and revealed a marked increase in bone formation [6–8]. When TGF-h1 was applied continuously to a healing osteotomy in rabbits, it resulted in increased callus formation and increased maximal bending strength of the osteotomy [9]. Transforming growth factor-h1 also stimulated chondrocyte proliferation and matrix synthesis [10,11]. The expression of TGF-h1 in bone and cartilage of the fracture callus and growth plate suggests an autocrine or paracrine role of the TGFh1 in osteogenesis and chondrogenesis [12,13]. Under cyclic mechanical stimulation, the human osteoblast culture increased the production of active TGF-h1 and cell proliferation [14]. In the present study, we investigated the responsible cells in the increase of TGF-h1 expression during distraction. By comparing the two regeneration processes in osseous tissues (fracture healing and distraction osteogenesis), we may understand better the in vivo effect of external mechanical stimulation on the biology of tissue regeneration.
Methods Animal model Forty-eight adolescent Chinese mountain goats were used in this study (Body weight: 27 F 3 kg). The operation was done under general anesthesia. Osteotomy and distraction protocol were done according to the method described previously [2]. Briefly, four transosseous Steinmann pins were inserted to the left tibia. A sub-periosteal osteotomy was made between the second and third pins. The periosteum was then sutured, followed by skin closure. The external fixator was then assembled carefully to make sure the osteotomy site was well aligned. After osteotomy, analgesics were given for 2 consecutive days. Distraction was started seven days after osteotomy. The distraction was done at the rate of 1 mm per day, 6 days per week. Distraction was continued for four weeks. For the welfare of the animal, the distraction was done once per day to minimize the disturbance. The animals in the fracture healing model were not distracted and the osteotomy site healed spontaneously. A weekly radiograph was taken to monitor the progress of both regeneration processes. The surgical procedures and distraction protocols were approved by the University Animal Research Ethics Committee. The staff in the University Laboratory Animal Service Center took care of all the animals in this study.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
69
Tissue sample collection Tissue samples of both regeneration processes were harvested at the identical time point for comparison. The surgically treated goats were divided into 6 sub-groups randomly. One group was euthanized on the last day of the latency phase (week 1 post-surgery; n = 4). Three groups were assigned during the distraction phase (weeks 2, 3, and 5 post-surgery; n = 4 in each group). During the consolidation phase, two groups of samples were obtained (week 8 and 12 post-surgery; n = 4 in each group). The animals were euthanized with an over-dosage of pentobarbital. As with the distraction group, the fracture healing control group was divided into six sub-groups. The distracted callus and fracture callus samples underwent decalcification in 9% formic acid/ formalin solution. The specimens were processed in graded alcohols and embedded in paraffin. 7 Am sections were obtained for haematoxylin and eosin staining and immunohistochemistry of TGF-h1. Immunohistochemistry Localization of TGF-h1 was performed using the indirect immunoperoxidase system. All paraffin sections were dewaxed by two changes of xylene and rehydrated through a graded series of aqueous ethanol solution. To quench the endogenous peroxidase, sections were treated with 3% hydrogen peroxide solution for 10 minutes. After being washed in phosphate buffered saline (pH 7.4), the sections were digested briefly with trypsin (0.25 Ag/mL; Gibco BRL, Gaitherburg, MD) and hyaluronidase (5 mg/mL; Sigma, St. Louis, MO) at 37 jC for 5 minutes and 7 minutes, respectively. Then, the nonspecific binding sites on the sections were blocked by 20% goat serum (Sigma, St. Louis, MO) in 0.5% bovine serum albumin (Sigma, St. Louis, MO) in phosphate buffered saline for 20 minutes. Then the sections were incubated with rabbit anti-Human TGF-h1 primary antibody (1:800; R&D System, Minneapolis, MN) diluted in 0.5% bovine serum albumin in phosphate buffered saline at 4 jC overnight in a humid chamber. After being washed three times with phosphate buffered saline, the sections were incubated with the biotinylated secondary antibody, rabbit anti-chick Immunoglobulin Type (1:100; Chemicon) for 30 minutes at room temperature. After being washed again with phosphate buffered saline, the sections were incubated with streptavidin-biotin-peroxidase complex (1:100) included in the DAKO Duet kit (DAKO, Glostrup, Denmark) for 30 minutes at room temperature. After another washing with phosphate buffered saline, the sections were reacted with diaminobenzidine solution containing hydrogen peroxide (DAB Chromogen Kit, DAKO, Glostrup, Denmark) in phosphate buffered saline. The reaction was stopped by washing with phosphate buffered saline and the sections were counterstained with Mayer’s hematoxylin for 1 minute before permanent organic mounting by DPX Mountant (Fluka, Milwaukee, WI). For negative control, 0.5% bovine serum albumin in phosphate buffered saline was used instead of the primary antibody. Evaluation of immunohistochemical staining results Slides were examined using a Leica DMRB microscope (Wetzlar, Germany) Immunostained slides were evaluated in a blinded fashion by three individuals, independently. The intensity of positive immunostaining was graded as +++, ++, +, for strong, moderate, weak, and negative, respectively. A
70
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
grade of +/ was used to represent a focal or questionable weakly positive cell in the tissue sections. Different types of cells were identified morphologically.
Results Animal model The goats in both models underwent distraction and external fixation without developing any major health problems. There was no non-union at the fracture site. Fig. 1 showed the typical radiographs of distraction osteogenesis and fracture healing in this study. From the radiographs of the distraction osteogenesis, the typical radiolucent zone and the faint radiodense columns near the osteotomy sites were observed during distraction. From the radiographs of fracture healing, a mineralized fracture callus was observed at the osteotomy site. Histology of distraction osteogenesis compared with fracture healing The histology of distraction osteogenesis and fracture healing is shown in Figs. 2 and 3. In the latency phase of distraction osteogenesis, the histology was the same as that in fracture healing. The
Fig. 1. Weekly radiograph follow-up was performed to monitor the mineralization of distraction osteogenesis and fracture healing. (a) The callus has been distracted for 4 weeks (Week 5 post-osteotomy). Mineralization of the callus was observed from the ends of the cortical bone. (b) The distraction callus was consolidated for 7 weeks. The callus was completely mineralized. (c) At week 5 post-fracture, mineralized fracture callus was observed around the fracture site. (d) At week 12 post-fracture, the fracture callus has been remodeled. The new cortex can be observed.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
71
Fig. 2. Histology of fracture healing. (a) At week 1 post-osteotomy, the fracture site (#) was filled with fibrous clot (FC). The newly formed woven bone (WB) was developed around the fractured cortex (C). FT: Fibrous Tissue. (Hematoxylin & Eosin, 38) (b) At week 2 post-osteotomy, cartilaginous tissue (CT) was observed at the middle of the fracture callus. The woven bone was formed by endochondral ossification (arrows). (Hematoxylin & Eosin, 38) (c) At week 5 post-osteotomy, the middle cartilaginous tissue was almost replaced by woven bone. (Hematoxylin & Eosin, 76) (d) At 12 week post-osteotomy, the fracture site has been filled with woven bone and the callus was under remodeling (Hematoxylin & Eosin, 18).
