The complexity and heterogeneity of bone material

Barbara M. MISOF, PhD
Ludwig Boltzmann Institute of Osteology, Hanusch
Hospital of WGKK and AUVA Trauma Centre
Meidling, 1st Med. Dept.
Hanusch Hospital – Vienna, AUSTRIA

The complexity and heterogeneity of bone material

by P. Roschger, B. M. Misof, and K. Klaushofer, Austria

Bone has a complex hierarchical structure which provides outstanding mechanical performance at minimal requisite mass. The geometry and inner architecture of trabecular and compact bone represent themacroand mesoscopic structural levels. At the lowest hierarchical level, bone material is a composite of two components with divergent mechanical characteristics: soft type I collagen fibrils and stiff calcium phosphate particles a few nanometers thick. To this basic heterogeneity of the composite at nanoscale is addedmicroscale heterogeneity in the formof continuous remodeling of the bone matrix. This process generates bone packets with different matrix mineralization and lamellar orientations coexisting within the bone material. Nondestructive techniques with high spatial resolution are therefore required to characterize material structure-function relationships in normal and diseased bone. The present article focuses on the characteristics of the material levels in bone, in particular on the two components collagen and mineral, the mineralized collagen fibril, the lamellar arrangement of fibrils, and bone packets in normal and diseased bone. Technical advances in recent years have yielded specific insights into the structural hierarchy of bone and a better appreciation of the impact of disease on bone material and its mechanical properties. This in turn should deepen understanding of the underlying pathophysiology, enhance prediction of fracture risk, and inform therapeutic decision-making.

Medicographia. 2012;34:155-162 (see French abstract on page 162)

Bone is an outstanding material in that it adapts to changes in mechanical demand and is self-healing. Normal function requires proper interplay between all sizes of its structural components.1 Bone is a lightweight structure providing maximal mechanical strength at minimal requisite mass thanks to a complex hierarchical organization extending from the nano , through the micro and meso, to the macroscopic level (Figure 1, page 156). The outer geometry and inner architecture of cancellous and cortical bone are clearly evident at the macro- and mesoscopic levels. However, the increase in fracture risk associated with aging or disease depends not only on the amount of bone, but also on its material properties.

This article focuses on the lower hierarchical levels comprising the intrinsic bone material. Technical advances have thrown fresh light on the structure-function relationships of the different components down to the nano level, thereby advancing our understanding of this material’s remarkable mechanical competence. Bone material is made up of four hierarchical levels of structural organization: the lowest (nano) level features collagen and mineral as the main components of the nanocomposite forming the mineralized fibrils. These fibrils are arranged in lamellae and bone structural units (BSUs, also known as bone packets or osteons).

Figure 1
Figure 1. The hierarchical structure of bone.

The 4-level structural hierarchy of bone material, beginning at the nanometer level with collagen and mineral, the basic components of the nanocomposite material organized into mineral fibrils, which are then arranged in lamellae and finally in bone packets (the basic structural unit of bone material formed by osteoblasts during a remodeling cycle).

Collagen and mineral

_ Collagen structure and cross-linking
The organic matrix of bone consists mainly of collagen type I, a triple helix of two α1 and one α2 collagen chains. It is synthesized by osteoblasts, assembled extracellularly into fibrils, and stabilized by cross-links.2 The matrix not only serves as a scaffold for the mineral in the composite material, but itself plays a decisive role in the biomechanical competence of the collagen-mineral composite described later. For instance, we know that the decrease in mechanical competence of the organic matrix with increasing age is partly responsible for bone fragility.3 Many characteristics of organic matrix, such as amount produced by the cell, fibril structure, and the character, number, and distribution of the fibril cross-links, are genetically determined. Consequently, mutations not only in the collagen encoding genes, but also in the proteins involved in synthesis can have an impact on the resulting material properties. This is exemplified in osteogenesis imperfecta (OI), which is caused by a variety of genetic mutations.4-8 The mechanical consequences of abnormal collagen molecules have been demonstrated in the OI mouse (oim), an animal model for human OI. Affected animals lose 50% of their tendon collagen strength compared with wild-type littermates.9 Indeed, the properties of collagen are also essential for the plastic (post-yield) behavior of bone during tension.10

