SEARCH

SEARCH BY CITATION

Keywords:

  • bone healing;
  • mineralized callus;
  • mineral particles;
  • ESEM;
  • scanning-SAXS

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Bone healing is known to occur through the successive formation and resorption of various tissues with different structural and mechanical properties. To get a better insight into this sequence of events, we used environmental scanning electron microscopy (ESEM) together with scanning small-angle X-ray scattering (sSAXS) to reveal the size and orientation of bone mineral particles within the regenerating callus tissues at different healing stages (2, 3, 6, and 9 weeks). Sections of 200 µm were cut from embedded blocks of midshaft tibial samples in a sheep osteotomy model with an external fixator. Regions of interest on the medial side of the proximal fragment were chosen to be the periosteal callus, middle callus, intercortical callus, and cortex. Mean thickness (T parameter), degree of alignment (ρ parameter), and predominant orientation (ψ parameter) of mineral particles were deduced from resulting sSAXS patterns with a spatial resolution of 200 µm. 2D maps of T and ρ overlapping with ESEM images revealed that the callus formation occurred in two waves of bone formation, whereby a highly disordered mineralized tissue was deposited first, followed by a bony tissue with more lamellar appearance in the ESEM and where the mineral particles were more aligned, as revealed by sSAXS. As a consequence, degree of alignment and mineral particle size within the callus increased with healing time, whereas at any given moment there were structural gradients, for example, from periosteal toward the middle callus. © 2010 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Bone healing is a complex process involving inflammation, chondrogenesis, osteogenesis, and remodeling.1–3 This process leads to the formation of different tissue types and is affected by various intrinsic factors4–9 and external environment conditions, such as the stability of the bone fragments.10–13 During secondary bone healing, the mineralized callus tissue and the newly formed bone undergo continuous changes in their structural and mechanical properties.14 Generally, the quality of newly formed bone tissue has been evaluated mostly by radiographic, histologic, and various macro- or micro-level biomechanical tests.15 On the nanoscale, bone materials are composed mainly of a collagen-rich matrix and mineral particles (carbonated hydroxyapatite). These components are assembled into a complex hierarchical structure that confers a combination of stiffness and extraordinary toughness to the bone material.16–20 For normal bone, it is known that the amount, size, and orientation of the mineral particles play an important role in its performance.21–24 It is therefore very likely that these nanoscale properties of the callus tissues are spatially and temporally associated with functional adaptation and are essential for the restoration of morphologic and mechanical integrity of the skeleton as a whole. Despite the relatively large amount of work documented in the literature focusing on bone healing, the material properties at the nanoscale level of the mineralized callus are not well characterized.

Over the last few decades, there have been an increasing number of studies focusing on the mineral phase of bone and callus formation in human or in different animal models. Among other techniques, diffraction measurements with X-rays, synchrotron, or neutron radiation25–29 can be used to quantify the mineral properties in bone or fracture callus. Wen and colleagues30 combined scanning electron microscopy (SEM), X-ray diffraction (XRD), and transmission electron microscopy (TEM) and proposed an evolutionary model for human callus formation that contains five representative stages of collagen-mineral organization, where fusion of mineralized fibrils is followed by the absorption of disordered and the deposition of ordered collagen matrix.31 However, the spatial variation together with the temporal evolution of callus mineral crystals in terms of their size and arrangement at the nanoscale level has not been investigated.

Scanning small-angle X-ray scattering (sSAXS) reveals properties of the nanosized mineral particles, such as their thickness and orientation distribution,32 and in particular allows mapping of these properties within mineralized callus sections by a methodology similar to what has been reported previously for bone biopsies.33–36 In addition, environmental scanning electron microscopy (ESEM) provides the possibility to study the degree of mineralization and the morphology of the mineralized tissue at the submicron level. Using our established sheep osteotomy model37 with the combination of sSAXS and ESEM techniques, we were able to map the distribution of mineral particle size and orientation in representative areas of the mineralized callus as a function of position and healing time.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Samples