periosteum expanded, and a fibrous clot and hematoma filled the osteotomy site. A thin layer of newly formed woven bone was observed around the cortical bone. The fibrous tissue contained spindle-shaped fibroblast-like cells (Fig. 2a). Once the distraction started, the fibrous tissue at the distraction site started to expand as the gap length increased (Fig. 3a). During the first week of distraction (week 2 post-surgery), the fibroblastlike cells orientated randomly. After the callus was distracted for 2 weeks (week 3 post-surgery), the fibroblast-like cells aligned with the direction of the distraction force (Fig. 3c). The woven bone formed at the distraction gap was observed and originated from the bone around the cortical ends during the latency phase. As the distraction continued, the newly formed woven bone continued to advance toward the center of the distraction callus. The new bone was well aligned with the distraction force (Fig. 3b). Microcolumns of woven bone were observed at the bone formation front (Fig. 3c). Unlike the distraction model, the fracture healing model showed that chondrogenesis right at the fracture site started at week 2 and 3 post-surgery (Fig. 2b). The initial fibrous tissue was replaced by cartilaginous tissue. The mode of bone formation was through endochondral ossifica-
72
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 3. Histology of distraction osteogenesis. (a) During the first week of distraction (week 2 post-osteotomy), the fibrous tissue (FT) from the periosteum filled the distraction gap. The woven bone (WB) formed during the latency period was advancing toward the center of the distraction callus. (C:Cortical bone; Hematoxylin & Eosin, 38) (b) After 4 weeks of distraction, microcolumn of new woven bone (arrows) was formed along the distraction force. (Hematoxylin & Eosin, 38) (c) At the tips of the microcolumn, the bone formation front (BFF) was observed. The cells at the fibrous tissue was well aligned to the direction of distraction force (arrow). However, at the BFF, the cells had a different orientation. (Hematoxylin & Eosin, 152) (d) After the distraction stopped for 7 weeks (week 12 post-osteotomy), the distraction callus was undergone remodeling. The large resorption cavity (RC) was observed (Hematoxylin & Eosin, 38).
tion. At week 5 post-surgery, the fracture callus was almost completely filled with woven bone and the endochondral ossification was completed (Fig. 2c). The observation above shows that the distraction force may induce the fibroblast-like cells derived from the periosteum to undergo osteogenesis rather than chondrogenesis. Three weeks after the distraction stopped (week 8 post-surgery), the distraction callus was in the consolidation phase. The distraction gap was filled with woven bone. Osteoblasts covered the newly woven bone. At the center of the distraction callus, cartilaginous tissue was observed. The callus had undergone endochondral ossification after the distraction stopped. This observation was similar to the histology of fracture healing. The fibrous tissue at the middle of the callus underwent chondrogenesis and endochondral ossification to form bone. At the same time point as in fracture healing, the fracture callus was in the remodeling phase. A large resorption cavity and fewer osteoblasts were observed in the fracture callus. The woven bone was covered with bone lining
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
73
cells. At week 12 post-surgery, the fracture gap had been bridged with new bone and the cortical ends had been remodeled to incorporate the new bone at the fracture site (Fig. 2d). The intramedullary callus of the fracture healing was almost completely resorbed. By contrast with the fracture healing group, the distraction callus in the distraction group was in the remodeling phase. Large resorptive cavities were observed in the distraction callus. The cartilaginous tissue observed at the early stage of consolidation was completely replaced with woven bone 7 weeks after distraction stopped (Fig. 3d). In the consolidation phase of distraction osteogenesis, the fibrous tissue at the middle of the distraction callus was no longer stimulated by distraction force. The histology was shown to be similar to that in fracture healing. Expression of transforming growth factor-b1 in distraction osteogenesis In Table 1, the change in transforming growth factor-h1 expression in different cell types in the distraction callus is compared with that in fracture healing. Briefly, the histology of the latency of distraction osteogenesis was the same as that for fracture healing at week 1 post-surgery. The expression of the TGF-h1 in the two regeneration processes was also the same as shown in Fig. 4a and b. Strong staining for TGF-h1 expression was observed at the osteoblasts, pre-osteoblasts, and fibroblast-like cells. The periosteal cells were stained moderately. During the distraction phase (from week 2 to week 5 post-surgery), osteoblasts and osteocytes in the newly formed distraction callus demonstrated strong staining for TGF-h1 (Fig. 5a and b). Moderate expression of TGF-h1 was observed in fibroblast-like cells in the fibrous zone of the distraction callus (Fig. 5c). The periosteal cells exhibited weak expression of TGF-h1 during the
Table 1 Expression of transforming growth factor-h1 in distraction osteogenesis Distraction osteogenesis Latency Phase Distraction Phase Consolidation Phase Fracture healing
Cell type in distraction callus Post-surgery
Osteoblasts
Osteocytes
Periosteal cells
Fibroblast-like cells
Chondrocytes
Week Week Week Week Week Week
+++ +++ +++ +++ ++/ ++/
ne +++ +++ ++ + +
++ + + + +/ +/
+++ ++ ++ ++ ne ne
ne ne ne ne +++ ne
1 2 3 5 8 12
Cell type in fracture callus Post-surgery
Osteoblasts
Osteocytes
Periosteal cells
Fibroblast-like cells
Chondrocytes
Week Week Week Week Week Week
+++ ++ ++ + + ++/
ne ++ ++ + + +
++ ++ ++ + + +/
+++ ++ + +/ ne ne
ne +++ +++ ne ne ne
1 2 3 5 8 12
+++ :Strong staining, ++ :Moderate staining, + :Weak staining, +/ :Focal weak staining, ++/ staining, ne :Non Existent.
:Focal moderate
74
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 4. Immunohistochemistry of TGF-h1 in fracture healing model. (a) At week 1 post-osteotomy, the fibroblast-like cells (arrows) in the fibrous tissue were stained for TGF-h1 (304). (b) At the newly formed woven bone around the cortex, the osteoblasts were also shown to express TGF-h1. (304) (c) At week 2 post-osteotomy, the callus was formed around the fracture site. The osteoblasts on the woven bone (WB) continued to express TGF-h1. The periosteal cells were also shown to express TGF-h1. (76) (d) At week 3 post-osteotomy, cartilaginous tissue and endochondral ossification were observed at the middle of the fracture callus. At the edge of the cartilaginous tissue (arrows) where endochondral ossification occurred, the expression of TGF-h1 was shown (76).
whole distraction phase (Fig. 5d). At the end of distraction phase, microcolumns of new bone were formed at the distraction callus. The osteoblasts covering the new microcolumns were shown to have strongest TGF-h1 expression (Fig. 5e). During the corresponding phase of fracture healing, the osteoblasts, osteocytes, and the periosteal cells showed moderate staining for TGF-h1 and reduced to a weak expression at week 5 post-surgery (Fig. 4c). Endochondral ossification was observed in the fracture callus at the week 3 post-osteotomy. TGF-h1 was present at the edge of the cartilaginous tissue (Fig. 4d). The staining for TGF-h1 at the fibroblast-like cells showed a weak expression at week 3 post-surgery from a strong signal at week 1 post-surgery. In the consolidation phase of distraction osteogenesis (week 8 to 12 post-operation), although not all the osteoblasts stained for TGF-h1, some showed a moderate signal (Fig. 6a). This moderate staining for TGF-h1 was present throughout the consolidation phase (week 8 to 12) (Fig 6b). At week 8 post-surgery, the chondrocytes in the cartilaginous tissue stained strongly for TGF-h1 (Fig. 6c). The osteocytes in the distraction callus exhibited weak expression of TGF-h1. In fracture
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
75
Fig. 5. Immunohistochemistry of TGF-h1 in distraction osteogenesis. (a) During the distraction period (week 3 post-osteotomy), the fibroblast-like cells (arrows) in the fibrous tissue of the distraction callus were shown to have TGF-h1 expression (304). (b) The osteoblasts (Ob) in the newly formed bone were shown to have TGF-h1 expression (304). (c) Microcolumns of new bone were formed along the direction of the distraction force. The osteoblasts on the new microcolumns of bone had strong TGF-h1 expression (152). (d) At the last week of distraction (week 5 post-osteotomy), the periosteum was expanded and the cells (arrows) were shown to have TGF-h1. (304) (e) At the mature trabecular bone of the distraction callus, the osteocytes (Oc) and osteoblasts (Ob) were stained for TGF-h1 (304). (f) The negative control omitting the primary antibody is shown (152).
