Endochondral ossification in vitro is influenced by mechanical bending

Bone development is influenced by the local mechanical environment. Experimental evidence suggests that altered loading can change cell proliferation and differentiation in chondro- and osteogenesis during endochondral ossification. This study investigated the effects of three-point bending of murine fetal metatarsal bone anlagen in vitro on cartilage differentiation, matrix mineralization and bone collar formation. This is of special interest because endochondral ossification is also an important process in bone healing and regeneration. Metatarsal preparations of 15 mouse fetuses stage 17.5 dpc were dissected en bloc and cultured for 7 days. After 3 days in culture to allow adherence they were stimulated 4 days for 20 min twice daily by a controlled bending of approximately 1000-1500 microstrain at 1 Hz. The paraffin-embedded bone sections were analyzed using histological and histomorphometrical techniques. The stimulated group showed an elongated periosteal bone collar while the total bone length was not different from controls. The region of interest (ROI), comprising the two hypertrophic zones and the intermediate calcifying diaphyseal zone, was greater in the stimulated group. The mineralized fraction of the ROI was smaller in the stimulated group, while the absolute amount of mineralized area was not different. These results demonstrate that a new device developed to apply three-point bending to a mouse metatarsal bone culture model caused an elongation of the periosteal bone collar, but did not lead to a modification in cartilage differentiation and matrix mineralization. The results corroborate the influence of biophysical stimulation during endochondral bone development in vitro. Further experiments with an altered loading regime may lead to more pronounced effects on the process of endochondral ossification and may provide further insights into the underlying mechanisms of mechanoregulation which also play a role in bone regeneration.

Fracture healing via periosteal callus formation requires macrophages for both initiation and progression of early endochondral ossification

The distribution, phenotype, and requirement of macrophages for fracture-associated inflammation and/or early anabolic progression during endochondral callus formation were investigated. A murine femoral fracture model [internally fixed using a flexible plate (MouseFix)] was used to facilitate reproducible fracture reduction. IHC demonstrated that inflammatory macrophages (F4/80+Mac-2+) were localized with initiating chondrification centers and persisted within granulation tissue at the expanding soft callus front. They were also associated with key events during soft-to-hard callus transition. Resident macrophages (F4/80+Mac-2neg), including osteal macrophages, predominated in the maturing hard callus. Macrophage Fas-induced apoptosis transgenic mice were used to induce macrophage depletion in vivo in the femoral fracture model. Callus formation was completely abolished when macrophage depletion was initiated at the time of surgery and was significantly reduced when depletion was delayed to coincide with initiation of early anabolic phase. Treatment initiating 5 days after fracture with the pro-macrophage cytokine colony stimulating factor-1 significantly enhanced soft callus formation. The data support that inflammatory macrophages were required for initiation of fracture repair, whereas both inflammatory and resident macrophages promoted anabolic mechanisms during endochondral callus formation. Overall, macrophages make substantive and prolonged contributions to fracture healing and can be targeted as a therapeutic approach for enhancing repair mechanisms. Thus, macrophages represent a viable target for the development of pro-anabolic fracture treatments with a potentially broad therapeutic window...

Kinetic characterization of a membrane-specific ATPase from rat osseous plate and its possible significance on endochondral ossification

Treatment with phosphatidylinositol-specific phospholipase C of rat osseous plate membranes released up to 90-95% of alkaline phosphatase, but a specific ATPase activity (optimum pH = 7.5) remained bound to the membrane. The hydrolysis of ATP by this ATPase was negligible in the absence of magnesium or calcium ions. However, at millimolar concentrations of magnesium and calcium ions, the membrane-specific ATPase activity increased to about 560-600 U/mg, exhibiting two classes of ATP-hydrolysing sites, and site-site interactions. GTP, UTP, ITP, and CTP were also hydrolyzed by the membrane-specific ATPase. Oligomycin, ouabain, bafilomycin A(1), thapsigargin, omeprazole, ethacrynic acid and EDTA slightly affected membrane-specific ATPase activity while vanadate produced a 18% inhibition. The membrane-specific ATPase activity was insensitive to theophylline, but was inhibited 40% by levamisole. These data suggested that the membrane-specific ATPase activity present in osseous plate membranes, and alkaline phosphatase, were different proteins. (C) 1998 Elsevier B.V. B.V.