The properties of collagen/organic matrix and specific noncollagenous proteins can be chemically analyzed with high spatial resolution by spectroscopic techniques such as Fourier transform infrared (FTIR) imaging11 and Raman microspectroscopy.12 FTIR imaging measures the absorption of infrared radiation at specific wavelengths. This is dependent on the molecular bonds and their modes of vibration. The absorbance spectra reveal the characteristics of collagen (amide bands and cross-links) at a typical spatial resolution of a few micrometers.11 In Raman microspectroscopy, the bone sample is irradiated by monochromatic laser light and the inelastically scattered light from the specimen is measured. The resulting vibrational bands are generally very sharp, enabling even small band shifts to be detected. High spatial resolution in the 1-μm range permits the analysis of selective small regions, such as newly formed bone between fluorescence-labeled bands.12

These techniques have shown newly formed bone material to possessmore immature divalent cross-links destined tomature into trivalent cross-links. Deviations from the normal ratio of trivalent over divalent cross-links (collagen cross-link ratio) are associated with bone fragility.11 Additionally, the formation of advanced glycation end products (AGEs) has come into focus during recent years. Increased AGEs are thought to predispose to bone fragility in postmenopausal osteoporosis and diabetes.13

_ Bone mineral
The inorganic component of bone material is the mineral consisting essentially of nanosized platelets (particles, crystals) made of carbonated hydroxyapatite (Ca5(PO4)3OH). However, the chemical composition of these platelets is not constant, but can change during mineralization and maturation. In particular, substitution of calciumand phosphate ions is frequent.14 The platelets are about 60 nm long15 and a few nanometers thick,16,17 and are embedded in collagen matrix. Their microscopic size is essential for the mechanical properties of the resulting nanocomposite and, moreover, makes platelet strength insensitive to flaws.18

Mineral characteristics such as maturity/crystallinity can be obtained from the intensity ratios of vibrational bands for phosphate, carbonate, and other constituents measured by FTIR imaging or Raman microspectroscopy. Other mineral characteristics, such as size, shape, and alignment, can be assessed from conventional histological bone sections by x-ray scattering down to micrometer resolutions using synchrotron radiation (SR). Small angle x-ray scattering (SAXS) measures the scattered x-ray intensities within angles no larger than 1° with respect to the direction of the incident beam. The intensities reflect objects between 1 and 50 nm thick within the sample. Information is given on the thickness, shape, and alignment of the mineral platelets.16 Transmission electron microscopy (TEM) can also be used to characterize platelet size and shape,15 but it is not applicable in routine as it needs sophisticated preparation, is rather time-consuming, and provides information only on single objects and not on an average of millions of particles, as does SAXS.

Figure 2
Figure 2. The effect of sodium fluoride treatment on trabecular bone structure.

Top: Backscattered electron image of part of a bone trabeculum from a patient treated with sodium fluoride. White circles: areas for synchrotron small angle x-ray scattering measurement. Bone formed during treatment (circles A and C) reveals altered structure compared with bone present before treatment (circle B).
Bottom: Corresponding G(x)-curves for the areas A, B, and C. G(x) was obtained from small angle x-ray scatter measurements and gives information on the shape and size of the mineral particles. G(x) differs qualitatively at positions A and C compared with position B.
Unpublished material related to reference 21.

Scattered x-ray intensities under wider angles (scanning xray diffraction [XRD]) give information about the crystal/lattice structure of the mineral particles. A useful elemental analysis technique is SR induced micro x-ray fluorescence (SR-μXRF). It measures the elemental distribution of trace elements, such as strontium and lead, with high sensitivity down to the ppm range.19 In contrast, electron-induced μXRF, as used in the scanning electron microscope (SEM)/energy-dispersive x-ray spectroscopy (EDX), is much less sensitive (limited to about 0.5 weight % elemental concentration). SAXS has revealed that average mineral platelet thickness increases rapidly in the first four years of life, then slows.20 It has also shown that sodium fluoride (NaF) administration significantly alters the size distribution of mineral particles in osteoporotic patients.21 Fluoride is incorporated into the mineral crystals and changes their chemical composition, size, and crystallinity (Figure 2). The resulting abnormalities in bone material may explain the absence of any increase in mechanical competence despite higher bone volume after NaF treatment.22 Strontium and lead are two other bone-seeking chemical elements. Strontium gets incorporated into newly formed bone packets during strontium ranelate therapy. Depending on the patient’s serum levels, strontium exchanges for up to 5% of calcium ions.23,24 The incorporated element does not modify the local mechanical properties (nanoindentation) or collagen crosslinking.23 As for lead, normal environmental exposure results in storage in bone mineral, specifically (up to 13 fold) in the tide mark of the transition zone between bone and articular cartilage, and within cement lines.19