76
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 6. Immunohistochemistry of TGF-h1 in consolidation period of distraction osteogenesis. (a) After the distraction stopped for 3 weeks, cartilaginous tissue was formed at the middle of the distraction callus. The chondrocytes in the cartilaginous tissue were stained for TGF-h1 (304). (b) The osteoblasts in the maturing distraction callus continued to be stained for TGF-h1 (304). (c) After the distraction stopped for 7 weeks (week 12 post-osteotomy), the lining cells (LC) and osteocytes (Oc) at the trabecular bone were weakly stained for TGF-h1 (304). (d) The osteoblasts were stained moderately for TGF-h1 at week 12 post-osteotomy (304).
healing, the fracture callus mainly consisted of osteoblasts, periosteal cells, and osteocytes. They were stained weakly for TGF-h1.
Discussion The mechanism of the biological response of osseous tissue to external mechanical stimulation in distraction osteogenesis has not been fully understood. In our previous related research, we have shown that the increase in the plasma activity of bone specific alkaline phosphatase was related to the distraction force applied to the callus [1,2]. The plasma activity of the bone specific alkaline phosphatase remained elevated as long as the tension within the callus was maintained by continuous distraction. The mechanism of activation of the enzymatic activity by mechanical stimulation may be related to the local factors present in the distraction callus. In this study, the expression of the TGF-h1 in the bone forming cells was increased throughout the distraction phase. The external mechanical stimulation induced the expression of the local growth factors in the cells involved in new bone formation.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
77
The immunohistochemical result of TGF-h1 in fracture healing was similar to that in other previous studies [6,7,16,17]. In addition to the localization of TGF-h1 at different cells within the fracture callus, we followed the temporal changes of TGF-h1 expression. Strong expression of TGF-h1 of fibroblastlike cells derived from periosteum and osteoblasts was observed at the initial stages of fracture repair. Subsequently, the signal of TGF-h1 in various cells declined gradually. Along with the changes in histology, this indicates that a specific concentration of the TGF-h1 may trigger a cascade events for chondrogenesis and osteogenesis in fracture healing [18,19]. Our present fracture healing serves a good control model for us to compare with distraction osteogenesis model. The osteoblasts in the distraction callus were shown to have strong TGF-h1 expression during the distraction phase. When compared with TGF-h1 expression in fracture healing, the strong expression was sustained as long as the callus was distracted. In other studies, the osteoblast-like cells increased the production of TGF-h1 when they were under a stretching force [14,20]. Lammens et al. [15] also showed that the tissue level of TGF-h1 was increased in the distraction callus as long as distraction applied. The present study further identifies the cellular components in the expression of TGF-h1. It confirms that the distraction has a stimulatory effect on the expression of TGF-h1 in osteoblasts during distraction osteogenesis. It is possible that mechanical force in the distraction osteogenesis model acts as a transduction signal to the osteoblasts that react with an increase in the expression of TGF-h1. The increased expression of TGF-h1 in the distraction callus acts as either autocrine or paracrine to stimulate the osteoblasts to synthesize collagen and alkaline phosphatase for bone formation and mineralization. This may explain our previous finding of the sustained elevation of the plasma bone specific alkaline phosphatase during distraction [1,2]. The hypothesis is further supported by other studies showing that TGF-h1 is a stimulant causing osteoblasts to increase the synthesis of collagen [19,21]. From our observation, increase of TGF-h1 expression was not only due to the osteoblasts, but also the fibroblastlike cells and the osteocytes in the distraction callus. The fibroblast-like cells in the distraction callus maintained a moderate expression of TGF-h1 in distraction phase. This temporal change was different from that in fracture healing. In fracture healing, the expression decreased gradually from the first week of fracture repair and became weak by week 3 post-surgery. The distraction force in the distraction model might induce the fibroblastlike cells in the fibrous zone of distraction callus to express a moderate amount of TGF-h1 protein. Together with other growth factor expression [22,23], the cells may differentiate directly to osteoblasts for bone formation instead of chondrogenesis, as in fracture healing. Therefore, bone was formed at the distraction gap by membraneous ossification. In fracture healing, the decrease of TGF-h1 expression might direct the fibroblast-like cells derived from the periosteum to differentiate to chondrocyte lineage, triggering the cascade of endochondral ossification. Thus, endochondral ossification was observed in the fracture healing model. In many in vitro studies, the effect of TGF-h1 on the differentiation of mesenchymal cells has been shown to be dose dependent [8,11,18,19,24–26]. Once distraction stopped, the distraction callus no longer experienced the distraction force and expression of TGF-h1 decreased. Transforming growth factor-h1 expression dropped to a low level at which the fibroblast-like cells were triggered to differentiate to chondrogenesis. Hence, endochondral ossification was observed in the consolidation phase of the distraction gap. This observation explains the fact that different modes of bone formation were observed during distraction osteogenesis. During the distraction phase, the osteocytes in the distraction callus were also shown to have strong expression of transforming growth factor-h1. Once the distraction stopped, the expression of trans-
78
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
forming growth factor-h1 declined. Osteocytes have long been thought to be the biomechanosensor in osseous tissue [27]. Osteocytes communicate with other bone cells to trigger a cascade of biological events in bone formation when an external mechanical force is applied to osseous tissue. Therefore, the observation of increased expression of transforming growth factor-h1 in the osteocytes may indicate that osteocytes in the distraction callus were also activated and modulated osteogenesis [28]. In this study, the sustained expression of TGF-h1 in various types of cells under mechanical stimulation was demonstrated in the distraction osteogenesis model. Different types of cells responded differently to the mechanical force. As a multi-functional growth factor, TGF-h1 is an important mediator in cell differentiation and cellular activity of the distraction osteogenesis to translate the external mechanical force into a biological response. Acknowledgements This project was funded by Research Grant Committee (CUHK 431/95M), Hong Kong, China.