Changes in cells's secretory organelles and extracellular matrix during endochondral ossification in the mandibular condyle of the growing rat

The mandibular condyle from 20-day-old rats was examined in the electron microscope with particular attention to intracellular secretory granules and extracellular matrix. Moreover, type II collagen was localized by an immunoperoxidase method. The condyle has been divided into five layers: (1) the most superficial, articular layer, (2) polymorphic cell layer, (3) flattened cell layer, (4) upper hypertrophic, and (5) lower hypertrophic cell layers. In the articular layer, the cells seldom divide, but in the polymorphic layer and upper part of the flattened cell layer, mitosis gives rise to new cells. In these layers, cells produce two types of secretory granules, usually in distinct stacks of the Golgi apparatus; type a, cylindrical granules, in which 300-nm-long threads are packed in bundles which appear lucent after formaldehyde fixation; and type b, spherical granules loaded with short, dotted filaments. The matrix is composed of thick banded lucent fibrils in a loose feltwork of short, dotted filaments. The cells arising from mitosis undergo endochondral differentiation, which begins in the lower part of the flattened cell layer and is completed in the upper hypertrophic cell layer; it is followed by gradual cell degeneration in the lower hypertrophic cell layer. The cells produce two main types of secretory granules: type b as above; and type c, ovoid granules containing 300-nm-long threads associated with short, dotted filaments. A possibly different secretory granule, type d, dense and cigar-shaped, is also produced. The matrix is composed of thin banded fibrils in a dense feltwork. In the matrix of the superficial layers, the lucency of the fibrils indicated that they were composed of collagen I, whereas the lucency of the cylindrical secretory granules suggested that they transported collagen I precursors to the matrix. Moreover, the use of ruthenium red indicated that the feltwork was composed of proteoglycan; the dotted filaments packed in spherical granules were similar to, and presumably the source of, the matrix feltwork. The superficial layers did not contain collagen II and were collectively referred to as perichondrium. In the deep layers, the ovoid secretory granules displayed collagen II antigenicity and were likely to transport precursors of this collagen to the matrix, where it appeared in the thin banded fibrils. That these granules also carried proteoglycan to the matrix was suggested by their content of short dotted filaments. Thus the deep layers contained collagen II and proteoglycan as in cartilage; they were collectively referred to as the hyaline cartilage region.

Characterization of the phosphatidylinositol-specific phospholipase C-released form of rat osseous plate alkaline phosphatase and its possible significance on endochondral ossification

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Effect of alendronate on endochondral ossification in mandibular condyles of growing rats

The replacement of the calcified cartilage by bone tissue during the endochondral ossification of the mandibular condyle is dependent of the resorbing activity of osteoclats. After partial resorption, calcified cartilage septa are covered by a primary bone matrix secreted by osteoblasts. Osteoadherin (OSAD) is a small proteoglycan present in bone matrix but absent in cartilage during the endochondral ossification. The aim of this study was to analyze the effect of alendronate, a drug known to inhibit bone resorption by osteoclasts, on the endochondral ossification of the mandibular condyle of young rats, by evaluating the distribution of osteoclasts and the presence of OSAD in the bone matrix deposited. Wistar newborn rats (n = 45) received daily injections of alendronate (n = 27) or sterile saline solution as control (n = 18) from the day of birth until the ages of 4, 14 and 30 days. At the days mentioned, the mandibular condyles were collected and processed for transmission electron microscopy analysis. Specimens were also submitted to tartrate resistant acid phosphatase (TRAP) histochemistry and ultrastructural immunodetection of OSAD. Alendronate treatment did not impede the recruitment and fusion of osteoclasts at the ossification zone during condyle growth, but they presented inactivated phenotype. The trabeculae at the ossification area consisted of cartilage matrix covered by a layer of primary bone matrix that was immunopositive to OSAD at all time points studied. Apparently, alendronate impeded the removal of calcified cartilage and maturation of bone trabeculae in the mandibular ramus, while in controls they occurred normally. These findings highlight for giving attention to the potential side-effects of bisphosphonates administered to young patients once it may represent a risk of disturbing maxillofacial development.