Longitudinal sections of callus samples had been collected from a previous study of an osteotomy model in sheep (healthy female Merino mix sheep, aged 2.5 years, mean weight 66 ± 8 kg). In brief, a monolateral external fixator (6 Schanz screws, 2 steel tubes) was attached to the medial aspect of the tibia. Osteotomy of the tibial diaphysis was performed and distracted to 3 mm intraoperatively.12, 37 The bones were harvested after 2, 3, 6, and 9 weeks of healing and were fixed and dehydrated with ethanol and embedded in polymethyl methacrylate (PMMA) resin without decalcifying. Thick sections of 200 µm were cut from embedded blocks using a low-speed diamond saw (Buehler Isomet, Buehler GmbH, Duesseldorf, Germany) for sSAXS measurements. Three sets of samples from each time point were used for this study and were investigated by ESEM and histology, the same set of specimen that had been used in the nanomechanical property study reported previously.38 One sample per time point was used for the complementary sSAXS experiments.

Environmental scanning electron microscopy

The morphology of the bone samples was observed with an ESEM device (FEI FEG-ESEM Quanta 600, FEI Company, Hillsboro, OR, USA) with a backscattered electron (BSE) detector operated at 12.5 kV with a 10-mm sample-detector distance, 4.0-nm spot size, and 0.75 torr pressure. To obtain an overview image, subimages were acquired consecutively with approximately 50% spatial overlap at ×100 magnification. Subimages were stitched together using ArcSoft Panorama Maker software (Version 3.0, ArcSoft, Inc., Fremont, CA, USA).

Higher-magnification images (×4000) were made for all four samples at each time point at three specific locations within the mineralized periosteal callus (Fig. 1A). Position A was chosen in an area of the mineralized callus formed by intramembranous ossification. Position C was selected to be in mineralized callus adjacent to the cortex and just above the level of the osteotomy. Position B was located between positions A and C but distanced periosteally farther away from the cortex.

thumbnail image

Figure 1. (A) Backscattered electron images (×4000) and (B) histologic sections with Movat-pentachrome staining at three different locations—A, B, and C—with the callus as shown in the overview image. Data are shown for 2, 3, 6, and 9 weeks of healing as indicated. Different tissue types are visible: (1) an unmineralized fibrous tissue as observed in region A at 2 and 3 weeks, (2) a mineralized but poorly organized woven tissue as observed in regions A and B at 2 weeks, and (3) mostly lamellar bone as observed in region A at 9 weeks.

Download figure to PowerPoint

Scanning small-angle X-ray scattering

sSAXS experiments were carried out using a laboratory instrument (Nanostar, Bruker AXS, Karlsruhe, Germany) with a rotating copper anode generator (M06XCE–SRA, Mac Science, Yokohama, Japan) operated at 40 kV and 100 mA. Scanning of the sample was realized through software-controlled movement of the translation stage together with the sample holder in x and y directions of sample plane perpendicular to the incident beam. The sample-to-detector distance was fixed at 600 mm. An X-ray radiograph of the sample was generated by using a pin diode that detected intensity values after transmission of the primary beam through the sample with a resolution of 100 µm. The measurement points within the regions of interest mentioned earlier were selected from this radiograph with a spacing of 200 µm. The scattered X-ray intensity was collected on a 2D detector with an accumulation time of 3600 seconds for each point. Calibration of the instrument parameters was done by measuring a standard [silver behenate (AgBh)]. An empty X-ray beam without sample was measured for background noise reduction, and a dark current was measured without X-ray beam and sample.

The regions of interest in mineralized callus and cortex chosen for scattering experiments are as illustrated in Fig. 2: (1) Region 1 (1.0 × 2.2 mm) lies in the periosteal callus, which is 3 mm away from the osteotomy site. This region was chosen for comparison with the previously reported mechanical properties of the mineralized callus from the same sample blocks.38 (2) Region 2 (2.2 × 3.6 mm) is in the middle of the callus that connects regions 1 and 4, defined only for 6- and 9-week samples. (3) Region 3 (1.4 × 2.6 mm) lies in the cortical bone above the osteotomy gap and was further classified into cortex near and away from osteotomy gap (later referred to as deep cortex, 1.4 × 0.8 mm). (4) Region 4 is the intercortical callus in the osteotomy gap, for which bone was found only in the 6- and 9-week samples.

thumbnail image

Figure 2. Distribution of mineral particle characteristics after (A) 6 weeks and (B) 9 weeks of healing mapped onto an ESEM overview image. The mean mineral thickness is illustrated with color coding. The direction of the pink bars indicates the predominant mineral orientation, and their length indicates the degree of mineral alignment (where 0 means randomly oriented, whereas 1 means perfectly aligned). At 6 weeks, there were only few areas with relatively thick and well-orientated mineral particles within both callus regions. At 9 weeks, these mineral properties reached values close to the cortex level, at least in callus region 1. At both time points, the mineral particle orientation (pink bars) in the callus followed the trabecular structure seen in the ESEM image.