References [1] Leung KS, Fung KP, Sher AH, Li CK, Lee KM. Plasma bone-specific alkaline phosphatase as an indicator of osteoblastic activity. Journal of Bone & Joint Surgery — British Volume 1993;75:288 – 92. [2] Leung KS, Lee KM, Chan CW, Mak A, Fung KP. Mechanical characterization of regenerated osseous tissue during callotasis and its related biological phenomenon. Life Sciences 2000;66:327 – 36. [3] Leung KS, Fung KP, Liu PP, Lee KM. Bone-specific alkaline phosphatase activities in plasma and callus during callotasis in rabbits. Life Sciences 1995;57:637 – 43. [4] Robey PG, Young MF, Flanders KC, et al. Osteoblasts synthesize and respond to transforming growth factor-type beta (TGF-beta) in vitro. Journal of Cell Biology 1987;105:457 – 63. [5] Sporn MB, Roberts AB. The multifunctional nature of growth factors. In: Sporn MB, Roberts AB, editors. Handbook of experimental pharmacology: Peptide growth factors and their receptors. New York: Springer-Verlag, 1990. pp. 3 – 15. [6] Joyce ME, Jingushi S, Bolander ME. Transforming growth factor-beta in the regulation of fracture repair. Orthopedic Clinics of North America 1990;21:199 – 209. [7] Joyce ME, Terek RM, Jingushi S, Bolander ME. Role of transforming growth factor-beta in fracture repair. Annals of the New York Academy of Sciences 1990;593:107 – 23. [8] Noda M, Camilliere JJ. In vivo stimulation of bone formation by transforming growth factor-beta. Endocrinology 1989;124:2991 – 4. [9] Lind M, Schumacker B, Soballe K, Keller J, Melsen F, Bunger C. Transforming growth factor-beta enhances fracture healing in rabbit tibiae. Acta Orthopaedica Scandinavica 1993;64:553 – 6. [10] Bolander ME. Regulation of fracture repair by growth factors. Proceedings of the Society for Experimental Biology & Medicine 1992;200:165 – 70. [11] Joyce ME, Roberts AB, Sporn MB, Bolander ME. Transforming growth factor-beta and the initiation of chondrogenesis and osteogenesis in the rat femur. Journal of Cell Biology 1990;110:2195 – 207. [12] Gelb DE, Rosier RN, Puzas JE. The production of transforming growth factor-beta by chick growth plate chondrocytes in short term monolayer culture. Endocrinology 1990;127:1941 – 7. [13] Thorp BH, Anderson I, Jakowlew SB. Transforming growth factor-beta 1, -beta 2 and -beta 3 in cartilage and bone cells during endochondral ossification in the chick. Development 1992;114:907 – 11. [14] Holbein O, Neidlinger-Wilke C, Suger G, Kinzl L, Claes L. Ilizarov callus distraction produces systemic bone cell mitogens. Journal of Orthopaedic Research 1995;13:629 – 38. [15] Lammens J, Liu Z, Aerssens J, Dequeker J, Fabry G. Distraction bone healing versus osteotomy healing: a comparative biochemical analysis. Journal of Bone & Mineral Research 1998;13:279 – 86.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
79
[16] Bourque WT, Gross M, Hall BK. Expression of four growth factors during fracture repair. International Journal of Developmental Biology 1993;37:573 – 9. [17] Matsumoto K, Matsunaga S, Imamura T, Ishidou Y, Yoshida H, Sakou T. Expression and distribution of transforming growth factor-beta and decorin during fracture healing. In Vivo 1994;8:215 – 9. [18] Iwasaki M, Nakata K, Nakahara H, et al. Transforming growth factor-beta 1 stimulates chondrogenesis and inhibits osteogenesis in high density culture of periosteum-derived cells. Endocrinology 1993;132:1603 – 8. [19] Iwasaki M, Nakahara H, Nakata K, Nakase T, Kimura T, Ono K. Regulation of proliferation and osteochondrogenic differentiation of periosteum-derived cells by transforming growth factor-beta and basic fibroblast growth factor. Journal of Bone & Joint Surgery - American Volume 1995;77:543 – 54. [20] Zhuang H, Wang W, Tahernia AD, Levitz CL, Luchetti WT, Brighton CT. Mechanical strain-induced proliferation of osteoblastic cells parallels increased TGF-beta 1 mRNA. Biochemical & Biophysical Research Communications 1996;229:449 – 53. [21] Terrell TG, Working PK, Chow CP, Green JD. Pathology of recombinant human transforming growth factor-beta 1 in rats and rabbits. International Review of Experimental Pathology 1993;34(Pt B):43 – 67. [22] Sato M, Ochi T, Nakase T, et al. Mechanical tension-stress induces expression of bone morphogenetic protein (BMP)-2 and BMP-4, but not BMP-6, BMP-7, and GDF-5 mRNA, during distraction osteogenesis. Journal of Bone & Mineral Research 1999;14:1084 – 95. [23] Yeung HY, Lee SK, Fung KP, Leung KS. Expression of basic fibroblast growth factor during distraction osteogenesis. Clinical Orthopaedics & Related Research 2001;219 – 29. [24] Izumi T, Scully SP, Heydemann A, Bolander ME. Transforming growth factor beta 1 stimulates type II collagen expression in cultured periosteum-derived cells. Journal of Bone & Mineral Research 1992;7:115 – 21. [25] Miura M, Sakagishi Y, Hata K, Komoda T. Differences between the sugar moieties of liver- and bone-type alkaline phosphatases: a re-evaluation. Annals of Clinical Biochemistry 1994;31:25 – 30. [26] Basic N, Basic V, Bulic K, et al. TGF-beta and basement membrane matrigel stimulate the chondrogenic phenotype in osteoblastic cells derived from fetal rat calvaria. Journal of Bone & Mineral Research 1996;11:384 – 91. [27] Mikuni-Takagaki Y, Suzuki Y, Kawase T, Saito S. Distinct responses of different populations of bone cells to mechanical stress. Endocrinology 1996;137:2028 – 35. [28] Lean JM, Mackay AG, Chow JW, Chambers TJ. Osteocytic expression of mRNA for c-fos and IGF-I: an immediate early gene response to an osteogenic stimulus. American Journal of Physiology 1996;270:E937 – 45.
