- Email: [email protected]
The role of mineral content in determining the micromechanical properties of discrete trabecular bone remodeling packets
Journal of Biomechanics 43 (2010) 3144–3149
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
The role of mineral content in determining the micromechanical properties of discrete trabecular bone remodeling packets Lachlan J. Smith a, Jeffrey P. Schirer b, Nicola L. Fazzalari a,c,n a
Bone and Joint Research Laboratory, Surgical Pathology, SA Pathology, Adelaide, Australia. Applications Research Laboratory, Hysitron Inc., Minneapolis, USA. c Discipline of Pathology, School of Medical Sciences, Faculty of Health Sciences, University of Adelaide, Australia. b
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 29 July 2010
In trabecular bone, each remodeling event results in the resorption and/or formation of discrete structural units called ‘packets’. These remodeling packets represent a fundamental level of bone’s structural hierarchy at which to investigate composition and mechanical behaviors. The objective of this study was to apply the complementary techniques of quantitative backscattered electron microscopy (qBSEM) and nanoindentation to investigate inter-relationships between packet mineralization, elastic modulus, contact hardness and plastic deformation resistance. Indentation arrays were performed across nine trabecular spicules from 3 human donors; these spicules were then imaged using qBSEM, and discretized into their composite remodeling packets (127 in total). Packets were classified spatially as peripheral or central, and mean contact hardness, plastic deformation resistance, elastic modulus and calcium content calculated for each. Inter-relationships between measured parameters were analysed using linear regression analyses, and dependence on location assessed using Student’s t-tests. Significant positive correlations were found between all mechanical parameters and calcium content. Elastic modulus and contact hardness were significantly correlated, however elastic modulus and plastic deformation resistance were not. Calcium content, contact hardness and elastic modulus were all significantly higher for central packets than for peripheral, confirming that packet mineral content contributes to micromechanical heterogeneity within individual trabecular spicules. Plastic deformation resistance, however, showed no such regional dependence, indicating that the plastic deformation properties in particular, are determined not only by mineral content, but also by the organic matrix and interactions between these two components. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Trabecular bone Remodeling packet Mineralization Hardness Elastic modulus
1. Introduction Remodeling packets are the basic structural units of trabecular bone (Roschger et al., 2008). Each packet represents a discrete remodeling event, reflecting the continual process of resorption and formation that functions to repair microdamage, adapt existing structures in response to altered mechanical environment, and regulate calcium homeostasis (Fazzalari et al., 2002; Hadjidakis and Androulakis, 2006). Remodeling packets therefore represent a fundamental level of trabecular bone’s structural hierarchy at which to investigate the relationships between function and composition. Bone is comprised of both an organic component (an organized network of type I collagen fibers) and an inorganic component (hydroxyapatite nano-crystals bound to those collagen fibers)
n Corresponding author at: Bone and Joint Research Laboratory, Surgical Pathology, SA Pathology, Frome Rd, Adelaide, SA 5000, Australia. Tel.: + 61 8 82223269; fax: + 61 8 82223293. E-mail address: [email protected] (N.L. Fazzalari).
0021-9290/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2010.07.038
(Ruppel et al., 2008). The mineral component is considered to play a fundamental role in determining bone’s strength and toughness (Roschger et al., 2008). Each bone packet has a unique mineral content, which is determined not only by mechanobiology and calcium metabolism, but also by time since deposition (Hadjidakis and Androulakis, 2006; Roschger et al., 2008). At the microscopic level, inter-relationships between micromechanical properties and mineral content have been quantitatively examined in numerous tissues, including human femoral calcified cartilage and subchondral bone, iliac crest and several others, by combining the techniques of nanoindentation and backscattered electron microscopy (Ferguson et al., 2003; Ferguson et al., 2008; Oyen et al., 2008). These studies have demonstrated positive correlations between micromechanical properties and the mineral content of the tissue. Parameters classically obtained from indentation tests are elastic modulus and hardness (contact hardness) (Oliver and Pharr, 1992). These properties vary throughout the cross sections of individual spicules, with higher values present near the spicule core, and lower values present near the spicule periphery
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
(Brennan et al., 2009; Mulder et al., 2007; Norman et al., 2008). As older and more highly mineralized bone packets are generally located near the center of the spicule, and newer, less mineralized packets are located near the spicule periphery, there is an implicit relationship between packet mineralization and micromechanical properties. Indeed, this compositional and functional gradient is considered to impart superior macro-scale biomechanical behavior with respect to toughness and stability (Busse et al., 2009). Contact hardness is derived from both the elastic and plastic deformation properties of the material, and as such displays strong linear interdependence with elastic modulus (Oyen, 2006). An expression has been developed that recognizes this interdependence by assuming in-series contributions of elastic and plastic contributions to deformation, and has recently been proposed for the analysis of mineralized tissues (Oyen, 2006; Sakai, 1999) This expression allows derivation of a specific measure of resistance to plastic deformation, independent of elastic properties. Examination of this resistance to plastic deformation parameter and its relationship to composition and other micromechanical properties may provide new insights into the mechanisms that underlie failure of the bone composite at the nanoscale. Improved understanding of these factors may be particularly important in terms of understanding the long term effects of biological therapies that modulate the rate and extent of bone remodeling, which are used in the treatment of diseases such as osteoporosis. The objective of this study was to investigate the importance of mineral content on the micromechanical behavior of discrete remodeling packets in non-diseased human femoral trabecular bone. We hypothesized that calcium content would correlate positively with packet elastic modulus, contact hardness and plastic deformation resistance. Additionally, we hypothesized that both calcium content and micromechanical properties would be significantly higher for packets at the center of trabecular spicules, than for those lining the spicule periphery.
2. Methods 2.1. Specimen preparation With institutional research ethics committee and next of kin approvals, intertrochanteric, femoral, trabecular bone samples were obtained at autopsy from three female cases aged 44, 61 and 68. Each had no history of musculoskeletal disease, or other illnesses or drug treatments considered likely to affect bone metabolism, as indicated by case notes. The intertrochanteric region was selected due to its location distal to articular surfaces and common fracture sites. Specimens were washed in phosphate buffered saline, then dehydrated progressively in serially graded ethanols and embedded in polymethylmethacrylate (PMMA) resin, according to established methodologies (Mittra et al., 2006; Norman et al., 2008), and mounted on an aluminum puck. A carbon–aluminum standard was mounted adjacent to each surface to be analysed, and both were polished progressively, firstly with silicon–carbide sandpaper then with dry alumina cloths to achieve a mirror finish. Polishing quality was validated in a pilot study using the indentation instrument under tapping mode, which demonstrated that the peak to trough distance was consistently less than ten percent of the peak indentation depth. Polishing quality was also assessed qualitatively for each sample using a stereo microscope. 2.2. Nanoindentation Nanoindentation was performed using a TI-900 TriboIndenter (Hysitron Inc., Eden Prairie, MN, USA). Within each sample, three trabecular spicules were randomly selected for analysis using an optical microscope, and imaged to enable subsequent relocation using the electron microscope; additionally, deep reference indents were placed adjacent to each selected spicule for this same purpose. Indentations were performed under displacement control using a Berkovich diamond tip, pre-calibrated with fused quartz, according to the following protocol: a ten-second linear approach to a peak displacement of 400 nm, a 40-second hold at peak displacement and a ten-second linear withdrawal to zero displacement. This protocol was optimised to compensate for potential time and strain-rate
3145
dependent material behavior. Arrays of indents spaced horizontally and vertically at 30 mm intervals and filling the pre-defined spicule areas were implemented (a total of 1790 indents, 199 7 70 per spicule, (mean 7 SD)).