Download figure to PowerPoint

Data analysis

Experimental data from sSAXS were treated with fit2d software (Version 12.077, Andy Hammersley/ESRF, Grenoble, France) and a self-developed GNU Octave program (Version 3.0.0). The 2D sSAXS patterns were integrated radially and azimuthally to obtain intensity versus q and intensity versus φ profiles, respectively (where q is the scattering vector related to the distance from the center of the sSAXS pattern and φ is the polar angle). The effective data range in this sSAXS study was chosen to be q from 0.34 to 2.83 nm−1 and φ from 0 to 360 degrees, and the q range for Porod fit was 2.00 to 2.83 nm−1. The 1D profiles of scattering intensity I derived from integration were calibrated as in Eq. (1) by considering local transmission rate τ while subtracting background air scattering Ibgr and instrumental noise from the dark current measurement IDC.

  • equation image(1)

By analyzing the calibrated radial and azimuthal sSAXS spectra, it is possible to calculate the mean thickness (T parameter), the degree of alignment (ρ parameter), and the predominant orientation (ψ parameter) of mineral particles within a unit volume of the specimen.32, 34, 39 The T parameter describes average particle characteristics as 4Φ(1 – Φ)/S,39 where Φ is the volume fraction of mineral particles and S is the surface area of mineral particles per unit volume of tissue. Supposing that the edge lengths of a mineral particle were a, b, and c in three dimensions, then the T parameter for platelike mineral particles can be simplified as40

  • equation image(2)

Assuming the volume fraction Φ to be 0.5 for bone material, the T parameter denotes the average thickness of mineral particles.

The degree of alignment (ρ parameter) indicates the percentage of aligned particles (ρ = 0 means randomly oriented, whereas ρ = 1 means perfectly aligned). Calculation of the ρ parameter was performed by fitting two Gaussian curves to the azimuthally integrated spectrum using the Levenberg-Marquardt algorithm. Given that A1 is the area below the curve above baseline, whereas A0 is the area below the baseline, the value of ρ parameter is then expressed as A1/(A0 + A1).34 The predominant orientation (ψ parameter) of the mineral particles within the plane of the sample section was represented by the mean value of the two fitted peak positions.

Data representation

2D maps of mineral particle properties obtained from sSAXS analysis were plotted with Sigmaplot software (Systat Software, Inc., Chicago, IL, USA). Mean thickness (T parameter) of the mineral particles was plotted in color-coded contour maps. Direction of the small bars indicates predominant orientation (ψ parameter) of the mineral particles, and length of the bars indicates their degree of alignment (ρ parameter). These plots were then overlaid on the ESEM images.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Morphology of the tissue within mineralized callus

Tissue morphology at the submicron level was first investigated by environmental scanning electron microscopy (ESEM) at four healing time points. At each time point, sections from three different animals were analyzed. Figure 1A shows high-magnification ESEM images from several positions of the callus as a function of healing time. Images from the other two sets of samples are given in Supplemental Fig. 1.

At position A, the mineralized callus tissue was characterized by an unorganized structure of woven bone after 2 weeks. Large holes (appearing dark, with no mineral content) were seen throughout the region. After 3 weeks of healing, some well-aligned bony structures were occurring. Most of the specimen in this region contained a mixture of poorly organized mineralized tissue (woven bone) and well-organized mineralized tissue (lamellar bone). At 6 weeks, this mineralized callus zone was almost completely filled with densely packed bony tissue. Well-aligned lamellar bone was observed in most areas. At 9 weeks, the woven bone structure was no longer present, and only lamellar bone was observed, occasionally with some visible osteocyte lacunae.