Mineralized collagen fibrils

The main building blocks of bone are mineralized collagen fibrils. These form the composite material consisting of the soft yet tough protein stiffened by the hard and brittle mineral platelets. The fibrils are about 100 nm15 in diameter and consist of collagen molecules staggered in parallel, but displaced by 67 nm, producing a structure with overlap and hole zones25 in which the mineral particles are thought to nucleate and grow. Not much is yet known about the nucleation of the mineral particles and their growth to final size. It has been suggested that both collagen itself26 and noncollagenous proteins27 significantly influence these processes. Interestingly, the amount of mineral with in the organic matrix is abnormally high, but similar in different forms of OI with altered or normal collagen structure,7 suggesting that the structure of the collagen itself is not the only regulator of the amount of mineral deposited.

Insight into the mechanical properties of mineralized collagen fibrils is essential for understanding whole-tissue mechanics. In general, the mineral platelets are oriented with their long axis (c axis) parallel to the long axis of the collagen fibril.15,28 Mechanical models have shown that the organic matrix in between the mineral platelets transfers the shear forces acting on bone material. The main determinant of the elasticity and hardness of the material is assumed to be the amount of mineral. However, platelet shape and large aspect ratio (length or width over thickness) are also important for stiffening the collagen fibrils.29 Sophisticated techniques using in-situ mechanical testing with high spatial resolution structural analysis at the European Synchrotron Radiation Facility (Grenoble, France [ESRF]) were recently introduced for studying the interface between collagen and mineral platelets, since the platelets are thought to be important for the deformation behavior of mineralized collagen fibrils. These tensile loading experiments revealed that bone material does not deform homogeneously in response to the external load, but the larger elements take up more strain than the small stiff elements, with characteristic strain contributions of 12:5:2 for whole tissue, fibril, and mineral, respectively, in the elastic deformation region.30 The results favor a staggered arrangement of mineral platelets within each fibril, and a staggered arrangement of fibrils to fibers and tissue within the extrafibrillar matrix (Figure 3).31

Figure 3
Figure 3. Schematic model for bone deformation in response to external tensile load at three levels in the structural hierarchy.

In response to an external load (elastic region of deformation), the bone material does not deform homogeneously, but the larger structures take up more strain than the small, stiff elements. This results in characteristic strain contributions of 12:5:2 for whole tissue, fibril, and mineral, respectively.
Modified from reference 30: Gupta et al. Proc Natl Acad Sci U S A. 2006;103(47):17741-17746. © 2006, The National Academy of Sciences.

Bone lamellae

In mature bone, themineralized collagen fibrils are assembled into lamellae.Within each lamella the fibrils are in a predominant orientation that changes from lamella to lamella. During bone formation, collagen is deposited onto the actual bone surface in a way that orients the lamellae parallel to this bone surface. This is mirrored by the orientation of the long axis of themineral platelets, which follows the direction of the trabeculae.20 Because of the birefringent properties of collagen fibrils, lamellar orientation can be clearly visualized under polarized light microscopy. Raman microspectroscopy has revealed the lamellar organization of the bone matrix by showing how the scattering intensities of certain collagen and phosphate bands strongly depend on the angle between fibril orientation, laser light polarization, and beam axis direction (Figure 4).32 Scanning SAXS combined with XRD has confirmed the rotated plywood arrangement of the fibrils consistent with the observed lamellar structures.28

Mechanical testing has shown that deformation is also not homogeneous at the lamellar level, but distributed between tensile deformation of the fibrils and shearing in the interfibrillar matrix (the so-called “glue”).31 Fibril anisotropy within the lamellae is essentially responsible for the high anisotropy in mechanical behavior found in bone material, in terms, for instance, of the nanoelasticity and nanohardness data obtained by nanoindentation.33 In this technique, the tip of an atomic force microscope is used as a nanosized indenter to measure the local elastic response of the sample in the submicrometer range. Evaluation of the elastic modulus and hardness is based on the measurement of load, indentation depth, and projected indentation area (contact area). The results greatly depend on the local orientation of lamellae and fibrils with respect to the indentation axis.34 Moreover, deformation experiments have clearly shown that lamellar organization is essential in controlling crack propagation:35 the energy required to propagate a crack is about two orders higher if the crack is perpendicular rather than parallel to the lamellar plane. In consequence, the impaired lamellar structure found in patients with pycnodysostosis36 is probably responsible for the bone fragility observed in this genetic disease.

