Nanoindentation discriminates the elastic properties of individual human bone lamellae under dry and physiological conditions

Abstract

Mechanical properties of single lamellae of human compact and trabecular bone tissue were measured with a combined atomic force microscopy (AFM) and nanoindentation technique. This combination allows for both characterization of bone surface topography and indentation of the bone extracellular matrix (ECM) with depths of between 100 and 600 nm. Four bone structural units (BSUs) were tested with 400 indents under dry conditions, and four BSUs with 160 indents were tested in a liquid cell under physiological conditions. A correspondence was established between the optical appearance of bone lamellae and the topography of the polished bone surface. The indentation modulus and hardness of bone ECM were investigated as a function of lamella type and indentation depth under wet and dry conditions. For low depth indents, thick lamellae showed a higher indentation modulus than thin lamellae. With increasing indentation depth, thick lamellae exhibited a significant decrease in indentation modulus and hardness, whereas, for thin lamellae, the effect of indentation depth was much less significant. These trends were similar for dry and physiological conditions and support compositional and/or ultrastructural differences between thick and thin lamellae.

Introduction

Besides reduction in bone mineral density, there is growing evidence that bone fragility might be linked to the degradation of the intrinsic mechanical properties of the bone extracellular matrix (ECM) that is associated with alteration of the bone remodeling process and fatigue damage accumulation.3, 15 It has also become widely accepted that the bone ECM determines the mechanical environment of the osteocytes and bone lining cells, and may therefore play an important role in mechanotransduction. These issues call for a better understanding of the intrinsic mechanical properties of the human bone ECM and their evolution with age and disease. The literature demonstrates a wide spectrum of studies that have focused on the characterization of intrinsic properties of bone tissue. Ascenzi and coworkers performed tension, compression, torsion, and bending tests of single osteons dissected from human bone and reported a strong dependence of Young’s modulus on average collagen fiber orientation and mineral content.1 Other investigators carried out microhardness tests2, 28 with imprint sizes of approximately 50 μm. Their results indicated a high correlation between microhardness and mineralization, anatomical site, Young’s modulus, yield strength, and tissue preparation. Surprisingly, Weaver did not observe a dependence of bone microhardness on donor, age, and osteoporosis.28 Based on microhardness tests, Hodgskinson et al. hypothesized that Young’s modulus for trabecular bone tissue and compact bone were comparable.10 Evans et al. and Ziv et al. found a strong dependence of bone tissue microhardness on collagen fiber and hydroxyapatite platelet orientation.6, 31 Unfortunately, hardness is a complex mechanical property that involves both elastic and postyield properties and cannot be easily converted to continuum-level properties such as Young’s modulus or shear strength. In addition, hardness can show a certain depth dependence, even for homogeneous materials.16

Another attractive technique for investigating the mechanical properties of individual bone structural units is ultrasound microscopy. The bone structural unit (BSU) represents the end result of a remodeling cycle; in cortical bone, it constitutes a haversian system (or cortical osteon) and, in cancellous bone, it is a wall or “packet” of bone (trabecular osteon).5 The experiments reported showed a relatively uniform acoustic reflectivity within each BSU, but significant differences between BSUs.12 However, acoustic reflectivity depends on both elastic properties and local material density, which makes the quantitative determination of Young’s modulus difficult.

Nanoindentation, which evolved from traditional Vickers microhardness testing, allows for measurement of mechanical properties at the nanometer scale. As a substantial improvement with respect to the aforementioned techniques, measured force displacement curves provide a local indentation modulus, a purely elastic property. The first nanoindentation studies applied on bone19, 20, 32 examined the influence of microstructure, drying, anatomical location, and age, and compared the mechanical properties of compact vs. trabecular bone. It was reported that drying increases Young’s modulus of compact bone by approximately 9%–16%, but does not change the relative stiffness of the bone constituents. For identical anatomical sites, interstitial bone showed the highest Young’s modulus, followed by osteonal bone, and then trabecular bone.

Further studies21, 23, 24 have investigated factors such as the anatomical orientation of the plane of indentation for vertebral and tibial bone and the site of indentation within secondary osteons. They measured significantly higher indentation moduli and hardness for compact and trabecular bone tissue tested in load-bearing directions with respect to transverse directions. Within single osteons, they observed a decrease of Young’s modulus with increasing distance to the haversian channel. Most of these studies reported indentations of 500–1000 nm depth, resulting in imprint sizes of 3–6 μm, and therefore test volumes exceeding the typical dimensions of single lamellae (typically 1–3 μm for thin lamellae and 2–4 μm for thick lamellae). In a recent study, thick lamellae were tested with a depth of 200 nm,23 but due to the difficulty in positioning the indenter tip in the submicron regime, no comparison between thick and thin lamellae was possible.

In an effort to extend our current knowledge of the mechanical properties of single bone lamellae, we sought to quantify the indentation modulus and hardness of both thin and thick lamellae selected from human trabecular and compact bone structural units (BSUs). The influence of lamella type (thick or thin) and indentation depth were examined under both dry and physiological conditions. For this purpose, we applied a combination of atomic force microscopy (AFM) and nanoindentation, which provided the required accuracy (better than 0.1 μm) to position the indenter tip in the center of a single lamella and perform reliable indentations with depths as low as 100 nm.