2.3. Microscopy and imaging Following nanoindentation, and without re-polishing in order to preserve indent locations, the surface of each sample was sputter-coated with carbon using a vacuum evaporator (Hitachi, Tokyo, Japan); specimens were then mounted in the chamber of a scanning electron microscope fitted with a backscattered electron detector (XL20; Philips, Amsterdam, Netherlands). Imaging was performed using an adaptation of established methods (Sutton-Smith et al., 2008). The microscope was switched on and left to stand for at least four hours prior to imaging for beam warm up and stabilization. Using an accelerating voltage of 15 kV, a working distance of 15 mm and at a magnification of 200 times, the carbon and aluminum standard grey levels were adjusted to approximately 30 and 225, respectively, by varying the contrast, brightness and spot size. Importantly, these standard grey levels were achieved for all three samples while maintaining a constant working distance. Each spicule that had been analysed using nanoindentation was then located from optical microscope images and imaged at 200 times magnification using backscattered electron (BSE) detection (Fig. 1A). Before and after each spicule the standards were imaged to account for minor temporal variations in the microscope beam current that might influence grey level stability. Indent locations were unable to be visualized easily using BSE detection; as such, following BSE imaging, multiple images covering the total area of each spicule were then taken at 1500 times magnification under secondary electron (SE) detection using an accelerating voltage of 10 kV, which improved surface contrast. These images were used to reconstruct a high resolution montage from which the precise locations of all indents could be mapped (Fig. 1B and C).
2.4. Analysis For each indent, contact hardness, Hc and elastic modulus, E were calculated according to the methods and definitions of Oliver and Pharr (1992) E¼
1n2B 1 Er
1n2I EI
where EI and nI are the elastic modulus and Poisson’s ratio of the diamond indenter (1140 GPa and 0.07, respectively), nB is the assumed Poisson’s ratio of bone (0.3) and Er is the reduced modulus of the sample, calculated from the formula pffiffiffiffi S p Er ¼ pffiffiffi 2 A where A is the projected contact area and S is the stiffness derived from an exponential curve fit to the unloading curve. Contact hardness, Hc was calculated using the formula Hc ¼
Pmax A
where Pmax is the maximum load. Additionally, plastic deformation resistance, H, was calculated using the model described by Sakai (1999) and previously proposed for mineralized tissue (Oyen, 2006). This model assumes in-series spring element contributions to elastic and plastic deformations, where the H is related to Hc and Er by the formula Hc ¼
1
a1 ðða2 Er Þ1=2 þ ða1 HÞ1=2 Þ2
with a1 ¼24.5 and a2 ¼4.4, being constants associated with the Berkovich indentation tip (Sakai, 1999). Using imaging software (Photoshop V5.0; Adobe Systems Inc., San Jose, CA, USA), indent locations from SE image montages were mapped to corresponding BSE images. Remodeling packets within each spicule were identified by tracing cement lines (Fig. 1D); each packet was then isolated and transferred to a separate image file for analysis. The mean hardness and elastic modulus for all indents falling within each packet was calculated. Indents falling close to bone-PMMA boundaries, close to or within osteocyte lacunae, or on cement lines were excluded from analyses. Grey level calibrations and conversions were carried out in accordance with previously published techniques (Roschger et al., 1998; SuttonSmith et al., 2008). Initially, a linear relationship between the average atomic numbers and grey levels of carbon, aluminum was confirmed against a hydroxyapatite standard. The relationship between calcium content and grey level was established from images of the hydroxyapatite standard and unmineralized collagen. The equation governing this relationship was wt% Ca ¼ 0.1733 grey level—4.3325. Our custom Matlab program (The Mathworks, Inc.; Natick, MA, USA) was able to calculate any minor temporal deviations in the grey levels of the carbon and aluminum standards from the target values of 25 and 225,
3146
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Fig. 1. (A) Representative BSE image of a trabecular spicule. (B) Corresponding SE image montage with mapped indents: ¼ included indents; ¼ indents excluded due to proximity to cell lacunae or spicule boundaries. (C) Higher magnification view showing actual indents spaced at 30 mm intervals. (D) BSEM image with traced remodeling packets and cross-mapped indents; ¼ additional indents excluded due to proximity to cement lines.
respectively, and correct the sample grey levels accordingly, before conversion to an estimate of calcium content.
2.5. Statistics For the analysis, packets from the nine spicules were initially pooled. The influence of calcium content on E, Hc and H was examined using linear regression analyses where strong, moderate and weak correlations were defined as r40.7, 0.7Z r 40.3 and r r0.3, respectively. Interdependence between Hc and H was examined in a similar way. Packets were then stratified into 2 groups, peripheral (with at least one edge at the spicule periphery) or central (no edges at the spicule periphery), and the same analyses performed. To examine whether interrelationships between properties varied between central and peripheral packets, the slopes and y-intercepts of the respective regression lines were compared using t-tests. Finally, differences in each of these properties between central and peripheral packets were examined using unpaired Student’s t-tests. All analyses were performed using GraphPad Prism V.5 (GraphPad Software Inc., San Diego, CA, USA), with 2-tailed pcritical ¼0.05.
3. Results A total of 127 discrete bone remodeling packets were resolved from the BSE images of the nine spicules, with a median of 8 (3–16) (interquartile range) indents falling within each packet. The mean7 SD peak load for all 1790 indents was 2.870.5 mN; a representative load vs. displacement curve is shown in Fig. 2. Mean7SD for E, Hc, H and Ca for all 127 packets were 22.347 3.01 GPa, 0.7470.07 GPa, 2.2070.25 GPa and 24.93 70.90 wt%, respectively. Linear regression analysis results demonstrated a strong positive correlation between Hc and E (r¼ 0.73, p o0.005, Fig. 3A); H and E, however, were not significantly correlated (r¼ 0.15, Fig. 3B). Calcium content correlated significantly with E, Hc and H (r ¼0.34, r¼ 0.61 and r ¼0.55, respectively, p o0.005, Fig. 4A–C). Of the 127 packets, 46 were classified as peripheral and the remaining 81 classified as central. Ca, E and Hc were found to be lower for peripheral packets (97%, p o0.005; 90%, po0.005 and 96%, p o0.05 of central, respectively, Fig. 5A–C). H, however, was not significantly different between the two groups (Fig. 5D).