Figure 4
Figure 4. Raman images of cortical bone with two Haversian canals in front of the imaged region.

The arrows indicate the polarization orientation of the laser beam. Note that dependent on the direction of polarization, the lamellae change their contrast in the images (those appearing dark in longitudinal orientation appear bright in transverse laser polarization and vice versa).
Modified from reference 32: Kazanci et al. Bone. 2007;41(3):456-461. © 2007, Elsevier B.V.

Figure 5
Figure 5. Formation of bone mineralization density distribution.

A. Backscattered electron images of trabecular area (left) in a parathyroid hormone-treated osteoporotic male patient and the corresponding bone mineralization density distribution (BMDD) represented by the solid red line (right). The dotted white line and the gray area in the background show the mean and 95% confidence interval of the normal reference BMDD, respectively.
B. Mineralization density profile perpendicular to the mineralization front from the enlarged trabeculum (left). In the first few micrometers from the mineralization front, the calcium content of the matrix increases sharply, and then flattens out further from the mineralization front. Using the tissue age data from fluorescence labeling of the bone sample, this profile can be assigned to the time sequence of mineralization (primary or secondary).
Abbreviations: BSU, basic structural unit; Ca, calcium; P, primary mineralization phase; PTH, parathyroid hormone; S, secondary mineralization phase.
Modified from reference 38: Roschger et al. Bone. 2008;42(3):456-466. © 2008, Elsevier B.V.

Basic structural units

As bone is remodeled throughout life, it is not a homogeneous material, but consists of tissue volumina of differing ages. The mean lifespan of bone material varies from a few to about 20 years, depending on the site considered.37 In consequence, bone material consists of younger and older bone volumina in BSUs, or bone packets for trabecular bone and osteons for compact bone. Each represents the amount of bone material formed by osteoblasts within one remodeling cycle. Consequently, these BSUs differ in lamellar orientation and mineral content. The differences in degree of mineralization are caused by the characteristic time courses involved. Newly formed unmineralized osteoid starts to mineralize after about 14 days of mineralization lag time. In a first rapid phase of primary mineralization, about 70% of the final mineral content is deposited within a few days (Figure 5).38 This is followed by much slower phases of secondary mineralization which last months to years before achieving the final mineral content,38-41 typically in interstitial bone.38 In pathological cases, however, final mineral content at BSU level can achieve values that are higher (hypermineralization) or lower (hypomineralization) than in normal interstitial bone, due to altered organic matrix, disordered mineralization kinetics, and/or changes inmineral particle size.38

In general, the amount of mineral and its distribution within the collagen matrix are important determinants of stiffness and strength.42 This is reflected in the positive correlation found between mineral content and nanoindentation outcomes.43 However, there is remarkable scatter in indentation outcomes at a given calcium content.44 This is because the measurement of mineral may be independent of orientation, but local elasticity and nanohardness greatly depend on the actual orientation of the mineralized collagen fibrils.34,44 Moreover, BSU heterogeneity in mineral content and lamellar orientation (including cement lines) may be essential in reducing and/or preventing crack propagation.43,45,46 However, bonematerial heterogeneity, especially at the bone surface, can also be a source of crack initiation. Local small changes in the mechanical properties of composite material are unlikely to affect the overall stiffness or modulus of bone material (considered as an average of local properties), but may have a profound impact on material strength, which depends essentially on the strength of its weakest component (in the same way as the weakest link determines the strength of a chain). Thus, it is difficult to predict the extent to which the heterogeneity introduced by BSUs benefits or impairs mechanical performance.