Sustained expression of transforming growth factor-h1 by distraction during distraction osteogenesis Hiu-Yan Yeung a, Kwong-Man Lee b, Kwok-Pui Fung c, Kwok-Sui Leung a,* a
Department of Orthopaedics and Traumatology, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China b Lee Hysan Clinical Research Laboratory, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China c Department of Biochemistry, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China Received 29 March 2001; accepted 17 December 2001
Abstract Distraction osteogenesis is a well-established clinical treatment for limb length discrepancy and skeletal deformities. In our previous studies, we have shown that the tension at the distraction gap correlated with the plasma bone specific alkaline phosphatase activity during distraction. Transforming growth factor-h1 (TGF-h1) has been shown to have a regulatory role in alkaline phosphatase activity during fracture healing. This study is to investigate the expression of TGF-h1 during distraction as a biological response to mechanically stimulated osteoblastic activity by immunohistochemistry. The expression of TGF-h1 in the distraction callus was compared with that in the fracture callus. During the distraction phase, the osteoblasts and osteocytes expressed a high level of TGF-h1. Moderate expression of TGF-h1 was observed in fibroblast-like cells in the fibrous zone of the distraction callus. After the distraction stopped, the expression of TGF-h1 in different cell types decreased. In fracture healing, the strong expression of TGF-h1 declined after the first week. Our results showed that the mechanical force induced and sustained TGF-h1 expression in osteoblasts and fibroblasts-like cells of the distraction callus. Transforming growth factor-h1 may play a role in transducing mechanical stimulation to biological tissue during in distraction osteogenesis. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Distraction Osteogenesis; Mechanical Stimulation; Transforming Growth Factor-h1
*
Corresponding author. Tel.: +852-2632-2724; fax: +852-2637-7889. E-mail address: [email protected] (K.-S. Leung). 0024-3205/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 4 - 3 2 0 5 ( 0 2 ) 0 1 5 7 5 - 8
68
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Introduction Distraction osteogenesis is a well-established clinical treatment for limb length discrepancy and skeletal deformities. This technique induces new bone formation at the osteotomy site by continuous distraction. In our previous studies [1–3], we have demonstrated that the tension in the newly formed osseous tissue increased gradually with distraction. The increase in tension correlated with the increase in the plasma activity of bone specific alkaline phosphatase in distraction osteogenesis. However, very little is known about the molecular basis of the increased bone formation activity by mechanical stimulation in the distraction model. We speculate that the mechanical stimulation is translated to a biological stimulation that activates the osteoblastic activity. Transforming growth factor-h1 is present in bone and platelets almost 100 times more than in any other tissues. Osteoblasts were shown to bear the highest amount of TGF-h1 receptors [4,5]. The initial in vivo studies demonstrating stimulatory effects of TGF-h1 on bone formation used injections of TGFh1 into rat and mouse calvaria and revealed a marked increase in bone formation [6–8]. When TGF-h1 was applied continuously to a healing osteotomy in rabbits, it resulted in increased callus formation and increased maximal bending strength of the osteotomy [9]. Transforming growth factor-h1 also stimulated chondrocyte proliferation and matrix synthesis [10,11]. The expression of TGF-h1 in bone and cartilage of the fracture callus and growth plate suggests an autocrine or paracrine role of the TGFh1 in osteogenesis and chondrogenesis [12,13]. Under cyclic mechanical stimulation, the human osteoblast culture increased the production of active TGF-h1 and cell proliferation [14]. In the present study, we investigated the responsible cells in the increase of TGF-h1 expression during distraction. By comparing the two regeneration processes in osseous tissues (fracture healing and distraction osteogenesis), we may understand better the in vivo effect of external mechanical stimulation on the biology of tissue regeneration.
Methods Animal model Forty-eight adolescent Chinese mountain goats were used in this study (Body weight: 27 F 3 kg). The operation was done under general anesthesia. Osteotomy and distraction protocol were done according to the method described previously [2]. Briefly, four transosseous Steinmann pins were inserted to the left tibia. A sub-periosteal osteotomy was made between the second and third pins. The periosteum was then sutured, followed by skin closure. The external fixator was then assembled carefully to make sure the osteotomy site was well aligned. After osteotomy, analgesics were given for 2 consecutive days. Distraction was started seven days after osteotomy. The distraction was done at the rate of 1 mm per day, 6 days per week. Distraction was continued for four weeks. For the welfare of the animal, the distraction was done once per day to minimize the disturbance. The animals in the fracture healing model were not distracted and the osteotomy site healed spontaneously. A weekly radiograph was taken to monitor the progress of both regeneration processes. The surgical procedures and distraction protocols were approved by the University Animal Research Ethics Committee. The staff in the University Laboratory Animal Service Center took care of all the animals in this study.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
69
Tissue sample collection Tissue samples of both regeneration processes were harvested at the identical time point for comparison. The surgically treated goats were divided into 6 sub-groups randomly. One group was euthanized on the last day of the latency phase (week 1 post-surgery; n = 4). Three groups were assigned during the distraction phase (weeks 2, 3, and 5 post-surgery; n = 4 in each group). During the consolidation phase, two groups of samples were obtained (week 8 and 12 post-surgery; n = 4 in each group). The animals were euthanized with an over-dosage of pentobarbital. As with the distraction group, the fracture healing control group was divided into six sub-groups. The distracted callus and fracture callus samples underwent decalcification in 9% formic acid/ formalin solution. The specimens were processed in graded alcohols and embedded in paraffin. 7 Am sections were obtained for haematoxylin and eosin staining and immunohistochemistry of TGF-h1. Immunohistochemistry Localization of TGF-h1 was performed using the indirect immunoperoxidase system. All paraffin sections were dewaxed by two changes of xylene and rehydrated through a graded series of aqueous ethanol solution. To quench the endogenous peroxidase, sections were treated with 3% hydrogen peroxide solution for 10 minutes. After being washed in phosphate buffered saline (pH 7.4), the sections were digested briefly with trypsin (0.25 Ag/mL; Gibco BRL, Gaitherburg, MD) and hyaluronidase (5 mg/mL; Sigma, St. Louis, MO) at 37 jC for 5 minutes and 7 minutes, respectively. Then, the nonspecific binding sites on the sections were blocked by 20% goat serum (Sigma, St. Louis, MO) in 0.5% bovine serum albumin (Sigma, St. Louis, MO) in phosphate buffered saline for 20 minutes. Then the sections were incubated with rabbit anti-Human TGF-h1 primary antibody (1:800; R&D System, Minneapolis, MN) diluted in 0.5% bovine serum albumin in phosphate buffered saline at 4 jC overnight in a humid chamber. After being washed three times with phosphate buffered saline, the sections were incubated with the biotinylated secondary antibody, rabbit anti-chick Immunoglobulin Type (1:100; Chemicon) for 30 minutes at room temperature. After being washed again with phosphate buffered saline, the sections were incubated with streptavidin-biotin-peroxidase complex (1:100) included in the DAKO Duet kit (DAKO, Glostrup, Denmark) for 30 minutes at room temperature. After another washing with phosphate buffered saline, the sections were reacted with diaminobenzidine solution containing hydrogen peroxide (DAB Chromogen Kit, DAKO, Glostrup, Denmark) in phosphate buffered saline. The reaction was stopped by washing with phosphate buffered saline and the sections were counterstained with Mayer’s hematoxylin for 1 minute before permanent organic mounting by DPX Mountant (Fluka, Milwaukee, WI). For negative control, 0.5% bovine serum albumin in phosphate buffered saline was used instead of the primary antibody. Evaluation of immunohistochemical staining results Slides were examined using a Leica DMRB microscope (Wetzlar, Germany) Immunostained slides were evaluated in a blinded fashion by three individuals, independently. The intensity of positive immunostaining was graded as +++, ++, +, for strong, moderate, weak, and negative, respectively. A
70
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
grade of +/ was used to represent a focal or questionable weakly positive cell in the tissue sections. Different types of cells were identified morphologically.