Fig. 2. Representative force vs. displacement curve for a single indent, showing regions corresponding to loading, 40 s hold and unloading.
Linear regression results for peripheral and central packets analysed separately are summarized in Table 1. For central spicules, E and Hc, were strongly correlated, E and H, and Ca and E were each weakly correlated, and Ca and Hc, and Ca and H were each moderately correlated. For peripheral spicules, E and Hc, Ca and Hc and Ca and H were each moderately correlated, while E and H, and Ca and E were not significantly correlated. Neither the slopes or y-intercepts of the regression lines for each of these comparisons were significantly different between central and peripheral packets.
4. Discussion In this study we examined inter-relationships between mineral content and micromechanical properties for 127 discrete remodeling packets in human trabecular bone spicules. Mean values for elastic modulus and contact hardness determined here for samples from donors with no identifiable musculoskeletal disease are similar to those published previously (Norman et al., 2008), with differences likely attributable to the specifics of the
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
3147
Fig. 3. Linear regression plots comparing remodeling packet micromechanical properties. (A) Contact hardness, Hc and elastic modulus, E exhibit strong positive correlation. (B) Elastic modulus and plastic deformation resistance, H, are not correlated. Strong, moderate and weak correlations are defined as r 40.7, 0.7 Zr 40.3 and r r0.3, respectively.
Fig. 4. Linear regression plots comparing remodeling packet calcium content with micromechanical properties. (A) Calcium content, Ca and elastic modulus, E exhibit moderate positive correlation. (B) Calcium content and contact hardness, Hc exhibit moderate positive correlation. (C) Calcium content and plastic deformation resistance, H, exhibit moderate positive correlation. Strong, moderate and weak correlations are defined as r 40.7, 0.7Z r4 0.3 and rr 0.3, respectively.
Fig. 5. Comparison of micromechanical properties and calcium content between central and peripheral remodeling packets. (A) Calcium content, Ca, (B) elastic modulus, E and (C) contact hardness, Hc are all significantly higher for central packets. (D) Plastic deformation resistance, H, however was not significantly different between regions. n p o0.05, nnp o0.005.
Table 1 Correlations between mechanical properties and calcium content for central and peripheral remodeling packets. E ¼elastic modulus, Hc ¼contact hardness, H¼ plastic deformation resistance and Ca ¼ wt% calcium and strong, moderate and weak correlations are defined as r 40.7, 0.7 Zr 40.3 and r r 0.3, respectively. Central n ¼83 E vs. Hc E vs. H Ca vs. E Ca vs. Hc Ca vs. H
r¼ 0.78, r¼ 0.28, r¼ 0.28, r¼ 0.59, r¼ 0.67,
p o0.005 (strong) p o0.05 (weak) p o0.05 (weak) p o0.005 (moderate) p o0.005 (moderate)
Peripheral n¼ 44 r ¼0.61, r ¼0.09, r ¼0.12, r ¼0.63, r ¼0.69,
p o 0.005 (moderate) p 40.05 p 40.05 p o 0.005 (moderate) p o 0.005 (moderate)
testing protocol. Similarly, mean values for wt% Ca determined here are very close to those determined previously (Sutton-Smith et al., 2008). Our results show that the heterogeneity in elastic modulus and contact hardness through the cross sections of trabecular spicules (Norman et al., 2008) may be attributed, at least in part, to the distribution of bone remodeling packets with unique mineral compositions, a property determined by multiple factors including time since deposition (and hence rate of remodeling), the local micromechanical environment and the systemic regulation of calcium homeostasis (Boivin and Meunier, 2002; Hadjidakis and Androulakis, 2006; Roschger et al., 2008).
3148
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Correlations between mineral content, contact hardness and elastic modulus are consistent with the results of previous studies at the nano-, micro- and macro-scales (Boivin et al., 2008; Currey, 1988; Ferguson et al., 2003; Mulder et al., 2007). In particular, the relationship between mineralization and hardness of bone structural units in cortical bone from the iliac crest was recently examined using micro-indentation and X-ray diffraction experiments (Boivin et al., 2008). This study found a very similar correlation between contact hardness and mineral content (r¼ 0.6) to the current study, suggesting a relationship between these properties that potentially extends across orders of scale, and across different anatomical sites. In this study we have, for the first time, examined the relationship between calcium content and plastic deformation resistance. Interestingly, while calcium was found to be a significant predictor of plastic deformation resistance, plastic deformation resistance itself was found not to be significantly different between central and peripheral packets, despite the observation that packets in these locations have significantly different calcium contents. With respect to function, this suggests that trabecular spicules may possess heterogeneous elastic mechanical properties while at the same time maintaining an overall homogeneous resistance to plastic deformation. This finding highlights the complex inter-relationships between bone composition, and elastic and plastic mechanical behavior at the microscopic level. Bone has very small nano-crystals in a collagen network in contrast to relatively large crystals found in metals (Gao et al., 2003). Hydroxyapatite nano-crystals in bone do not fracture when bone fractures. Bone fractures as a result of disruption of the bonds between hydroxyapatite and collagen as well as bonds within the collagen matrix (Gao et al., 2003; Nalla et al., 2003). There is no evidence that plastic deformation in bone is associated with dislocations in hydroxyapatite crystals (in contrast to plastic deformation in metals). While mineral content has classically been used as the predictor of bone strength, there is increasing evidence that the organic matrix, particularly collagen cross-link biochemistry, plays a very important role in its mechanical performance. In fact, recent work has demonstrated that the organic matrix may account for as much as one third of the variation in contact hardness (Boivin et al., 2008). Like degree of mineralization, the nature of collagen intermolecular cross-links varies with distance from the spicule periphery (Paschalis et al., 2003). We hypothesize that the plastic deformation properties of bone packets are determined by both the interactions between hydroxyapatite crystals and the collagen fibers, as well as the degree of collagen cross-linking, both of which are likely to be strongly dependent on packet maturation state. A limitation of this study was the fact that spicule selection was random. No distinction was made between rod-like and plate-like spicules, nor spicule orientation. Qualitatively, the nine spicules analysed appeared to comprise a broad cross-section of spicule types and orientations. Additionally, our study did not account for variability in the tissue microstructure, including peaks and troughs in the lamellar structure and the orientation of collagen fibers. Future studies could address this final point by complementing the described techniques with polarized light microscopy of sections taken from the sample surface. Our multidisciplinary testing regime necessitated the use of dehydrated, PMMA embedded samples. This preparation has been demonstrated to result in overall higher elastic modulus, however this increase reflects a uniform offset rather than a change in range (Bushby et al., 2004). In fact, embedding in PMMA may be considered an asset as infilling of nanometer sized pores within the bone material may support the structure during nanoindentation. Finally, for our calculations of mechanical properties we
have assumed an isotropic material with constant Poisson’s ratio. While this is common practice for many nanoindentation studies, we recognize that bone is in fact an orthotropic material; this assumption therefore may contribute to some additional variability in our results. In this study we present new data describing the interrelationships between micro-scale elastic and plastic deformation properties and mineral content in trabecular bone, at the fundamental structural level of bone remodeling packets. Future work will seek to investigate potential alterations in these relationships with osteoporosis, and following anti-resorptive therapies.
