Multiscale and multidisciplinary analysis of aging processes in bone

Abstract

The world population is increasingly aging, deeply affecting our society by challenging our healthcare systems and presenting an economic burden, thus turning the spotlight on aging-related diseases: exempli gratia, osteoporosis, a silent disease until you suddenly break a bone. The increase in bone fracture risk with age is generally associated with a loss of bone mass and an alteration in the skeletal architecture. However, such changes cannot fully explain increased fragility with age. To successfully tackle age-related bone diseases, it is paramount to comprehensively understand the fundamental mechanisms responsible for tissue degeneration. Aging mechanisms persist at multiple length scales within the complex hierarchical bone structure, raising the need for a multiscale and multidisciplinary approach to resolve them. This paper aims to provide an overarching analysis of aging processes in bone and to review the most prominent outcomes of bone aging. A systematic description of different length scales, highlighting the corresponding techniques adopted at each scale and motivating the need for combining diverse techniques, is provided to get a comprehensive description of the multi-physics phenomena involved.

Fig. 1. Multiscale structure of bone.

a Human femur as an example of bone; b femur cross section showing the cortical tissue (outer layer) and the trabecular tissue that accomodate bone marrow (inner part); c magnification of the previous panel, with a focus on the Haversian structure: the main building block is the osteon, a hollow composite cylinder, made of several concentric fiber-reinforced layers, the lamellae, each one showing a preferential orientation of the mineralized collagen fibers; d collagen fiber, represented as a collagen fibril bundle (the surrounding matrix is removed for clarity); e mineralized collagen fibril, base constituent of the collagen fiber, which includes both intra- and extra-fibrillar hydroxyapatite (HA) mineral crystals; f mineral aggregate, including HA in both the forms of g) platelet and h) acicular crystals; i molecular component of the collagen fibers, aka tropocollagen, consisting of three polypeptide chains coiled around each other; j magnification of a polypeptide chain; k focus on the osteon structure, to reveal the osteocytes lying inside lacunae in the extracellular matrix of bones. Lacunocanalicular networks formed by dendrites propagating from osteocytes is clearly visible; l focus on bone cells responsible for the continous remodeling process. In particular, osteoclasts are responsible for bone resorption (on the left) and osteoblasts are involved in bone formation (right). The whole figure is adapted from Gabriele Grezzana et al. “Probing the Role of Bone Lamellar Patterns through Collagen Microarchitecture Mapping, Numerical Modeling, and 3D-Printing.” Advanced Engineering Materials 22.10 (2020): 2000387. Copyright 2020, with permission from Elsevier. Subfigure (k) and (l) are built with Servier Medical Art.

Fig. 2. Multiscale and multidisciplinary approaches used to study bone aging.

Overview of experimental and computational techniques suitable for studying various aspects of the aging process in bone tissue. Each “bubble” represents a specific approach and the “bubble” height spans the length scale of its applicability.

Fig. 3. Different techniques to study bone aging at the nanoscale.

a Map of crystal dimensions (respectively thickness and length) inside a healthy mature rat femoral cortex obtained with small- and wide-angle X-ray scattering (SAXS and WAXS) experiments from ref. ; b TEM micrographs of longitudinally-sectioned mineralized collagen fibrils in human normal trabecular bone. Distinct individual apatite crystals are seen with plate-like shape (arrows) and tablet-like shape (representing plates on edge) (dotted arrows). Reprinted from Bone 33 (2003) 270–282, Matthew A. Rubin et al., “TEM analysis of the nanostructure of normal and osteoporotic human trabecular bone”, Copyright 2003, with permission from Elsevier; c Schematic showing B-type carbonate substitution with atoms explicitly shown, for phosphate groups (only bonds shown) in the MD computational model, for different amounts (shown as wt%) of carbonate substitutions. Biomaterials 127 (2017) 75–88, Alix C. Deymier et al., “Protein-free formation of bone-like apatite: New insights into the key role of carbonation”, Copyright 2017, with permission from Elsevier; d Schematic of a computer-simulated collagen fibril. The characteristic gap and overlap region are clearly visible. Focus on the immature divalent and mature trivalent cross-links (top) and their representation in the molecular model (bottom). Reprinted from Bone 110 (2018) 107–114, Baptiste Depalle et al., “The different distribution of enzymatic collagen cross-links found in adult and children bone result in different mechanical behavior of collagen”, Copyright 2018, with permission from Elsevier; e Schematic representing the different locations of water inside HAP-collagen nanocomposites in bone. Water can be tightly bound to collagen triple helix, loosely bound at the interface between collagen and mineral, and present between the mineral platelets. Reprinted from Bone 120 (2019) 85–93, David B. Burr, “Changes in bone matrix properties with aging”, Copyright 2019, with permission from Elsevier.