At position B, the sequence appeared to be delayed. Indeed, at 2 weeks, approximately half the area was composed of woven bone, the rest being unmineralized tissue or embedding material. After 3 weeks of healing, morphologic characteristics of woven bone were detected, and randomly distributed areas of voids were found, similar to position A at 2 weeks. At 6 weeks, a mixture of woven bone and lamellar bone structure was observed, similar to the 3-week appearance at position A. Over half the region was filled with organized lamellar fibrils after 9 weeks of healing, which was similar to the mineralized callus appearance at position A at 6 weeks.

At position C, neither the 2-week specimen nor the 3-week specimen showed signs of any mineralized tissue. After 6 weeks, mineralized callus tissue was observed that resembled the morphology of the 2-week specimen at position A and that of the 3-week specimen at position B. At 9 weeks, mostly woven tissue, sometimes with a granular appearance, was found, and invasion of blood vessels was seen only sparingly. Isolated areas of lamellar bone were detected as well.

Similar observations were made in the histology sections stained with Movat-pentachrome, as shown in Fig. 1B. Around position C at early time points (2 and 3 weeks) of healing, only dense fibrous tissue that was very rich in cells was observed. At these early time points, the regions farther away from the osteotomy gap (positions A and B) were crossed by single-bone trabeculae consisting of woven bone. The woven bone was characterized by large, roundish osteocytes in the bone and broad, constant osteoid seams on the bone surface. At 6 weeks, remnants of cartilage existed at position C, whereas at positions A and B the mineralized regions grew larger, with the proportion of fibrous tissue decreasing. The bone tissue consisted mostly of woven bone but was already partially replaced by lamellar bone, as characterized by smaller, flattened osteocytes. Osteoid seams were visible at most of the bone surfaces. Nine weeks after osteotomy, lamellar bone was predominant in all positions.

Time evolution of mineral particle properties within mineralized callus and cortex

The average values, determined by sSAXS, of mean thickness and degree of alignment of mineral particles in the mineralized periosteal callus (region 1) and cortical bone (region 3) are listed in Table 1 for all healing time points. The mean thickness of the mineral particles within the mineralized periosteal callus region increased with healing time. Meanwhile, their degree of alignment did not show an increase. The mineral particles in the cortex near the osteotomy gap were similar to the mineral particles deep inside the cortex at 2, 3, and 6 weeks, whereas at the later time point, 9 weeks, the mean thickness of mineral particles near the osteotomy gap was smaller with a lesser degree of alignment than those in the deep cortex.

Table 1. List of the Effective Average Values of the Mean Thickness (T Parameter) and Degree of Alignment (ρ Parameter) of the Mineral Particles in Regions 1 and 3 After 2, 3, 6, and 9 Weeks of Bone Healing
Bone healing timeT (nm) (mean ± SD)ρ (mean ± SD)
Mineralized periosteal callusCortex near osteotomyDeep cortexMineralized periosteal callusCortex near osteotomyDeep cortex
2 weeks2.42 ± 0.122.67 ± 0.032.67 ± 0.030.34 ± 0.130.67 ± 0.020.67 ± 0.01
3 weeks2.41 ± 0.112.62 ± 0.052.64 ± 0.070.23 ± 0.050.61 ± 0.020.61 ± 0.03
6 weeks2.57 ± 0.092.78 ± 0.042.79 ± 0.040.36 ± 0.110.68 ± 0.020.68 ± 0.02
9 weeks2.82 ± 0.062.74 ± 0.072.76 ± 0.060.48 ± 0.060.62 ± 0.070.64 ± 0.02

Spatial distribution of mineral particle properties within mineralized callus and cortex

The spatial maps of mineral particle thickness and orientation in the mineralized callus and cortical bone are shown in Fig. 2 for 6- and 9-week samples overlaid on their respective ESEM overview image of the osteotomized bone. At 6 weeks, mineral particles in the mineralized callus had a much lower value of both mean thickness and degree of alignment than those of the cortex. At 9 weeks, the mean thickness increased and reached values similar to those of the cortex in most of region 1 and parts of region 2. Moreover, the distribution of thicker (>2.5 nm) and well-aligned particles in regions 1 and 2 followed the direction of trabeculae, as revealed by ESEM.