Conflict of interest statement Jeffrey Schirer is an employee of Hysitron, Inc., manufacturer of the nanoindentation instrument used in this study. The authors have no other potential conflicts of interest to declare.
Acknowledgments This project was funded by a National Health and Medical Research Council Project Grant, an Australian Research Council Discovery Grant and a grant in aid from Alliance for Better Bone Health (Proctor and Gamble/Sanofi-Aventis). Cadaveric material donated through the Royal Adelaide Hospital is gratefully acknowledged. The program for grey level calibration and conversion was developed by Dr. Arash Badiei. The authors would also like to acknowledge Dr. Michelle Dickinson for her assistance with the initial study design, and Dr. Peter Sutton-Smith for providing technical advice.
References Boivin, G., Bala, Y., Doublier, A., Farlay, D., Ste-Marie, L.G., Meunier, P.J., Delmas, P.D., 2008. The role of mineralization and organic matrix in the microhardness of bone tissue from controls and osteoporotic patients. Bone 43 (3), 532–538. Boivin, G., Meunier, P.J., 2002. Changes in bone remodeling rate influence the degree of mineralization of bone. Connective Tissue Research 43 (2–3), 535–537. Brennan, O., Kennedy, O.D., Lee, T.C., Rackard, S.M., O’Brien, F.J., 2009. Biomechanical properties across trabeculae from the proximal femur of normal and ovariectomised sheep. Journal of Biomechanics 42 (4), 498–503. Bushby, A.J., Ferguson, V.L., Boyde, A., 2004. Nanoindentation of bone: comparison of specimens tested in liquid and embedded in polymethylmethacrylate. Journal of Materials Research 19 (1), 249–259. Busse, B., Hahn, M., Soltau, M., Zustin, J., Puschel, K., Duda, G.N., Amling, M., 2009. Increased calcium content and inhomogeneity of mineralization render bone toughness in osteoporosis: mineralization, morphology and biomechanics of human single trabeculae. Bone 45 (6), 1034–1043. Currey, J.D., 1988. The effect of porosity and mineral content on the Young’s modulus of elasticity of compact bone. Journal of Biomechanics 21 (2), 131–139. Fazzalari, N.L., Kuliwaba, J.S., Forwood, M.R., 2002. Cancellous bone microdamage in the proximal femur: influence of age and osteoarthritis on damage morphology and regional distribution. Bone 31 (6), 697–702. Ferguson, V.L., Bushby, A.J., Boyde, A., 2003. Nanomechanical properties and mineral concentration in articular calcified cartilage and subchondral bone. Journal of Anatomy 203 (2), 191–202. Ferguson, V.L., Bushby, A.J., Firth, E.C., Howell, P.G., Boyde, A., 2008. Exercise does not affect stiffness and mineralisation of third metacarpal condylar subarticular calcified tissues in 2 year old thoroughbred racehorses. European Cells and Materials 16, 40–46. Gao, H., Ji, B., Jager, I.L., Arzt, E., Fratzl, P., 2003. Materials become sensitive to flaws at nanoscale: lessons from nature. Proceedings of the National Academy of Science of the United States of America 13 (100), 5597–5600. Hadjidakis, D.J., Androulakis, I.I., 2006. Bone remodeling. Annals of the New York Academy of Sciences 1092, 385–396. Mittra, E., Akella, S., Qin, Y.X., 2006. The effects of embedding material, loading rate and magnitude, and penetration depth in nanoindentation of trabecular bone. Journal of Biomedical Materials Research Part A 79 (1), 86–93.
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Mulder, L., Koolstra, J.H., den Toonder, J.M., van Eijden, T.M., 2007. Intratrabecular distribution of tissue stiffness and mineralization in developing trabecular bone. Bone 41 (2), 256–265. Nalla, R.K., Kinney, J.H., Ritchie, R.O., 2003. Mechanistic fracture criteria for the failure of human cortical bone. Nature Materials 2 (3), 164–168. Norman, J., Shapter, J.G., Short, K., Smith, L.J., Fazzalari, N.L., 2008. Micromechanical properties of human trabecular bone: a hierarchical investigation using nanoindentation. Journal of Biomedical Materials Research Part A 87 (1), 196–202. Oliver, W.C., Pharr, G.M., 1992. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research 7 (6), 1564–1583. Oyen, M.L., 2006. Nanoindentation hardness of mineralized tissues. Journal of Biomechanics 39 (14), 2699–2702. Oyen, M.L., Ferguson, V.L., Bembey, A.K., Bushby, A.J., Boyde, A., 2008. Composite bounds on the elastic modulus of bone. Journal of Biomechanics 41 (11), 2585–2588.
3149
Paschalis, E.P., Recker, R., DiCarlo, E., Doty, S.B., Atti, E., Boskey, A.L., 2003. Distribution of collagen cross-links in normal human trabecular bone. Journal of Bone and Mineral Research 18 (11), 1942–1946. Roschger, P., Fratzl, P., Eschberger, J., Klaushofer, K., 1998. Validation of quantitative backscattered electron imaging for the measurement of mineral density distribution in human bone biopsies. Bone 23 (4), 319–326. Roschger, P., Paschalis, E.P., Fratzl, P., Klaushofer, K., 2008. Bone mineralization density distribution in health and disease. Bone 42 (3), 456–466. Ruppel, M.E., Miller, L.M., Burr, D.B., 2008. The effect of the microscopic and nanoscale structure on bone fragility. Osteoporosis International 19 (9), 1251–1265. Sakai, M., 1999. The Meyer hardness: a measure for plasticity. Journal of Materials Research 14 (9), 3630–3639. Sutton-Smith, P., Beard, H., Fazzalari, N., 2008. Quantitative backscattered electron imaging of bone in proximal femur fragility fracture and medical illness. Journal of Microscopy 229 (Pt 1), 60–66.