Site-1 protease is essential for endochondral bone formation in mice

Site-1 protease (S1P) has an essential function in the conversion of latent, membrane-bound transcription factors to their free, active form. In mammals, abundant expression of S1P in chondrocytes suggests an involvement in chondrocyte function. To determine the requirement of S1P in cartilage and bone development, we have created cartilage-specific S1P knockout mice (S1P(cko)). S1P(cko) mice exhibit chondrodysplasia and a complete lack of endochondral ossification even though Runx2 expression, Indian hedgehog signaling, and osteoblastogenesis is intact. However, there is a substantial increase in chondrocyte apoptosis in the cartilage of S1P(cko) mice. Extraction of type II collagen is substantially lower from S1P(cko) cartilage. In S1P(cko) mice, the collagen network is disorganized and collagen becomes entrapped in chondrocytes. Ultrastructural analysis reveals that the endoplasmic reticulum (ER) in S1P(cko) chondrocytes is engorged and fragmented in a manner characteristic of severe ER stress. These data suggest that S1P activity is necessary for a specialized ER stress response required by chondrocytes for the genesis of normal cartilage and thus endochondral ossification.

Targeted overexpression of parathyroid hormone-related peptide in chondrocytes causes chondrodysplasia and delayed endochondral bone formation.

Parathyroid hormone-related peptide (PTHrP) was initially identified as a product of malignant tumors that mediates paraneoplastic hypercalcemia. It is now known that the parathyroid hormone (PTH) and PTHrP genes are evolutionarily related and that the products of these two genes share a common receptor, the PTH/PTHrP receptor. PTHrP and the PTH/PTHrP receptor are widely expressed in both adult and fetal tissues, and recent gene-targeting and disruption experiments have implicated PTHrP as a developmental regulatory molecule. Apparent PTHrP functions include the regulation of endochondral bone development, of hair follicle formation, and of branching morphogenesis in the breast. Herein, we report that overexpression of PTHrP in chondrocytes using the mouse type II collagen promoter induces a novel form of chondrodysplasia characterized by short-limbed dwarfism and a delay in endochondral ossification. This features a delay in chondrocyte differentiation and in bone collar formation and is sufficiently marked that the mice are born with a cartilaginous endochondral skeleton. In addition to the delay, chondrocytes in the transgenic mice initially become hypertrophic at the periphery of the developing long bones rather than in the middle, leading to a seeming reversal in the pattern of chondrocyte differentiation and ossification. By 7 weeks, the delays in chondrocyte differentiation and ossification have largely corrected, leaving foreshortened and misshapen but histologically near-normal bones. These findings confirm a role for PTHrP as an inhibitor of the program of chondrocyte differentiation. PTHrP may function in this regard to maintain the stepwise differentiation of chondrocytes that initiates endochondral ossification in the midsection of endochondral bones early in development and that also permits linear growth at the growth plate later in development.