Results Animal model The goats in both models underwent distraction and external fixation without developing any major health problems. There was no non-union at the fracture site. Fig. 1 showed the typical radiographs of distraction osteogenesis and fracture healing in this study. From the radiographs of the distraction osteogenesis, the typical radiolucent zone and the faint radiodense columns near the osteotomy sites were observed during distraction. From the radiographs of fracture healing, a mineralized fracture callus was observed at the osteotomy site. Histology of distraction osteogenesis compared with fracture healing The histology of distraction osteogenesis and fracture healing is shown in Figs. 2 and 3. In the latency phase of distraction osteogenesis, the histology was the same as that in fracture healing. The
Fig. 1. Weekly radiograph follow-up was performed to monitor the mineralization of distraction osteogenesis and fracture healing. (a) The callus has been distracted for 4 weeks (Week 5 post-osteotomy). Mineralization of the callus was observed from the ends of the cortical bone. (b) The distraction callus was consolidated for 7 weeks. The callus was completely mineralized. (c) At week 5 post-fracture, mineralized fracture callus was observed around the fracture site. (d) At week 12 post-fracture, the fracture callus has been remodeled. The new cortex can be observed.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
71
Fig. 2. Histology of fracture healing. (a) At week 1 post-osteotomy, the fracture site (#) was filled with fibrous clot (FC). The newly formed woven bone (WB) was developed around the fractured cortex (C). FT: Fibrous Tissue. (Hematoxylin & Eosin, 38) (b) At week 2 post-osteotomy, cartilaginous tissue (CT) was observed at the middle of the fracture callus. The woven bone was formed by endochondral ossification (arrows). (Hematoxylin & Eosin, 38) (c) At week 5 post-osteotomy, the middle cartilaginous tissue was almost replaced by woven bone. (Hematoxylin & Eosin, 76) (d) At 12 week post-osteotomy, the fracture site has been filled with woven bone and the callus was under remodeling (Hematoxylin & Eosin, 18).
periosteum expanded, and a fibrous clot and hematoma filled the osteotomy site. A thin layer of newly formed woven bone was observed around the cortical bone. The fibrous tissue contained spindle-shaped fibroblast-like cells (Fig. 2a). Once the distraction started, the fibrous tissue at the distraction site started to expand as the gap length increased (Fig. 3a). During the first week of distraction (week 2 post-surgery), the fibroblastlike cells orientated randomly. After the callus was distracted for 2 weeks (week 3 post-surgery), the fibroblast-like cells aligned with the direction of the distraction force (Fig. 3c). The woven bone formed at the distraction gap was observed and originated from the bone around the cortical ends during the latency phase. As the distraction continued, the newly formed woven bone continued to advance toward the center of the distraction callus. The new bone was well aligned with the distraction force (Fig. 3b). Microcolumns of woven bone were observed at the bone formation front (Fig. 3c). Unlike the distraction model, the fracture healing model showed that chondrogenesis right at the fracture site started at week 2 and 3 post-surgery (Fig. 2b). The initial fibrous tissue was replaced by cartilaginous tissue. The mode of bone formation was through endochondral ossifica-
72
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 3. Histology of distraction osteogenesis. (a) During the first week of distraction (week 2 post-osteotomy), the fibrous tissue (FT) from the periosteum filled the distraction gap. The woven bone (WB) formed during the latency period was advancing toward the center of the distraction callus. (C:Cortical bone; Hematoxylin & Eosin, 38) (b) After 4 weeks of distraction, microcolumn of new woven bone (arrows) was formed along the distraction force. (Hematoxylin & Eosin, 38) (c) At the tips of the microcolumn, the bone formation front (BFF) was observed. The cells at the fibrous tissue was well aligned to the direction of distraction force (arrow). However, at the BFF, the cells had a different orientation. (Hematoxylin & Eosin, 152) (d) After the distraction stopped for 7 weeks (week 12 post-osteotomy), the distraction callus was undergone remodeling. The large resorption cavity (RC) was observed (Hematoxylin & Eosin, 38).
tion. At week 5 post-surgery, the fracture callus was almost completely filled with woven bone and the endochondral ossification was completed (Fig. 2c). The observation above shows that the distraction force may induce the fibroblast-like cells derived from the periosteum to undergo osteogenesis rather than chondrogenesis. Three weeks after the distraction stopped (week 8 post-surgery), the distraction callus was in the consolidation phase. The distraction gap was filled with woven bone. Osteoblasts covered the newly woven bone. At the center of the distraction callus, cartilaginous tissue was observed. The callus had undergone endochondral ossification after the distraction stopped. This observation was similar to the histology of fracture healing. The fibrous tissue at the middle of the callus underwent chondrogenesis and endochondral ossification to form bone. At the same time point as in fracture healing, the fracture callus was in the remodeling phase. A large resorption cavity and fewer osteoblasts were observed in the fracture callus. The woven bone was covered with bone lining
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
73
cells. At week 12 post-surgery, the fracture gap had been bridged with new bone and the cortical ends had been remodeled to incorporate the new bone at the fracture site (Fig. 2d). The intramedullary callus of the fracture healing was almost completely resorbed. By contrast with the fracture healing group, the distraction callus in the distraction group was in the remodeling phase. Large resorptive cavities were observed in the distraction callus. The cartilaginous tissue observed at the early stage of consolidation was completely replaced with woven bone 7 weeks after distraction stopped (Fig. 3d). In the consolidation phase of distraction osteogenesis, the fibrous tissue at the middle of the distraction callus was no longer stimulated by distraction force. The histology was shown to be similar to that in fracture healing. Expression of transforming growth factor-b1 in distraction osteogenesis In Table 1, the change in transforming growth factor-h1 expression in different cell types in the distraction callus is compared with that in fracture healing. Briefly, the histology of the latency of distraction osteogenesis was the same as that for fracture healing at week 1 post-surgery. The expression of the TGF-h1 in the two regeneration processes was also the same as shown in Fig. 4a and b. Strong staining for TGF-h1 expression was observed at the osteoblasts, pre-osteoblasts, and fibroblast-like cells. The periosteal cells were stained moderately. During the distraction phase (from week 2 to week 5 post-surgery), osteoblasts and osteocytes in the newly formed distraction callus demonstrated strong staining for TGF-h1 (Fig. 5a and b). Moderate expression of TGF-h1 was observed in fibroblast-like cells in the fibrous zone of the distraction callus (Fig. 5c). The periosteal cells exhibited weak expression of TGF-h1 during the
Table 1 Expression of transforming growth factor-h1 in distraction osteogenesis Distraction osteogenesis Latency Phase Distraction Phase Consolidation Phase Fracture healing
Cell type in distraction callus Post-surgery
Osteoblasts
Osteocytes
Periosteal cells
Fibroblast-like cells
Chondrocytes
Week Week Week Week Week Week
+++ +++ +++ +++ ++/ ++/
ne +++ +++ ++ + +
++ + + + +/ +/
+++ ++ ++ ++ ne ne
ne ne ne ne +++ ne
1 2 3 5 8 12
Cell type in fracture callus Post-surgery
Osteoblasts
Osteocytes
Periosteal cells
Fibroblast-like cells
Chondrocytes
Week Week Week Week Week Week
+++ ++ ++ + + ++/
ne ++ ++ + + +
++ ++ ++ + + +/
+++ ++ + +/ ne ne
ne +++ +++ ne ne ne
1 2 3 5 8 12
+++ :Strong staining, ++ :Moderate staining, + :Weak staining, +/ :Focal weak staining, ++/ staining, ne :Non Existent.