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
The role of mineral content in determining the micromechanical properties of discrete trabecular bone remodeling packets Lachlan J. Smith a, Jeffrey P. Schirer b, Nicola L. Fazzalari a,c,n a
Bone and Joint Research Laboratory, Surgical Pathology, SA Pathology, Adelaide, Australia. Applications Research Laboratory, Hysitron Inc., Minneapolis, USA. c Discipline of Pathology, School of Medical Sciences, Faculty of Health Sciences, University of Adelaide, Australia. b
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 29 July 2010
In trabecular bone, each remodeling event results in the resorption and/or formation of discrete structural units called ‘packets’. These remodeling packets represent a fundamental level of bone’s structural hierarchy at which to investigate composition and mechanical behaviors. The objective of this study was to apply the complementary techniques of quantitative backscattered electron microscopy (qBSEM) and nanoindentation to investigate inter-relationships between packet mineralization, elastic modulus, contact hardness and plastic deformation resistance. Indentation arrays were performed across nine trabecular spicules from 3 human donors; these spicules were then imaged using qBSEM, and discretized into their composite remodeling packets (127 in total). Packets were classified spatially as peripheral or central, and mean contact hardness, plastic deformation resistance, elastic modulus and calcium content calculated for each. Inter-relationships between measured parameters were analysed using linear regression analyses, and dependence on location assessed using Student’s t-tests. Significant positive correlations were found between all mechanical parameters and calcium content. Elastic modulus and contact hardness were significantly correlated, however elastic modulus and plastic deformation resistance were not. Calcium content, contact hardness and elastic modulus were all significantly higher for central packets than for peripheral, confirming that packet mineral content contributes to micromechanical heterogeneity within individual trabecular spicules. Plastic deformation resistance, however, showed no such regional dependence, indicating that the plastic deformation properties in particular, are determined not only by mineral content, but also by the organic matrix and interactions between these two components. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Trabecular bone Remodeling packet Mineralization Hardness Elastic modulus
1. Introduction Remodeling packets are the basic structural units of trabecular bone (Roschger et al., 2008). Each packet represents a discrete remodeling event, reflecting the continual process of resorption and formation that functions to repair microdamage, adapt existing structures in response to altered mechanical environment, and regulate calcium homeostasis (Fazzalari et al., 2002; Hadjidakis and Androulakis, 2006). Remodeling packets therefore represent a fundamental level of trabecular bone’s structural hierarchy at which to investigate the relationships between function and composition. Bone is comprised of both an organic component (an organized network of type I collagen fibers) and an inorganic component (hydroxyapatite nano-crystals bound to those collagen fibers)
n Corresponding author at: Bone and Joint Research Laboratory, Surgical Pathology, SA Pathology, Frome Rd, Adelaide, SA 5000, Australia. Tel.: + 61 8 82223269; fax: + 61 8 82223293. E-mail address: [email protected] (N.L. Fazzalari).
0021-9290/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2010.07.038
(Ruppel et al., 2008). The mineral component is considered to play a fundamental role in determining bone’s strength and toughness (Roschger et al., 2008). Each bone packet has a unique mineral content, which is determined not only by mechanobiology and calcium metabolism, but also by time since deposition (Hadjidakis and Androulakis, 2006; Roschger et al., 2008). At the microscopic level, inter-relationships between micromechanical properties and mineral content have been quantitatively examined in numerous tissues, including human femoral calcified cartilage and subchondral bone, iliac crest and several others, by combining the techniques of nanoindentation and backscattered electron microscopy (Ferguson et al., 2003; Ferguson et al., 2008; Oyen et al., 2008). These studies have demonstrated positive correlations between micromechanical properties and the mineral content of the tissue. Parameters classically obtained from indentation tests are elastic modulus and hardness (contact hardness) (Oliver and Pharr, 1992). These properties vary throughout the cross sections of individual spicules, with higher values present near the spicule core, and lower values present near the spicule periphery
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
(Brennan et al., 2009; Mulder et al., 2007; Norman et al., 2008). As older and more highly mineralized bone packets are generally located near the center of the spicule, and newer, less mineralized packets are located near the spicule periphery, there is an implicit relationship between packet mineralization and micromechanical properties. Indeed, this compositional and functional gradient is considered to impart superior macro-scale biomechanical behavior with respect to toughness and stability (Busse et al., 2009). Contact hardness is derived from both the elastic and plastic deformation properties of the material, and as such displays strong linear interdependence with elastic modulus (Oyen, 2006). An expression has been developed that recognizes this interdependence by assuming in-series contributions of elastic and plastic contributions to deformation, and has recently been proposed for the analysis of mineralized tissues (Oyen, 2006; Sakai, 1999) This expression allows derivation of a specific measure of resistance to plastic deformation, independent of elastic properties. Examination of this resistance to plastic deformation parameter and its relationship to composition and other micromechanical properties may provide new insights into the mechanisms that underlie failure of the bone composite at the nanoscale. Improved understanding of these factors may be particularly important in terms of understanding the long term effects of biological therapies that modulate the rate and extent of bone remodeling, which are used in the treatment of diseases such as osteoporosis. The objective of this study was to investigate the importance of mineral content on the micromechanical behavior of discrete remodeling packets in non-diseased human femoral trabecular bone. We hypothesized that calcium content would correlate positively with packet elastic modulus, contact hardness and plastic deformation resistance. Additionally, we hypothesized that both calcium content and micromechanical properties would be significantly higher for packets at the center of trabecular spicules, than for those lining the spicule periphery.
2. Methods 2.1. Specimen preparation With institutional research ethics committee and next of kin approvals, intertrochanteric, femoral, trabecular bone samples were obtained at autopsy from three female cases aged 44, 61 and 68. Each had no history of musculoskeletal disease, or other illnesses or drug treatments considered likely to affect bone metabolism, as indicated by case notes. The intertrochanteric region was selected due to its location distal to articular surfaces and common fracture sites. Specimens were washed in phosphate buffered saline, then dehydrated progressively in serially graded ethanols and embedded in polymethylmethacrylate (PMMA) resin, according to established methodologies (Mittra et al., 2006; Norman et al., 2008), and mounted on an aluminum puck. A carbon–aluminum standard was mounted adjacent to each surface to be analysed, and both were polished progressively, firstly with silicon–carbide sandpaper then with dry alumina cloths to achieve a mirror finish. Polishing quality was validated in a pilot study using the indentation instrument under tapping mode, which demonstrated that the peak to trough distance was consistently less than ten percent of the peak indentation depth. Polishing quality was also assessed qualitatively for each sample using a stereo microscope. 2.2. Nanoindentation Nanoindentation was performed using a TI-900 TriboIndenter (Hysitron Inc., Eden Prairie, MN, USA). Within each sample, three trabecular spicules were randomly selected for analysis using an optical microscope, and imaged to enable subsequent relocation using the electron microscope; additionally, deep reference indents were placed adjacent to each selected spicule for this same purpose. Indentations were performed under displacement control using a Berkovich diamond tip, pre-calibrated with fused quartz, according to the following protocol: a ten-second linear approach to a peak displacement of 400 nm, a 40-second hold at peak displacement and a ten-second linear withdrawal to zero displacement. This protocol was optimised to compensate for potential time and strain-rate
3145
dependent material behavior. Arrays of indents spaced horizontally and vertically at 30 mm intervals and filling the pre-defined spicule areas were implemented (a total of 1790 indents, 199 7 70 per spicule, (mean 7 SD)).