In the cortical region near the osteotomy site (region 3 in Fig. 2), a typical lamellar structure was seen on the ESEM overview, and mineral particles appeared thick and well aligned. The ESEM images showed that there was a distinct boundary between the mineralized intercortical callus (region 4) and the cortex (region 3) at 6 weeks, whereas the newly formed bone in the intercortical region rejoined with the existing cortex at 9 weeks. There also was a boundary in mean thickness of mineral particles between the mineralized intercortical callus (blue color, region 4 in Fig. 2A) and cortex (orange-yellow color, region 3 in Fig. 2A) at 6 weeks, whereas the distinction disappeared at 9 weeks, showing that the mineral particles in the osteotomy gap were growing in thickness (green color, region 4 in Fig. 2B), approaching the size of particles in the cortex near the osteotomy gap. There was no change in the thickness and orientation of mineral particles in the deep cortex between 6 and 9 weeks. The orientation of mineral particles inside the cortex was vertical, whereas the orientation of mineral particles in the mineralized callus within the osteotomy gap was nearly horizontal.

Mineral particle properties within mineralized callus as a function of distance from the periosteal cortex and from the osteotomy gap

As the mineralized periosteal callus developed during healing, the thickness of its mineral particles differed with distance from the periosteal cortex. Figure 3 presents the change in mineral particle thickness and orientation as a function of distance from the cortex at all time points of healing in region 1 of the mineralized periosteal callus (Fig. 3D). The plots of normalized values of particle thickness (normalized T = T/Tdeep, where Tdeep is the average T value within the previously defined “deep cortex” region of the same specimen) are presented in Fig. 3B to interpret the variation of T within each of the specimens. This representation was chosen to better visualize the variation within a given cortex independent of the interindividual variations between different animals.

thumbnail image

Figure 3. Plots (with standard error bars) of the distribution of (A) mineral thickness T, (B) normalized values of mineral thickness (normalized with respect to the average value within the deep cortex of the same specimen), and (C) degree of mineral alignment ρ as a function of distance from the cortex for the mineralized periosteal callus region marked in the ESEM image (D). Regions near the cortex always had larger values of both mineral thickness and mineral orientation at any given time point.

Download figure to PowerPoint

The thickness of the mineral particles in the callus decreased at the earlier healing time (2 and 3 weeks) as a function of distance from the cortex, whereas at 6 and 9 weeks it was independent of this distance (Fig. 3A). The normalized data (normalized with respect to the deep cortex level) in Fig. 3B showed that the mineral particles in this mineralized callus region were comparable between 2 and 6 weeks but slightly lower in mean thickness compared with the mineral particles in the deep cortex. At 9 weeks, however, the thickness reached values higher than those of the cortical bone particles. The degree of orientation of the mineral particles revealed a similar pattern as a function of position, apart from the fact that the overall value never came close to the cortex level (which was between 0.6 and 0.7 for all the four specimens).

To analyze the effect of distance from the osteotomy site, regions 1 and 2 of the mineralized periosteal callus were combined as shown in Fig. 4D, and the change in mineral particle thickness and orientation in this combined region was plotted as a function of distance from the osteotomy site in Fig. 4A–C. There were no data from the mineralized middle callus region (region 2) in the 2- and 3-week samples owing to the absence of mineralized tissue at these early time points. It is evident that mineral particles at about 4 mm away from the osteotomy site always were thicker and more aligned than those lying closer to the osteotomy gap, given a certain period of healing. The values of thickness of mineral particles expressed a clear gradient along the long bone axis that can be seen both with and without normalization of the data (normalized with respect to the deep cortex level) in Fig. 4A, B. This gradient was not affected by the healing time. The normalized mean thickness of mineral particles increased most dramatically from 6 to 9 weeks of healing (Fig. 4B), whereas such an increase was not present for particle alignment (Fig. 4C). The organization of mineral particles depicted by the degree of alignment in Fig. 4C showed that the mineral particles were not well aligned, as also seen from Fig. 2. It can be noted that at 9 weeks, the degree of alignment of mineral particles at distances of 3 mm and more away from the osteotomy site was essentially unchanged.

thumbnail image

Figure 4. Plots (with standard error bars) of the distribution of (A) mineral thickness T, (B) normalized values of mineral thickness (normalized with respect to the average value within the deep cortex of the same specimen), and (C) degree of mineral alignment ρ as a function of distance from the osteotomy site for the mineralized periosteal callus region marked in the ESEM image (D). Regions about 4 mm away from the osteotomy site always had larger values of both mineral thickness and mineral orientation at any given time point.