The genetic basis of human height : the role of estrogen

Height is a complex physical trait that displays strong heritability. Adult height is related to length of the long bones, which is determined by growth at the epiphyseal growth plate. Longitudinal bone growth occurs via the process of endochondral ossification, where bone forms over the differentiating cartilage template at the growth plate. Estrogen plays a major role in regulating longitudinal bone growth and is responsible for inducing the pubertal growth spurt and fusion of the epiphyseal growth plate. However, the mechanism by which estrogen promotes epiphyseal fusion is poorly understood. It has been hypothesised that estrogen functions to regulate growth plate fusion by stimulating chondrocyte apoptosis, angiogenesis and bone cell invasion in the growth plate. Another theory has suggested that estrogen exposure exhausts the proliferative capacity of growth plate chondrocytes, which accelerates the process of chondrocyte senescence, leading to growth plate fusion. The overall objective of this study was to gain a greater understanding of the molecular mechanisms behind estrogen-mediated growth and height attainment by examining gene regulation in chondrocytes and the role of some of these genes in normal height inheritance. With the heritability of height so well established, the initial hypothesis was that genetic variation in candidate genes associated with longitudinal bone growth would be involved in normal adult height variation. The height-related genes FGFR3, CBFA1, ER and CBFA1 were screened for novel polymorphisms using denaturing HPLC and RFLP analysis. In total, 24 polymorphisms were identified. Two SNPs in ER (rs3757323 C>T and rs1801132 G>C) were strongly associated with adult male height and displayed an 8 cm and 9 cm height difference between homozygous genotypes, respectively. The TC haplotype of these SNPs was associated with a 6 cm decrease in height and remarkably, no homozygous carriers of the TC haplotype were identified in tall subjects. No significant associations with height were found for polymorphisms in the FGFR3, CBFA1 or VDR genes. In the epiphyseal growth plate, chondrocyte proliferation, matrix synthesis and chondrocyte hypertrophy are all major contributors to long bone growth. As estrogen plays such a significant role in both growth and final height attainment, another hypothesis of this study was that estrogen exerted its effects in the growth plate by influencing chondrocyte proliferation and mediating the expression of chondrocyte marker genes. The examination of genes regulated by estrogen in chondrocyte-like cells aimed to identify potential regulators of growth plate fusion, which may further elucidate mechanisms involved in the cessation of linear growth. While estrogen did not dramatically alter the proliferation of the SW1353 cell line, gene expression experiments identified several estrogen regulated genes. Sixteen chondrocyte marker genes were examined in response to estrogen concentrations ranging from 10-12 M to 10-8 M over varying time points. Of the genes analysed, IHH, FGFR3, collagen II and collagen X were not readily detectable and PTHrP, GHR, ER, BMP6, SOX9 and TGF1 mRNAs showed no significant response to estrogen treatments. However, the expression of MMP13, CBFA1, BCL-2 and BAX genes were significantly decreased. Interestingly, the majority of estrogen regulated genes in SW1353 cells are expressed in the hypertrophic zone of the growth plate. Estrogen is also known to regulate systemic GH secretion and local GH action. At the molecular level, estrogen functions to inhibit GH action by negatively regulating GH signalling. GH treated SW1353 cells displayed increases in MMP9 mRNA expression (4.4-fold) and MMP13 mRNA expression (64-fold) in SW1353 cells. Increases were also detected in their respective proteins. Treatment with AG490, an established JAK2 inhibitor, blocked the GH mediated stimulation of both MMP9 and MMP13 mRNA expression. The application of estrogen and GH to SW1353 cells attenuated GH-stimulated MMP13 levels, but did not affect MMP9 levels. Investigation of GH signalling revealed that SW1353 cells have high levels of activated JAK2 and exposure to GH, estrogen, AG490 and other signalling inhibitors did not affect JAK2 phosphorylation. Interestingly, AG490 treatment dramatically decreased ERK2 signalling, although GH did stimulate ERK2 phosphorylation above control levels. AG490 also decreased CBFA1 expression, a transcription factor known to activate MMP9 and MMP13. Finally, GH and estrogen treatment increased expression of SOCS3 mRNA, suggesting that SOCS3 may regulate JAK/STAT signalling in SW1353 cells. The modulation of GH-mediated MMP expression by estrogen in SW1353 cells represents a potentially novel mechanism by which estrogen may regulate longitudinal bone growth. However, further investigation is required in order to elucidate the precise mechanisms behind estrogen and GH regulation of MMP13 expression in SW1353 cells. This study has provided additional evidence that estrogen and the ER gene are major factors in the regulation of growth and the determination of adult height. Newly identified polymorphisms in the ER gene not only contribute to our understanding of the genetic basis of human height, but may also be useful in association studies examining other complex traits. This study also identified several estrogen regulated genes and indicated that estrogen modifies the expression of genes which are primarily expressed in the hypertrophic region of the epiphyseal growth plate. Furthermore, synergistic studies incorporating GH and estrogen have revealed the ability of estrogen to attenuate the effects of GH on MMP13 expression, revealing potential pathways by which estrogen may modulate growth plate fusion, longitudinal bone growth and even arthritis.