2.3. Microscopy and imaging Following nanoindentation, and without re-polishing in order to preserve indent locations, the surface of each sample was sputter-coated with carbon using a vacuum evaporator (Hitachi, Tokyo, Japan); specimens were then mounted in the chamber of a scanning electron microscope fitted with a backscattered electron detector (XL20; Philips, Amsterdam, Netherlands). Imaging was performed using an adaptation of established methods (Sutton-Smith et al., 2008). The microscope was switched on and left to stand for at least four hours prior to imaging for beam warm up and stabilization. Using an accelerating voltage of 15 kV, a working distance of 15 mm and at a magnification of 200 times, the carbon and aluminum standard grey levels were adjusted to approximately 30 and 225, respectively, by varying the contrast, brightness and spot size. Importantly, these standard grey levels were achieved for all three samples while maintaining a constant working distance. Each spicule that had been analysed using nanoindentation was then located from optical microscope images and imaged at 200 times magnification using backscattered electron (BSE) detection (Fig. 1A). Before and after each spicule the standards were imaged to account for minor temporal variations in the microscope beam current that might influence grey level stability. Indent locations were unable to be visualized easily using BSE detection; as such, following BSE imaging, multiple images covering the total area of each spicule were then taken at 1500 times magnification under secondary electron (SE) detection using an accelerating voltage of 10 kV, which improved surface contrast. These images were used to reconstruct a high resolution montage from which the precise locations of all indents could be mapped (Fig. 1B and C).
2.4. Analysis For each indent, contact hardness, Hc and elastic modulus, E were calculated according to the methods and definitions of Oliver and Pharr (1992) E¼
1n2B 1 Er
1n2I EI
where EI and nI are the elastic modulus and Poisson’s ratio of the diamond indenter (1140 GPa and 0.07, respectively), nB is the assumed Poisson’s ratio of bone (0.3) and Er is the reduced modulus of the sample, calculated from the formula pffiffiffiffi S p Er ¼ pffiffiffi 2 A where A is the projected contact area and S is the stiffness derived from an exponential curve fit to the unloading curve. Contact hardness, Hc was calculated using the formula Hc ¼
Pmax A
where Pmax is the maximum load. Additionally, plastic deformation resistance, H, was calculated using the model described by Sakai (1999) and previously proposed for mineralized tissue (Oyen, 2006). This model assumes in-series spring element contributions to elastic and plastic deformations, where the H is related to Hc and Er by the formula Hc ¼
1
a1 ðða2 Er Þ1=2 þ ða1 HÞ1=2 Þ2
with a1 ¼24.5 and a2 ¼4.4, being constants associated with the Berkovich indentation tip (Sakai, 1999). Using imaging software (Photoshop V5.0; Adobe Systems Inc., San Jose, CA, USA), indent locations from SE image montages were mapped to corresponding BSE images. Remodeling packets within each spicule were identified by tracing cement lines (Fig. 1D); each packet was then isolated and transferred to a separate image file for analysis. The mean hardness and elastic modulus for all indents falling within each packet was calculated. Indents falling close to bone-PMMA boundaries, close to or within osteocyte lacunae, or on cement lines were excluded from analyses. Grey level calibrations and conversions were carried out in accordance with previously published techniques (Roschger et al., 1998; SuttonSmith et al., 2008). Initially, a linear relationship between the average atomic numbers and grey levels of carbon, aluminum was confirmed against a hydroxyapatite standard. The relationship between calcium content and grey level was established from images of the hydroxyapatite standard and unmineralized collagen. The equation governing this relationship was wt% Ca ¼ 0.1733 grey level—4.3325. Our custom Matlab program (The Mathworks, Inc.; Natick, MA, USA) was able to calculate any minor temporal deviations in the grey levels of the carbon and aluminum standards from the target values of 25 and 225,
3146
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Fig. 1. (A) Representative BSE image of a trabecular spicule. (B) Corresponding SE image montage with mapped indents: ¼ included indents; ¼ indents excluded due to proximity to cell lacunae or spicule boundaries. (C) Higher magnification view showing actual indents spaced at 30 mm intervals. (D) BSEM image with traced remodeling packets and cross-mapped indents; ¼ additional indents excluded due to proximity to cement lines.
respectively, and correct the sample grey levels accordingly, before conversion to an estimate of calcium content.
2.5. Statistics For the analysis, packets from the nine spicules were initially pooled. The influence of calcium content on E, Hc and H was examined using linear regression analyses where strong, moderate and weak correlations were defined as r40.7, 0.7Z r 40.3 and r r0.3, respectively. Interdependence between Hc and H was examined in a similar way. Packets were then stratified into 2 groups, peripheral (with at least one edge at the spicule periphery) or central (no edges at the spicule periphery), and the same analyses performed. To examine whether interrelationships between properties varied between central and peripheral packets, the slopes and y-intercepts of the respective regression lines were compared using t-tests. Finally, differences in each of these properties between central and peripheral packets were examined using unpaired Student’s t-tests. All analyses were performed using GraphPad Prism V.5 (GraphPad Software Inc., San Diego, CA, USA), with 2-tailed pcritical ¼0.05.
3. Results A total of 127 discrete bone remodeling packets were resolved from the BSE images of the nine spicules, with a median of 8 (3–16) (interquartile range) indents falling within each packet. The mean7 SD peak load for all 1790 indents was 2.870.5 mN; a representative load vs. displacement curve is shown in Fig. 2. Mean7SD for E, Hc, H and Ca for all 127 packets were 22.347 3.01 GPa, 0.7470.07 GPa, 2.2070.25 GPa and 24.93 70.90 wt%, respectively. Linear regression analysis results demonstrated a strong positive correlation between Hc and E (r¼ 0.73, p o0.005, Fig. 3A); H and E, however, were not significantly correlated (r¼ 0.15, Fig. 3B). Calcium content correlated significantly with E, Hc and H (r ¼0.34, r¼ 0.61 and r ¼0.55, respectively, p o0.005, Fig. 4A–C). Of the 127 packets, 46 were classified as peripheral and the remaining 81 classified as central. Ca, E and Hc were found to be lower for peripheral packets (97%, p o0.005; 90%, po0.005 and 96%, p o0.05 of central, respectively, Fig. 5A–C). H, however, was not significantly different between the two groups (Fig. 5D).