Download figure to PowerPoint

Mineral particle properties within the cortex as a function of distance from the osteotomy gap

The variations in particle thickness within the cortex and osteotomy gap can be seen from the maps in Fig. 2, where the particle alignment appeared to be the same at 6 and 9 weeks in the cortex. A more detailed analysis of particle thickness as a function of distance from the osteotomy site in the cortex (region 3) and mineralized callus in the osteotomy gap (region 4) is given in Fig. 5. The representative ESEM image in Fig. 5B shows the region of interest used for the plot in Fig. 5A, and it has to be mentioned that at 2 and 3 weeks of bone healing there was no mineralized callus tissue occurring inside the gap.

thumbnail image

Figure 5. (A) Plots (with standard error bars) of the distribution of mineral thickness T as a function of distance from the osteotomy site for the cortex and for the osteotomy gap region (at 6- and 9-week time points) marked in the ESEM image (B). The inset shows normalized values of mineral thickness in cortex alone (normalized with respect to the average value within deep cortex of the same specimen). During healing, mineral thickness in the cortex decreased (4% at 9 weeks) in proximity of the osteotomy site.

Download figure to PowerPoint

The thickness of the mineral particles inside the cortex was above 2.5 nm at all time points. The mineral particles within the osteotomy gap, however, always were much thinner (<2.2 nm) even after 6 weeks of healing. Only after 9 weeks, this difference decreased owing to a strong increase of particle thickness within the osteotomy gap. The inset in Fig. 5A shows the normalized data of particle thickness for the cortical bone alone (not including the osteotomy site). The normalized particle thickness did not show any changes from deep cortex to the osteotomy site at 2 weeks, as expected. A decrease in particle thickness was observed in the cortical bone near the osteotomy site in the 6- and 9-week specimens. Specifically, in the 9-week specimen, the mineral particles in the cortex closer to the osteotomy site were smaller by 4% in normalized values of mean thickness than those in the cortex further away from the osteotomy site.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

During bone healing, the structural and mechanical properties of the newly formed bone undergo constant changes at all levels of hierarchy. Based on a detailed structural investigation of a sheep osteotomy model using scanning electron microscopy and small-angle x-ray scattering, we report here that intramembranous ossification of the callus occurs in a biphasic process. It was found that woven bone deposition is followed by lamellar bone apposition (Fig. 1), perhaps by a successive action of mesenchymal osteoblasts and surface osteoblasts.14 The organization of bony tissue in callus suggests a certain propagation of formation waves, as indicated by ESEM imaging, represented by a scheme as shown in Fig. 6. The first wave of tissue transformation from unmineralized tissue to unorganized mineralized tissue reached positions A and B at 2 weeks and position C only at 6 weeks. The second wave of tissue transformation corresponding to lamellar bone apposition reached position A at 6 weeks and position B at 9 weeks but never occurred in position C before the end of the experiment at 9 weeks.

thumbnail image

Figure 6. Based on the spatial and temporal occurrence of the three different tissue types shown in Fig. 1, it is proposed that the transformation from structure (1) to (2) and from structure (2) to (3) occurs in two sequential “waves” propagating along the cortex toward the osteotomy gap. Numbers in parenthesis—(1), (2), and (3)—refer to the description in the legend of Fig. 1.

Download figure to PowerPoint

The ultrastructural properties of the mineral particles are known to depend on species, age, and type of tissue in which they are located,41, 42 that is, trabecular bone or cortical bone in humans, for example.33, 34 From micromechanical tests and histologic analyses, the region of mineralized periosteal callus (region 1) was known first to achieve rapid woven-bone formation characterized by high mineralization close to the fully mature cortex level38 and afterwards to undergo significant remodeling characterized by increased osteoclastic activity (number and density) together with decreased bone density.37 Here we found that at 2 weeks of healing, the mean particle thickness of this mineralized callus region already was close to the values in the cortex, but the degree of particle alignment was only half. In contrast, the mean particle thickness at 9 weeks was found to be even larger than in the cortex, whereas the degree of particle alignment was still far below that of cortical bone despite an increase during healing (seen both in Table 1 and Fig. 2B).