:Focal moderate
74
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 4. Immunohistochemistry of TGF-h1 in fracture healing model. (a) At week 1 post-osteotomy, the fibroblast-like cells (arrows) in the fibrous tissue were stained for TGF-h1 (304). (b) At the newly formed woven bone around the cortex, the osteoblasts were also shown to express TGF-h1. (304) (c) At week 2 post-osteotomy, the callus was formed around the fracture site. The osteoblasts on the woven bone (WB) continued to express TGF-h1. The periosteal cells were also shown to express TGF-h1. (76) (d) At week 3 post-osteotomy, cartilaginous tissue and endochondral ossification were observed at the middle of the fracture callus. At the edge of the cartilaginous tissue (arrows) where endochondral ossification occurred, the expression of TGF-h1 was shown (76).
whole distraction phase (Fig. 5d). At the end of distraction phase, microcolumns of new bone were formed at the distraction callus. The osteoblasts covering the new microcolumns were shown to have strongest TGF-h1 expression (Fig. 5e). During the corresponding phase of fracture healing, the osteoblasts, osteocytes, and the periosteal cells showed moderate staining for TGF-h1 and reduced to a weak expression at week 5 post-surgery (Fig. 4c). Endochondral ossification was observed in the fracture callus at the week 3 post-osteotomy. TGF-h1 was present at the edge of the cartilaginous tissue (Fig. 4d). The staining for TGF-h1 at the fibroblast-like cells showed a weak expression at week 3 post-surgery from a strong signal at week 1 post-surgery. In the consolidation phase of distraction osteogenesis (week 8 to 12 post-operation), although not all the osteoblasts stained for TGF-h1, some showed a moderate signal (Fig. 6a). This moderate staining for TGF-h1 was present throughout the consolidation phase (week 8 to 12) (Fig 6b). At week 8 post-surgery, the chondrocytes in the cartilaginous tissue stained strongly for TGF-h1 (Fig. 6c). The osteocytes in the distraction callus exhibited weak expression of TGF-h1. In fracture
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
75
Fig. 5. Immunohistochemistry of TGF-h1 in distraction osteogenesis. (a) During the distraction period (week 3 post-osteotomy), the fibroblast-like cells (arrows) in the fibrous tissue of the distraction callus were shown to have TGF-h1 expression (304). (b) The osteoblasts (Ob) in the newly formed bone were shown to have TGF-h1 expression (304). (c) Microcolumns of new bone were formed along the direction of the distraction force. The osteoblasts on the new microcolumns of bone had strong TGF-h1 expression (152). (d) At the last week of distraction (week 5 post-osteotomy), the periosteum was expanded and the cells (arrows) were shown to have TGF-h1. (304) (e) At the mature trabecular bone of the distraction callus, the osteocytes (Oc) and osteoblasts (Ob) were stained for TGF-h1 (304). (f) The negative control omitting the primary antibody is shown (152).
76
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
Fig. 6. Immunohistochemistry of TGF-h1 in consolidation period of distraction osteogenesis. (a) After the distraction stopped for 3 weeks, cartilaginous tissue was formed at the middle of the distraction callus. The chondrocytes in the cartilaginous tissue were stained for TGF-h1 (304). (b) The osteoblasts in the maturing distraction callus continued to be stained for TGF-h1 (304). (c) After the distraction stopped for 7 weeks (week 12 post-osteotomy), the lining cells (LC) and osteocytes (Oc) at the trabecular bone were weakly stained for TGF-h1 (304). (d) The osteoblasts were stained moderately for TGF-h1 at week 12 post-osteotomy (304).
healing, the fracture callus mainly consisted of osteoblasts, periosteal cells, and osteocytes. They were stained weakly for TGF-h1.
Discussion The mechanism of the biological response of osseous tissue to external mechanical stimulation in distraction osteogenesis has not been fully understood. In our previous related research, we have shown that the increase in the plasma activity of bone specific alkaline phosphatase was related to the distraction force applied to the callus [1,2]. The plasma activity of the bone specific alkaline phosphatase remained elevated as long as the tension within the callus was maintained by continuous distraction. The mechanism of activation of the enzymatic activity by mechanical stimulation may be related to the local factors present in the distraction callus. In this study, the expression of the TGF-h1 in the bone forming cells was increased throughout the distraction phase. The external mechanical stimulation induced the expression of the local growth factors in the cells involved in new bone formation.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
77
The immunohistochemical result of TGF-h1 in fracture healing was similar to that in other previous studies [6,7,16,17]. In addition to the localization of TGF-h1 at different cells within the fracture callus, we followed the temporal changes of TGF-h1 expression. Strong expression of TGF-h1 of fibroblastlike cells derived from periosteum and osteoblasts was observed at the initial stages of fracture repair. Subsequently, the signal of TGF-h1 in various cells declined gradually. Along with the changes in histology, this indicates that a specific concentration of the TGF-h1 may trigger a cascade events for chondrogenesis and osteogenesis in fracture healing [18,19]. Our present fracture healing serves a good control model for us to compare with distraction osteogenesis model. The osteoblasts in the distraction callus were shown to have strong TGF-h1 expression during the distraction phase. When compared with TGF-h1 expression in fracture healing, the strong expression was sustained as long as the callus was distracted. In other studies, the osteoblast-like cells increased the production of TGF-h1 when they were under a stretching force [14,20]. Lammens et al. [15] also showed that the tissue level of TGF-h1 was increased in the distraction callus as long as distraction applied. The present study further identifies the cellular components in the expression of TGF-h1. It confirms that the distraction has a stimulatory effect on the expression of TGF-h1 in osteoblasts during distraction osteogenesis. It is possible that mechanical force in the distraction osteogenesis model acts as a transduction signal to the osteoblasts that react with an increase in the expression of TGF-h1. The increased expression of TGF-h1 in the distraction callus acts as either autocrine or paracrine to stimulate the osteoblasts to synthesize collagen and alkaline phosphatase for bone formation and mineralization. This may explain our previous finding of the sustained elevation of the plasma bone specific alkaline phosphatase during distraction [1,2]. The hypothesis is further supported by other studies showing that TGF-h1 is a stimulant causing osteoblasts to increase the synthesis of collagen [19,21]. From our observation, increase of TGF-h1 expression was not only due to the osteoblasts, but also the fibroblastlike cells and the osteocytes in the distraction callus. The fibroblast-like cells in the distraction callus maintained a moderate expression of TGF-h1 in distraction phase. This temporal change was different from that in fracture healing. In fracture healing, the expression decreased gradually from the first week of fracture repair and became weak by week 3 post-surgery. The distraction force in the distraction model might induce the fibroblastlike cells in the fibrous zone of distraction callus to express a moderate amount of TGF-h1 protein. Together with other growth factor expression [22,23], the cells may differentiate directly to osteoblasts for bone formation instead of chondrogenesis, as in fracture healing. Therefore, bone was formed at the distraction gap by membraneous ossification. In fracture healing, the decrease of TGF-h1 expression might direct the fibroblast-like cells derived from the periosteum to differentiate to chondrocyte lineage, triggering the cascade of endochondral ossification. Thus, endochondral ossification was observed in the fracture healing model. In many in vitro studies, the effect of TGF-h1 on the differentiation of mesenchymal cells has been shown to be dose dependent [8,11,18,19,24–26]. Once distraction stopped, the distraction callus no longer experienced the distraction force and expression of TGF-h1 decreased. Transforming growth factor-h1 expression dropped to a low level at which the fibroblast-like cells were triggered to differentiate to chondrogenesis. Hence, endochondral ossification was observed in the consolidation phase of the distraction gap. This observation explains the fact that different modes of bone formation were observed during distraction osteogenesis. During the distraction phase, the osteocytes in the distraction callus were also shown to have strong expression of transforming growth factor-h1. Once the distraction stopped, the expression of trans-
78
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
forming growth factor-h1 declined. Osteocytes have long been thought to be the biomechanosensor in osseous tissue [27]. Osteocytes communicate with other bone cells to trigger a cascade of biological events in bone formation when an external mechanical force is applied to osseous tissue. Therefore, the observation of increased expression of transforming growth factor-h1 in the osteocytes may indicate that osteocytes in the distraction callus were also activated and modulated osteogenesis [28]. In this study, the sustained expression of TGF-h1 in various types of cells under mechanical stimulation was demonstrated in the distraction osteogenesis model. Different types of cells responded differently to the mechanical force. As a multi-functional growth factor, TGF-h1 is an important mediator in cell differentiation and cellular activity of the distraction osteogenesis to translate the external mechanical force into a biological response. Acknowledgements This project was funded by Research Grant Committee (CUHK 431/95M), Hong Kong, China.