Fig. 2. Representative force vs. displacement curve for a single indent, showing regions corresponding to loading, 40 s hold and unloading.
Linear regression results for peripheral and central packets analysed separately are summarized in Table 1. For central spicules, E and Hc, were strongly correlated, E and H, and Ca and E were each weakly correlated, and Ca and Hc, and Ca and H were each moderately correlated. For peripheral spicules, E and Hc, Ca and Hc and Ca and H were each moderately correlated, while E and H, and Ca and E were not significantly correlated. Neither the slopes or y-intercepts of the regression lines for each of these comparisons were significantly different between central and peripheral packets.
4. Discussion In this study we examined inter-relationships between mineral content and micromechanical properties for 127 discrete remodeling packets in human trabecular bone spicules. Mean values for elastic modulus and contact hardness determined here for samples from donors with no identifiable musculoskeletal disease are similar to those published previously (Norman et al., 2008), with differences likely attributable to the specifics of the
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
3147
Fig. 3. Linear regression plots comparing remodeling packet micromechanical properties. (A) Contact hardness, Hc and elastic modulus, E exhibit strong positive correlation. (B) Elastic modulus and plastic deformation resistance, H, are not correlated. Strong, moderate and weak correlations are defined as r 40.7, 0.7 Zr 40.3 and r r0.3, respectively.
Fig. 4. Linear regression plots comparing remodeling packet calcium content with micromechanical properties. (A) Calcium content, Ca and elastic modulus, E exhibit moderate positive correlation. (B) Calcium content and contact hardness, Hc exhibit moderate positive correlation. (C) Calcium content and plastic deformation resistance, H, exhibit moderate positive correlation. Strong, moderate and weak correlations are defined as r 40.7, 0.7Z r4 0.3 and rr 0.3, respectively.
Fig. 5. Comparison of micromechanical properties and calcium content between central and peripheral remodeling packets. (A) Calcium content, Ca, (B) elastic modulus, E and (C) contact hardness, Hc are all significantly higher for central packets. (D) Plastic deformation resistance, H, however was not significantly different between regions. n p o0.05, nnp o0.005.
Table 1 Correlations between mechanical properties and calcium content for central and peripheral remodeling packets. E ¼elastic modulus, Hc ¼contact hardness, H¼ plastic deformation resistance and Ca ¼ wt% calcium and strong, moderate and weak correlations are defined as r 40.7, 0.7 Zr 40.3 and r r 0.3, respectively. Central n ¼83 E vs. Hc E vs. H Ca vs. E Ca vs. Hc Ca vs. H
r¼ 0.78, r¼ 0.28, r¼ 0.28, r¼ 0.59, r¼ 0.67,
p o0.005 (strong) p o0.05 (weak) p o0.05 (weak) p o0.005 (moderate) p o0.005 (moderate)
Peripheral n¼ 44 r ¼0.61, r ¼0.09, r ¼0.12, r ¼0.63, r ¼0.69,
p o 0.005 (moderate) p 40.05 p 40.05 p o 0.005 (moderate) p o 0.005 (moderate)
testing protocol. Similarly, mean values for wt% Ca determined here are very close to those determined previously (Sutton-Smith et al., 2008). Our results show that the heterogeneity in elastic modulus and contact hardness through the cross sections of trabecular spicules (Norman et al., 2008) may be attributed, at least in part, to the distribution of bone remodeling packets with unique mineral compositions, a property determined by multiple factors including time since deposition (and hence rate of remodeling), the local micromechanical environment and the systemic regulation of calcium homeostasis (Boivin and Meunier, 2002; Hadjidakis and Androulakis, 2006; Roschger et al., 2008).
3148
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Correlations between mineral content, contact hardness and elastic modulus are consistent with the results of previous studies at the nano-, micro- and macro-scales (Boivin et al., 2008; Currey, 1988; Ferguson et al., 2003; Mulder et al., 2007). In particular, the relationship between mineralization and hardness of bone structural units in cortical bone from the iliac crest was recently examined using micro-indentation and X-ray diffraction experiments (Boivin et al., 2008). This study found a very similar correlation between contact hardness and mineral content (r¼ 0.6) to the current study, suggesting a relationship between these properties that potentially extends across orders of scale, and across different anatomical sites. In this study we have, for the first time, examined the relationship between calcium content and plastic deformation resistance. Interestingly, while calcium was found to be a significant predictor of plastic deformation resistance, plastic deformation resistance itself was found not to be significantly different between central and peripheral packets, despite the observation that packets in these locations have significantly different calcium contents. With respect to function, this suggests that trabecular spicules may possess heterogeneous elastic mechanical properties while at the same time maintaining an overall homogeneous resistance to plastic deformation. This finding highlights the complex inter-relationships between bone composition, and elastic and plastic mechanical behavior at the microscopic level. Bone has very small nano-crystals in a collagen network in contrast to relatively large crystals found in metals (Gao et al., 2003). Hydroxyapatite nano-crystals in bone do not fracture when bone fractures. Bone fractures as a result of disruption of the bonds between hydroxyapatite and collagen as well as bonds within the collagen matrix (Gao et al., 2003; Nalla et al., 2003). There is no evidence that plastic deformation in bone is associated with dislocations in hydroxyapatite crystals (in contrast to plastic deformation in metals). While mineral content has classically been used as the predictor of bone strength, there is increasing evidence that the organic matrix, particularly collagen cross-link biochemistry, plays a very important role in its mechanical performance. In fact, recent work has demonstrated that the organic matrix may account for as much as one third of the variation in contact hardness (Boivin et al., 2008). Like degree of mineralization, the nature of collagen intermolecular cross-links varies with distance from the spicule periphery (Paschalis et al., 2003). We hypothesize that the plastic deformation properties of bone packets are determined by both the interactions between hydroxyapatite crystals and the collagen fibers, as well as the degree of collagen cross-linking, both of which are likely to be strongly dependent on packet maturation state. A limitation of this study was the fact that spicule selection was random. No distinction was made between rod-like and plate-like spicules, nor spicule orientation. Qualitatively, the nine spicules analysed appeared to comprise a broad cross-section of spicule types and orientations. Additionally, our study did not account for variability in the tissue microstructure, including peaks and troughs in the lamellar structure and the orientation of collagen fibers. Future studies could address this final point by complementing the described techniques with polarized light microscopy of sections taken from the sample surface. Our multidisciplinary testing regime necessitated the use of dehydrated, PMMA embedded samples. This preparation has been demonstrated to result in overall higher elastic modulus, however this increase reflects a uniform offset rather than a change in range (Bushby et al., 2004). In fact, embedding in PMMA may be considered an asset as infilling of nanometer sized pores within the bone material may support the structure during nanoindentation. Finally, for our calculations of mechanical properties we
have assumed an isotropic material with constant Poisson’s ratio. While this is common practice for many nanoindentation studies, we recognize that bone is in fact an orthotropic material; this assumption therefore may contribute to some additional variability in our results. In this study we present new data describing the interrelationships between micro-scale elastic and plastic deformation properties and mineral content in trabecular bone, at the fundamental structural level of bone remodeling packets. Future work will seek to investigate potential alterations in these relationships with osteoporosis, and following anti-resorptive therapies.