Characterisation of mesenchymal cells from osteophytes in osteoarthritis

Osteophytes form through the process of chondroid metamorphosis of fibrous tissue followed by endochondral ossification. Osteophytes have been found to consist of three different mesenchymal tissue regions including endochondral bone formation within cartilage residues, intra-membranous bone formation within fibrous tissue and bone formation within bone marrow spaces. All these features provide evidence of mesenchymal stem cells (MSC) involvement in osteophyte formation; nevertheless, it remains to be characterised. MSC from numerous mesenchymal tissues have been isolated but bone marrow remains the “ideal” due to the ease of ex vivo expansion and multilineage potential. However, the bone marrow stroma has a relatively low number of MSC, something that necessitates the need for long-term culture and extensive population doublings in order to obtain a sufficient number of cells for therapeutic applications. MSC in vitro have limited proliferative capacity and extensive passaging compromises differentiation potential. To overcome this barrier, tissue derived MSC are of strong interest for extensive study and characterisation, with a focus on their potential application in therapeutic tissue regeneration. To date, no MSC type cell has been isolated from osteophyte tissue, despite this tissue exhibiting all the hallmark features of a regenerative tissue. Therefore, this study aimed to isolate and characterise cells from osteophyte tissues in relation to their phenotype, differentiation potential, immuno-modulatory properties, proliferation, cellular ageing, longevity and chondrogenesis in in vitro defect model in comparison to patient matched bone marrow stromal cells (bMSC). Osteophyte derived cells were isolated from osteophyte tissue samples collected during knee replacement surgery. These cells were characterised by the expression of cell surface antigens, differentiation potential into mesenchymal lineages, growth kinetics and modulation of allo-immune responses. Multipotential stem cells were identified from all osteophyte samples namely osteophyte derived mesenchymal stem cells (oMSC). Extensively expanded cell cultures (passage 4 and 9 respectively) were used to confirm cytogenetic stability and study signs of cellular aging, telomere length and telomerase activity. Cultured cells at passage 4 were used to determine 84 pathway focused stem cell related gene expression profile. Micro mass pellets were cultured in chondrogenic differentiation media for 21 days for phenotypic and chondrogenic related gene expression. Secondly, cell pellets differentiated overnight were placed into articular cartilage defects and cultured for further 21 days in control medium and chondrogenic medium to study chondrogenesis and cell behaviour. The surface antigen expression of oMSC was consistent with that of mesenchymal stem cells, such as lacking the haematopoietic and common leukocyte markers (CD34, CD45) while expressing those related to adhesion (CD29, CD166, CD44) and stem cells (CD90, CD105, CD73). The proliferation capacity of oMSC in culture was superior to that of bMSC, and they readily differentiated into tissues of the mesenchymal lineages. oMSC also demonstrated the ability to suppress allogeneic T-cell proliferation, which was associated with the expression of tryptophan degrading enzyme indoleamine 2,3 dioxygenase (IDO). Cellular aging was more prominent in late passage bMSC than in oMSC. oMSC had longer telomere length in late passages compared with bMSC, although there was no significant difference in telomere lengths in the early passages in either cell type. Telomerase activity was detectable only in early passage oMSC and not in bMSC. In osteophyte tissues telomerase positive cells were found to be located peri vascularly and were Stro-1 positive. Eighty-four pathway-focused genes were investigated and only five genes (APC, CCND2, GJB2, NCAM and BMP2) were differentially expressed between bMSC and oMSC. Chondrogenically induced micro mass pellets of oMSC showed higher staining intensity for proteoglycans, aggrecan and collagen II. Differential expression of chondrogenic related genes showed up regulation of Aggrecan and Sox 9 in oMSC and collagen II in bMSC. The in vitro defect models of oMSC in control medium showed rounded and aggregated cells staining positively for proteoglycan and presence of some extracellular matrix. In contrast, defects with bMSC showed fragmentation and loss of cells, fibroblast-like cell morphology staining positively for proteoglycans. For defects maintained in chondrogenic medium, rounded, aggregated and proteoglycan positive cells were found in both oMSC and bMSC cultures. Extracellular matrix and cellular integration into newly formed matrix was evident only in oMSC defects. For analysis of chondrocyte hypertrophy, strong expression of type X collagen could be noticed in the pellet cultures and transplanted bMSC. In summary, this study demonstrated that osteophyte derived cells had similar properties to mesenchymal stem cells in the expression of antigen phenotype, differential potential and suppression of allo-immune response. Furthermore, when compared to bMSC, oMSC maintained a higher proliferative capacity due to a retained level of telomerase activity in vitro, which may account for the relatively longer telomeres delaying growth arrest by replicative senescence compared with bMSC. oMSC behaviour in defects supported chondrogenesis which implies that cells derived from regenerative tissue can be an alternative source of stem cells and have a potential clinical application for therapeutic stem cell based tissue regeneration.

Engineering tubular bone constructs

Cell-sheet techniques have been proven effective in various soft tissue engineering applications. In this experiment, we investigated the feasibility of bone tissue engineering using a hybrid of mesenchymal stem cell (MSC) sheets and PLGA meshes. Porcine MSCs were cultured to a thin layer of cell sheets via osteogenic induction. Tube-like long bones were constructed by wrapping the cell sheet on to PLGA meshes resulting in constructs which could be cultured in spinner flasks, prior to implantation in nude rats. Our results showed that the sheets were composed of viable cells and dense matrix with a thickness of about 80–120 mm, mineral deposition was also observed in the sheet. In vitro cultures demonstrated calcified cartilage-like tissue formation and most PLGA meshes were absorbed during the 8-week culture period. In vivo experiments revealed that dense mineralized tissue was formed in subcutaneous sites and the 8- week plants shared similar micro-CT characteristics with native bone. The neo tissue demonstrated histological markers for both bone and cartilage, indicating that the bone formation pathway in constructs was akin to endochondral ossification, with the residues of PLGA having an effect on the neo tissue organization and formation. These results indicate that cell-sheet approaches in combination with custom-shaped scaffolds have potential in producing bone tissue.