References [1] Leung KS, Fung KP, Sher AH, Li CK, Lee KM. Plasma bone-specific alkaline phosphatase as an indicator of osteoblastic activity. Journal of Bone & Joint Surgery — British Volume 1993;75:288 – 92. [2] Leung KS, Lee KM, Chan CW, Mak A, Fung KP. Mechanical characterization of regenerated osseous tissue during callotasis and its related biological phenomenon. Life Sciences 2000;66:327 – 36. [3] Leung KS, Fung KP, Liu PP, Lee KM. Bone-specific alkaline phosphatase activities in plasma and callus during callotasis in rabbits. Life Sciences 1995;57:637 – 43. [4] Robey PG, Young MF, Flanders KC, et al. Osteoblasts synthesize and respond to transforming growth factor-type beta (TGF-beta) in vitro. Journal of Cell Biology 1987;105:457 – 63. [5] Sporn MB, Roberts AB. The multifunctional nature of growth factors. In: Sporn MB, Roberts AB, editors. Handbook of experimental pharmacology: Peptide growth factors and their receptors. New York: Springer-Verlag, 1990. pp. 3 – 15. [6] Joyce ME, Jingushi S, Bolander ME. Transforming growth factor-beta in the regulation of fracture repair. Orthopedic Clinics of North America 1990;21:199 – 209. [7] Joyce ME, Terek RM, Jingushi S, Bolander ME. Role of transforming growth factor-beta in fracture repair. Annals of the New York Academy of Sciences 1990;593:107 – 23. [8] Noda M, Camilliere JJ. In vivo stimulation of bone formation by transforming growth factor-beta. Endocrinology 1989;124:2991 – 4. [9] Lind M, Schumacker B, Soballe K, Keller J, Melsen F, Bunger C. Transforming growth factor-beta enhances fracture healing in rabbit tibiae. Acta Orthopaedica Scandinavica 1993;64:553 – 6. [10] Bolander ME. Regulation of fracture repair by growth factors. Proceedings of the Society for Experimental Biology & Medicine 1992;200:165 – 70. [11] Joyce ME, Roberts AB, Sporn MB, Bolander ME. Transforming growth factor-beta and the initiation of chondrogenesis and osteogenesis in the rat femur. Journal of Cell Biology 1990;110:2195 – 207. [12] Gelb DE, Rosier RN, Puzas JE. The production of transforming growth factor-beta by chick growth plate chondrocytes in short term monolayer culture. Endocrinology 1990;127:1941 – 7. [13] Thorp BH, Anderson I, Jakowlew SB. Transforming growth factor-beta 1, -beta 2 and -beta 3 in cartilage and bone cells during endochondral ossification in the chick. Development 1992;114:907 – 11. [14] Holbein O, Neidlinger-Wilke C, Suger G, Kinzl L, Claes L. Ilizarov callus distraction produces systemic bone cell mitogens. Journal of Orthopaedic Research 1995;13:629 – 38. [15] Lammens J, Liu Z, Aerssens J, Dequeker J, Fabry G. Distraction bone healing versus osteotomy healing: a comparative biochemical analysis. Journal of Bone & Mineral Research 1998;13:279 – 86.
H.-Y. Yeung et al. / Life Sciences 71 (2002) 67–79
79
[16] Bourque WT, Gross M, Hall BK. Expression of four growth factors during fracture repair. International Journal of Developmental Biology 1993;37:573 – 9. [17] Matsumoto K, Matsunaga S, Imamura T, Ishidou Y, Yoshida H, Sakou T. Expression and distribution of transforming growth factor-beta and decorin during fracture healing. In Vivo 1994;8:215 – 9. [18] Iwasaki M, Nakata K, Nakahara H, et al. Transforming growth factor-beta 1 stimulates chondrogenesis and inhibits osteogenesis in high density culture of periosteum-derived cells. Endocrinology 1993;132:1603 – 8. [19] Iwasaki M, Nakahara H, Nakata K, Nakase T, Kimura T, Ono K. Regulation of proliferation and osteochondrogenic differentiation of periosteum-derived cells by transforming growth factor-beta and basic fibroblast growth factor. Journal of Bone & Joint Surgery - American Volume 1995;77:543 – 54. [20] Zhuang H, Wang W, Tahernia AD, Levitz CL, Luchetti WT, Brighton CT. Mechanical strain-induced proliferation of osteoblastic cells parallels increased TGF-beta 1 mRNA. Biochemical & Biophysical Research Communications 1996;229:449 – 53. [21] Terrell TG, Working PK, Chow CP, Green JD. Pathology of recombinant human transforming growth factor-beta 1 in rats and rabbits. International Review of Experimental Pathology 1993;34(Pt B):43 – 67. [22] Sato M, Ochi T, Nakase T, et al. Mechanical tension-stress induces expression of bone morphogenetic protein (BMP)-2 and BMP-4, but not BMP-6, BMP-7, and GDF-5 mRNA, during distraction osteogenesis. Journal of Bone & Mineral Research 1999;14:1084 – 95. [23] Yeung HY, Lee SK, Fung KP, Leung KS. Expression of basic fibroblast growth factor during distraction osteogenesis. Clinical Orthopaedics & Related Research 2001;219 – 29. [24] Izumi T, Scully SP, Heydemann A, Bolander ME. Transforming growth factor beta 1 stimulates type II collagen expression in cultured periosteum-derived cells. Journal of Bone & Mineral Research 1992;7:115 – 21. [25] Miura M, Sakagishi Y, Hata K, Komoda T. Differences between the sugar moieties of liver- and bone-type alkaline phosphatases: a re-evaluation. Annals of Clinical Biochemistry 1994;31:25 – 30. [26] Basic N, Basic V, Bulic K, et al. TGF-beta and basement membrane matrigel stimulate the chondrogenic phenotype in osteoblastic cells derived from fetal rat calvaria. Journal of Bone & Mineral Research 1996;11:384 – 91. [27] Mikuni-Takagaki Y, Suzuki Y, Kawase T, Saito S. Distinct responses of different populations of bone cells to mechanical stress. Endocrinology 1996;137:2028 – 35. [28] Lean JM, Mackay AG, Chow JW, Chambers TJ. Osteocytic expression of mRNA for c-fos and IGF-I: an immediate early gene response to an osteogenic stimulus. American Journal of Physiology 1996;270:E937 – 45.