Microradiography,41,47 SR microcomputed tomography (SR μCT)48 and quantitative backscattered electron imaging (qBEI)6,38 are the techniques used to measure the intrinsic bone mineralization pattern. Microradiography has been used for this purpose for some decades. However, it requires bone sections about 100 μm thick, leading to partial volume effects (mimicking lower mineral content) that impair its accuracy. SR μCT is by comparison a rather novel technique analogous to x-ray tomography, but with a specific beam energy (monochromatic beam) and high spatial resolution (a few micrometers in beam diameter) that confer the bonus of full 3D information. However, its disadvantage is that it is time-consuming and requires the synchrotron facility, thereby excluding conventional applications. Attempts to harness laboratory μCT devices (usually used to measure structural indices of bone microarchitecture) to measure matrix mineralization have to be viewed with caution because of the beam-hardening effects introduced by the non-monochromatic x-ray beam used by such devices. Yet another technique for analyzing mineralization density is qBEI, a 2D method that gives information from the bone surface layer <1.5 μm in thickness, but provides high spatial resolution and sensitivity in the detection of different degrees of mineralization. In bone, the backscattered electron intensity signal is dominated by its calcium content. Thus, the different gray levels in qBEI images reflect different local calcium concentrations and can be further analyzed in histograms revealing the percentage of bone areas with a specific calcium concentration (bone mineralization density distribution [BMDD]). Comprehensive studies on samples from healthy adult individuals have shown minor variations in cancellous BMDD with gender, ethnicity, and skeletal site.47,49 These amazingly small variations may indicate that BMDD is an evolutionary optimum in terms of biology andmechanics. Thus, even small deviations from the normal BMDD appear of biological and/or clinical relevance in that they reflect altered mineralization kinetics and/or bone turnover rates (Figure 5). Metabolic bone diseases and treatments that increase bone turnover push BMDD values towards lower mineralization and/or into a wider range (ie, more heterogeneousmatrixmineralization),38,47 as in postmenopausal osteoporosis.38 Antiresorptive treatments that reduce bone turnover (alendronate, risedronate, zoledronic acid, etc), on the other hand, shift BMDD values from lower to normal calcium concentrations and transiently narrow the range (more homogeneousmatrixmineralization).38 Higher than normal degrees of mineralization have been found in patients with different types of OI, making their bone harder and more brittle than normal.6-8 Interestingly, this shift of BMDD to higher calcium concentrations occurs despite an increased bone turnover, indicating that mineralization must be accelerated in this genetic disease, leading to a hypermineralized matrix. Mathematical modeling of BMDD has greatly contributed to the understanding and interpretation of experimentally measured BMDD in health and disease.50


Bone material is heterogeneous and anisotropic based on the hierarchical organization of its principal components (collagen and mineral) into several structural levels starting from the nanometer scale. Metabolic and genetic disease associated with bone fragility usually affects one or more of these structural levels, indicating that the normal structure of healthy bone material is optimized for mechanical performance. Any deviation fromnormal is likely to compromisemechanical competence. Yet it still remains difficult to predict mechanical strength from changes in single structural components due to the complex interplay of all structural levels and the sensitivity of material strength to local defects, according to the principle of the weakest link in a chain. _

Acknowledgement. We acknowledge all the colleagues from the Ludwig Boltzmann Institute of Osteology who have contributed to the work described. In particular, we want to express our gratitude to Peter Fratzl (Max Planck Institute of Colloids and Interfaces, Golm, Germany) for longstanding cooperation in the analysis of bone tissue properties. We also wish to thank the Vienna Health Insurance Fund (WGKK), Austrian Workers’ Compensation Board (AUVA), and Ludwig Boltzmann-Gesellschaft for their financial support of our work.