Instability prolongs the chondral phase during bone healing in sheep

In this sheep study, we investigated the influence of fixation stability on the temporal and spatial distribution of tissues in the fracture callus. As the initial mechanical conditions have been cited as being especially important for the healing outcome, it was hypothesized that differences in the path of healing would be seen as early as the initial phase of healing. ----- ----- Sixty-four sheep underwent a mid-shaft tibial osteotomy that was treated with either a rigid or a semi-rigid external fixator. Animals were sacrificed at 2, 3, 6 and 9 weeks postoperatively and the fracture calluses were analyzed using radiological, biomechanical and histological techniques. Statistical comparison between the groups was performed using the Mann–Whitney U test for unpaired non-parametric data. ----- ----- In the callus of the tibia treated with semi-rigid fixation, remnants of the fracture haematoma remained present for longer, although new periosteal bone formation during early healing was similar in both groups. The mechanical competence of the healing callus at 6 weeks was inferior compared to tibiae treated with rigid fixation. Semi-rigid fixation resulted in a larger cartilage component of the callus, which persisted longer. Remodeling processes were initiated earlier in the rigid group, while new bone formation continued throughout the entire investigated period in the semi-rigid group. ----- ----- In this study, evidence is provided that less rigid fixation increased the time required for healing. The process of intramembranous ossification appeared during the initial stages of healing to be independent of mechanical stability. However, the delay in healing was related to a prolonged chondral phase.

The mechanical heterogeneity of the hard callus influences local tissue strains during bone healing : a finite element study based on sheep experiments

During secondary fracture healing, various tissue types including new bone are formed. The local mechanical strains play an important role in tissue proliferation and differentiation. To further our mechanobiological understanding of fracture healing, a precise assessment of local strains is mandatory. Until now, static analyses using Finite Elements (FE) have assumed homogenous material properties. With the recent quantification of both the spatial tissue patterns (Vetter et al., 2010) and the development of elastic modulus of newly formed bone during healing (Manjubala et al., 2009), it is now possible to incorporate this heterogeneity. Therefore, the aim of this study is to investigate the effect of this heterogeneity on the strain patterns at six successive healing stages. The input data of the present work stemmed from a comprehensive cross-sectional study of sheep with a tibial osteotomy (Epari et al., 2006). In our FE model, each element containing bone was described by a bulk elastic modulus, which depended on both the local area fraction and the local elastic modulus of the bone material. The obtained strains were compared with the results of hypothetical FE models assuming homogeneous material properties. The differences in the spatial distributions of the strains between the heterogeneous and homogeneous FE models were interpreted using a current mechanobiological theory (Isakson et al., 2006). This interpretation showed that considering the heterogeneity of the hard callus is most important at the intermediate stages of healing, when cartilage transforms to bone via endochondral ossification.

Influence of internal fixator stiffness on murine fracture healing : two types of fracture healing lead to two distinct cellular events and FGF-2 expressions

This study aimed to clarify the relationship between the mechanical environment at the fracture site and endogenous fibroblast growth factor-2 (FGF-2). We compared two types of fracture healing with different callus formations and cellular events using MouseFix(TM) plate fixation systems for murine fracture models. Left femoral fractures were induced in 72 ten-week-old mice and then fixed with a flexible (Group F) or rigid (Group R) Mouse Fix(TM) plate. Mice were sacrificed on days 3, 5, 7, 10, 14, and 21. The callus volumes were measured by 3D micro-CT and tissues were histologically stained with hematoxylin & eosin or safranin-O. Sections from days 3, 5, and 7 were immunostained for FGF-2 and Proliferating Cell Nuclear Antigen (PCNA). The callus in Group F was significantly larger than that in Group R. The rigid plate allowed bone union without a marked external callus or chondrogenesis. The flexible plate formed a large external callus as a result of endochondral ossification. Fibroblastic cells in the granulation tissue on days 5 and 7 in Group F showed marked FGF-2 expression compared with Group R. Fibroblastic cells showed ongoing proliferation in granulation tissue in group F, as indicated by PCNA expression, which explained the relative granulation tissue increase in group F. There were major differences in early phase endogenous FGF-2 expression between these two fracture healing processes, due to different mechanical environments.