In previous nanomechanical studies38 we reported that even at 2 weeks of bone healing, a high mineral content (35 vol% and 22 wt% in some regions) was observed in region 1. This dramatic accumulation of mineral in the early phase of healing might contribute to the early appearance of relatively large particles detected by sSAXS (Table 1). Despite this high mineral content, the average indentation modulus in this mineralized periosteal callus region was much lower (13 GPa) than that of normal sheep cortical bone (26 GPa), even at 9 weeks.38 This could be due to the fact that the mineral particles were still not as well aligned as in cortical bone (Fig. 2B). It is well known that the orientation of both the mineral and the collagen fibrils plays an important role in mechanical properties.43, 44 In previous histologic studies,37 significant bone turnover was observed in the mineralized periosteal callus region, with an increase in the osteoclastic density occurring at the 6- and 9-week time points. This implies that bone remodeling was significant only at late healing stages.

Present mathematical models of bone healing45–47 usually are based on the experimental observation that the main source of the stem cells is the periosteum,48–50 and thus new bone tissue within the external mineralized callus is predicted to form first on the periosteal side and then into the osteotomy/fracture gap. Our study does support this concept in a way that callus mineral particles appeared to develop much earlier/faster in the periosteal region near the cortical bone (Fig. 3) with a distance of around 4 mm from the osteotomy site (Fig. 4). It also was observed that the most significant increase in particle thickness took place at the latest stage (from 6 to 9 weeks), whereas the degree of particle alignment changed less throughout the healing process under investigation (Figs. 3B and 4B). Furthermore, since the mineral particles within the mineralized periosteal callus at 9 weeks were still less well oriented (although slightly thicker) than the ones within old cortical bone, it is also possible to predict that the rearrangement of mineral particles in mineralized callus still might continue even after the bony bridge had been achieved at 6 and 9 weeks if the bone is not resorbed before this occurs.

Apart from mineralized periosteal callus (region 1), cortical bone (region 3), mineralized middle callus (region 2), and mineralized intercortical callus (region 4) were evaluated by sSAXS in the 6- and 9-week specimens. It was observed that the mean particle thickness and degree of particle alignment in mineralized middle callus (region 2) increased during healing time (Fig. 2). A steady increase in mean particle thickness and degree of particle alignment also was present in mineralized intercortical callus (region 4) from 6 to 9 weeks. It is interesting to see that the mineralized callus tissue formed within the osteotomy gap at 9 weeks had an orientation of its mineral particles perpendicular to that of the mineral particles in the cortex (Fig. 2B), which could be an indirect outcome of the invasion of mesenchymal cells into the osteotomy gap. Knowing that the fibrous tissue that forms in the early fracture gap is orientated transversely in our sheep osteotomy model,13 the alignment or orientation of mineral particles therefore is maintained in the initial bone formation but probably would be remodeled later to be vertical. Incidentally, this also indicates that an elevated fracture risk still remains owing to the weak interface remaining at the original osteotomy site even at 9 weeks when a bony bridge has formed based on both histologic analysis and X-ray radiologic examination.13

Another interesting event was a decrease in mineral particle thickness toward the osteotomy site in the cortex near the fracture even at 9 weeks. While the resorption processes in the intact cortical bone adjacent to the fracture have been observed previously as an increased porosity,13 it was until now not known that the remaining bone also undergoes microstructural changes. The decrease in average particle thickness started early from 3 weeks of healing, as detected by sSAXS, suggesting a dissolution of mineral close to the osteotomy site (Fig. 5A). A pronounced remodeling and resorption activity of osteoclasts in the cortex occurred only after 6 weeks of healing according to previous histologic studies.37 This might indicate a potential role of the osteocytes in dissolving parts of the cortical mineral phase near the fracture gap.51

In summary, detailed analysis of mineral particle characteristics in the mineralized callus of sheep reveals an intricate bone-formation process whereby the formation of highly disordered woven bone is followed by the deposition of lamellar tissue in an orderly spatial and temporal fashion. Of course, an interspecies comparison certainly would be quite interesting and is currently ongoing, comparing the sheep and rat models.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

This study was supported by a grant from the German Research Foundation (DFG) under the collaborative research centre scheme (SFB 760, Project B1). The authors appreciate the help of G. Dinst and A. Martins with sample preparation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
jbmr_84_sm_supplMat.doc411KSupplementary Material

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.