1. Fratzl P, Gupta HS, Paschalis EP, Roschger P. Structure and mechanical quality of the collagen-mineral nano-composite in bone. J Mater Chem. 2004;14: 2115-2123.
2. Hulmes DJ. Building collagen molecules, fibrils, and suprafibrillar structures. J Struct Biol. 2002;137(1-2):2-10.
3. Wang X, Shen X, Li X, Agrawal CM. Age-related changes in the collagen network and toughness of bone. Bone. 2002;31(1):1-7.
4. Rauch F, Glorieux FH. Osteogenesis imperfecta. Lancet. 2004;363(9418): 1377-1385.
5. Marini JC, CabralWA, Barnes AM. Null mutations in LEPRE1 and CRTAP cause severe recessive osteogenesis imperfecta. Cell Tissue Res. 2010;339(1):59-70.
6. Boyde A, Travers R, Glorieux FH, Jones SJ. The mineralization density of iliac crest bone from children with osteogenesis imperfecta. Calcif Tissue Int. 1999; 64(3):185-190.
7. Roschger P, Fratzl-Zelman N, Misof BM, Glorieux FH, Klaushofer K, Rauch F. Evidence that abnormal high bone mineralization in growing children with osteogenesis imperfecta is not associated with specific collagen mutations. Calcif Tissue Int. 2008;82(4):263-270.
8. Fratzl-Zelman N, Morello R, Lee B, et al. CRTAP deficiency leads to abnormally high bone matrix mineralization in a murine model and in children with osteogenesis imperfecta type VII. Bone. 2010;46(3):820-826.
9. Misof K, LandisWJ, Klaushofer K, Fratzl P. Collagen from the osteogenesis imperfecta mouse model (oim) shows reduced resistance against tensile stress. J Clin Invest. 1997;100(1):40-45.
10. Jepsen KJ, Goldstein SA, Kuhn JL, Schaffler MB, Bonadio J. Type-I collagen mutation compromises the post-yield behavior of Mov13 long bone. J Orthop Res. 1996;14(3):493-499.
11. Paschalis EP, Mendelsohn R, Boskey AL. Infrared assessment of bone quality: a review. Clin Orthop Relat Res. 2011;469(8):2170-2178.
12. Gamsjaeger S, Buchinger B, Zwettler E, et al. Bone material properties in actively bone-forming trabeculae in postmenopausal women with osteoporosis after three years of treatment with once-yearly zoledronic acid. J Bone Miner Res. 2011;26(1):12-18.
13. Saito M, Marumo K. Collagen cross-links as a determinant of bone quality: a possible explanation for bone fragility in aging, osteoporosis, and diabetes mellitus. Osteoporos Int. 2010;21(2):195-214.
14. Posner AS. The structure of bone apatite surfaces. J Biomed Mater Res. 1985; 19(3):241-250.
15. Rubin MA, Jasiuk I, Taylor J, Rubin J, Ganey T, Apkarian RP. TEM analysis of the nanostructure of normal and osteoporotic human trabecular bone. Bone. 2003;33(3):270-282.
16. Fratzl P, Groschner M, Vogl G, et al. Mineral crystals in calcified tissues: a comparative study by SAXS. J Bone Miner Res. 1992;7(3):329-334.
17. Weiner S, Wagner HD. The material bone: Structure-mechanical function relations. Annu Rev Mater Sci. 1998;28:271-298.
18. Gao H, Ji B, Jager IL, Arzt E, Fratzl P. Materials become insensitive to flaws at nanoscale: lessons from nature. Proc Natl Acad Sci U S A. 2003;100(10):5597- 5600.
19. Zoeger N, Roschger P, Hoefstaetter JG, et al. Lead accumulation in the tidemark of articular cartilage. Osteoarthritis Cartilage. 2006;14(9):906-913.
20. Roschger P, Grabner BM, Rinnerthaler S, et al. Structural development of the mineralized tissue in the human L4 vertebral body. J Struct Biol. 2001;136(2): 126-136.
21. Gourrier A, Li C, Seigel S, et al. Scanning small-angle X-ray scattering analysis of the size and organization of the mineral nanoparticles in fluorotic bone using a stack of cards model. J Appl Cryst. 2010;43(6):1385-1392.
22. Fratzl P, Schreiber S, Roschger P, Lafage MH, Rodan G, Klaushofer K. Effects of sodium fluoride and alendronate on the bone mineral in minipigs: a smallangle X-ray scattering and backscattered electron imaging study. J Bone Miner Res. 1996;11(2):248-253.
23. Roschger P, Manjubala I, Zoeger N, et al. Bone material quality in transiliac bone biopsies of postmenopausal osteoporotic women after 3 years of strontium ranelate treatment. J Bone Miner Res. 2010;25(4):891-900.
24. Li C, Paris O, Siegel S, et al. Strontium is incorporated into mineral crystals only in newly formed bone during strontium ranelate treatment. J Bone Miner Res. 2010;25(5):968-975.
25. Hodge AJ, Petruska JA. Recent studies with the electron microscope on ordered aggregates of the tropocollagen molecule. In: Ramachandran GN, ed. Aspects of protein structure. New York, NY: Academic Press; 1963;289-300.