Conflict of interest statement Jeffrey Schirer is an employee of Hysitron, Inc., manufacturer of the nanoindentation instrument used in this study. The authors have no other potential conflicts of interest to declare.
Acknowledgments This project was funded by a National Health and Medical Research Council Project Grant, an Australian Research Council Discovery Grant and a grant in aid from Alliance for Better Bone Health (Proctor and Gamble/Sanofi-Aventis). Cadaveric material donated through the Royal Adelaide Hospital is gratefully acknowledged. The program for grey level calibration and conversion was developed by Dr. Arash Badiei. The authors would also like to acknowledge Dr. Michelle Dickinson for her assistance with the initial study design, and Dr. Peter Sutton-Smith for providing technical advice.
References Boivin, G., Bala, Y., Doublier, A., Farlay, D., Ste-Marie, L.G., Meunier, P.J., Delmas, P.D., 2008. The role of mineralization and organic matrix in the microhardness of bone tissue from controls and osteoporotic patients. Bone 43 (3), 532–538. Boivin, G., Meunier, P.J., 2002. Changes in bone remodeling rate influence the degree of mineralization of bone. Connective Tissue Research 43 (2–3), 535–537. Brennan, O., Kennedy, O.D., Lee, T.C., Rackard, S.M., O’Brien, F.J., 2009. Biomechanical properties across trabeculae from the proximal femur of normal and ovariectomised sheep. Journal of Biomechanics 42 (4), 498–503. Bushby, A.J., Ferguson, V.L., Boyde, A., 2004. Nanoindentation of bone: comparison of specimens tested in liquid and embedded in polymethylmethacrylate. Journal of Materials Research 19 (1), 249–259. Busse, B., Hahn, M., Soltau, M., Zustin, J., Puschel, K., Duda, G.N., Amling, M., 2009. Increased calcium content and inhomogeneity of mineralization render bone toughness in osteoporosis: mineralization, morphology and biomechanics of human single trabeculae. Bone 45 (6), 1034–1043. Currey, J.D., 1988. The effect of porosity and mineral content on the Young’s modulus of elasticity of compact bone. Journal of Biomechanics 21 (2), 131–139. Fazzalari, N.L., Kuliwaba, J.S., Forwood, M.R., 2002. Cancellous bone microdamage in the proximal femur: influence of age and osteoarthritis on damage morphology and regional distribution. Bone 31 (6), 697–702. Ferguson, V.L., Bushby, A.J., Boyde, A., 2003. Nanomechanical properties and mineral concentration in articular calcified cartilage and subchondral bone. Journal of Anatomy 203 (2), 191–202. Ferguson, V.L., Bushby, A.J., Firth, E.C., Howell, P.G., Boyde, A., 2008. Exercise does not affect stiffness and mineralisation of third metacarpal condylar subarticular calcified tissues in 2 year old thoroughbred racehorses. European Cells and Materials 16, 40–46. Gao, H., Ji, B., Jager, I.L., Arzt, E., Fratzl, P., 2003. Materials become sensitive to flaws at nanoscale: lessons from nature. Proceedings of the National Academy of Science of the United States of America 13 (100), 5597–5600. Hadjidakis, D.J., Androulakis, I.I., 2006. Bone remodeling. Annals of the New York Academy of Sciences 1092, 385–396. Mittra, E., Akella, S., Qin, Y.X., 2006. The effects of embedding material, loading rate and magnitude, and penetration depth in nanoindentation of trabecular bone. Journal of Biomedical Materials Research Part A 79 (1), 86–93.
L.J. Smith et al. / Journal of Biomechanics 43 (2010) 3144–3149
Mulder, L., Koolstra, J.H., den Toonder, J.M., van Eijden, T.M., 2007. Intratrabecular distribution of tissue stiffness and mineralization in developing trabecular bone. Bone 41 (2), 256–265. Nalla, R.K., Kinney, J.H., Ritchie, R.O., 2003. Mechanistic fracture criteria for the failure of human cortical bone. Nature Materials 2 (3), 164–168. Norman, J., Shapter, J.G., Short, K., Smith, L.J., Fazzalari, N.L., 2008. Micromechanical properties of human trabecular bone: a hierarchical investigation using nanoindentation. Journal of Biomedical Materials Research Part A 87 (1), 196–202. Oliver, W.C., Pharr, G.M., 1992. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research 7 (6), 1564–1583. Oyen, M.L., 2006. Nanoindentation hardness of mineralized tissues. Journal of Biomechanics 39 (14), 2699–2702. Oyen, M.L., Ferguson, V.L., Bembey, A.K., Bushby, A.J., Boyde, A., 2008. Composite bounds on the elastic modulus of bone. Journal of Biomechanics 41 (11), 2585–2588.
3149
Paschalis, E.P., Recker, R., DiCarlo, E., Doty, S.B., Atti, E., Boskey, A.L., 2003. Distribution of collagen cross-links in normal human trabecular bone. Journal of Bone and Mineral Research 18 (11), 1942–1946. Roschger, P., Fratzl, P., Eschberger, J., Klaushofer, K., 1998. Validation of quantitative backscattered electron imaging for the measurement of mineral density distribution in human bone biopsies. Bone 23 (4), 319–326. Roschger, P., Paschalis, E.P., Fratzl, P., Klaushofer, K., 2008. Bone mineralization density distribution in health and disease. Bone 42 (3), 456–466. Ruppel, M.E., Miller, L.M., Burr, D.B., 2008. The effect of the microscopic and nanoscale structure on bone fragility. Osteoporosis International 19 (9), 1251–1265. Sakai, M., 1999. The Meyer hardness: a measure for plasticity. Journal of Materials Research 14 (9), 3630–3639. Sutton-Smith, P., Beard, H., Fazzalari, N., 2008. Quantitative backscattered electron imaging of bone in proximal femur fragility fracture and medical illness. Journal of Microscopy 229 (Pt 1), 60–66.