26. Landis WJ, Silver FH. Mineral deposition in the extracellular matrices of vertebrate tissues: identification of possible apatite nucleation sites on type I collagen. Cells Tissues Organs. 2009;189(1-4):20-24.
27. Boskey AL. Matrix proteins and mineralization: an overview. Connect Tissue Res. 1996;35(1-4):357-363.
28. Wagermaier W, Gupta HS, Gourrier A, et al. Scanning texture analysis of lamellar bone using microbeam synchrotron X-ray radiation. J Appl Cryst. 2007;40(1): 115-120.
29. Jager I, Fratzl P. Mineralized collagen fibrils: a mechanical model with a staggered arrangement of mineral particles. Biophys J. 2000;79(4):1737-1746.
30. Gupta HS, Seto J, Wagermaier W, Zaslansky P, Boesecke P, Fratzl P. Cooperative deformation of mineral and collagen in bone at the nanoscale. Proc Natl Acad Sci U S A. 2006;103(47):17741-17746.
31. Gupta HS, Wagermaier W, Zickler GA, et al. Nanoscale deformation mechanisms in bone. Nano Lett. 2005;5(10):2108-2111.
32. Kazanci M, Wagner HD, Manjubala NI, et al. Raman imaging of two orthogonal planes within cortical bone. Bone. 2007;41(3):456-461.
33. Zysset PK. Indentation of bone tissue: a short review. Osteoporos Int. 2009;20 (6):1049-1055.
34. Rho JY, Tsui TY, Pharr GM. Elastic properties of human cortical and trabecular lamellar bone measured by nanoindentation. Biomaterials. 1997;18(20): 1325-1330.
35. Peterlik H, Roschger P, Klaushofer K, Fratzl P. From brittle to ductile fracture of bone. Nat Mater. 2006;5(1):52-55.
36. Fratzl-Zelman N, Valenta A, Roschger P, et al. Decreased bone turnover and deterioration of bone structure in two cases of pycnodysostosis. J Clin Endocrinol Metab. 2004;89(4):1538-1547.
37. Parfitt AM. Misconceptions (2): turnover is always higher in cancellous than in cortical bone. Bone. 2002;30(6):807-809.
38. Roschger P, Paschalis EP, Fratzl P, Klaushofer K. Bone mineralization density distribution in health and disease. Bone. 2008;42(3):456-466.
39. Akkus O, Polyakova-Akkus A, Adar F, Schaffler MB. Aging of microstructural compartments in human compact bone. J Bone Miner Res. 2003;18(6):1012- 1019.
40. Fuchs RK, Allen MR, Ruppel ME, et al. In situ examination of the time-course for secondary mineralization of Haversian bone using synchrotron Fourier transform infrared microspectroscopy. Matrix Biol. 2008;27(1):34-41.
41. Bala Y, Farlay D, Delmas PD, Meunier PJ, Boivin G. Time sequence of secondary mineralization and microhardness in cortical and cancellous bone from ewes. Bone. 2010;46(4):1204-1212.
42. Currey JD. The mechanical consequences of variation in the mineral content of bone. J Biomech. 1969;2(1):1-11.
43. Gupta HS, Schratter S, Tesch W, et al. Two different correlations between nanoindentationmodulus andmineral content in the bone-cartilage interface. J Struct Biol. 2005;149(2):138-148.
44. Oyen ML, Ferguson VL, Bembey AK, Bushby AJ, Boyde A. Composite bounds on the elastic modulus of bone. J Biomech. 2008;41(11):2585-2588.
45. Fratzl P. Bone fracture: When the cracks begin to show. Nat Mater. 2008;7(8): 610-612.
46. Zioupos P, Gresle M, Winwood K. Fatigue strength of human cortical bone: age, physical, and material heterogeneity effects. J Biomed Mater Res A. 2008; 86(3):627-636.
47. Boivin G, Meunier PJ. The degree of mineralization of bone tissue measured by computerized quantitative contact microradiography. Calcif Tissue Int. 2002; 70(6):503-511.
48. Nuzzo S, Lafage-Proust MH, Martin-Badosa E, et al. Synchrotron radiation microtomography allows the analysis of three-dimensional microarchitecture and degree of mineralization of human iliac crest biopsy specimens: Effect of etidronate treatment. J Bone Miner Res. 2002;17(8):1372-1382.
49. Roschger P, Gupta HS, Berzlanovich A, et al. Constant mineralization density distribution in cancellous human bone. Bone. 2003;32(3):316-323.
50. Ruffoni D, Fratzl P, Roschger P, Phipps R, Klaushofer K, Weinkamer R. The effect of temporal changes in bone turnover on the bone mineralization density distribution: A computer simulation study. J Bone Miner Res. 2008;23(12): 1905-1914.

Keywords: bone mineralization density distribution; collagen; Fourier transform infrared imaging; nanocomposite; nanoindentation; quantitative backscattered electron imaging; Raman imaging; x-ray scattering