Fig. 4. Different aspects of bone aging at the microscale.

a Schematic representation of aging changes in bone cells. (a.I) In osteoblasts, osteoclasts, and osteocytes, telomeres decrease in length with each cell division due to the inability of the cell to fully replicate this region, and upon reaching a critical level of telomere length, cells undergo cellular senescence or apoptosis. (a.II) Altered rate of bone remodeling during aging. In osteoblasts, there is an increased secretion of RANKL, triggering the process of osteoclast differentiation (+), and a decrease of OPG, which competes with RANKL for binding to the RANK receptor inhibiting osteoclast formation and activity (−). The expression levels of all Wnt proteins, particularly Wnt10b, are significantly decreased in osteoblasts and osteocytes, thus reducing the bone anabolism (−). With aging, reactive oxygen species (ROS) increase and indirectly inhibit β-catenin, thus decreasing osteoblast formation (−). Subfigure (a) is built with Servier Medical Art; b Collagen fibrils coming from demineralized tibiae of a 10 months old female rat (control, I) and of an ovariectomized (used as an animal model for aging, II) specimen of same age viewed with electron microscopy. Fibrils in the ovariectomized case have an irregular arrangement in contrast to control. Scale bar is 0.5 μm. Reprinted from Kafantari et al. “Structural alterations in rat skin and bone collagen fibrils induced by ovariectomy.” Bone 26.4 (2000): 349–353, Copyright 2000, with permission from Elsevier; c Circularly polarized light microscopy (CPL) experiments allow the visualization of osteons with different collagen fiber orientation: osteons composed of oblique orientated collagen fibers appear bright (II and zoomed in IV), while osteons composed of collagen fibers parallel with the osteon axis appear dark (I and zoomed in III). It has to been noticed that the two types of osteons coexist in bone samples of the same subject. Reprinted from ref. ; d Changes in lacunocanalicular network and single osteocytes in young and aged individuals resulting from scanning electron micrograph of resin-embedded acid-etched osteonal bone performed by Milovanovic et al. The number of osteocytes lacunae in the young sample (I) is clearly higher than in the aged one (III), as well as the number of dendrites departing from a single osteocyte (II and IV). Reprinted with permission from Petar Milovanovic et al. “Osteocytic canalicular networks: morphological implications for altered mechanosensitivity.” ACS nano 7.9 (2013): 7542–7551. Copyright 2013 American Chemical Society; e Comparison of cortical bone porosity between young, aged, osteoporotic (Opo), and immobilized (Immo) individuals in quantitative Backscattered Electron Images (qBEI) of human bone samples obtained using scanning electron microscopy (SEM) by ref. .

Fig. 5. Studying effects of aging on bone tissue at the macroscale.

a Dual X-ray absorptiometry (DXA) images of human spine allowing the comparison between Bone Mineral Density (BMD) and Trabecular Bone Score (TBS) scoring techniques. Two different patients with similar BMD could show different TBSs. Thanks to the experimental variograms, TBS takes into account the degradation of bone microarchitecture. Figure reprinted from ref. ; b Two-dimensional computed X-ray tomographic reconstructions showing crack propagation in bone longitudinal direction. Each image represents a transverse section taken at different distances from the nominal crack tip (see numbers on top of each figure). The comparison is between young (top row) and old (bottom row) human bone samples. Differences in micro-crack dimensions and uncracked-ligament bridges (indicated by black arrows) are clearly visible. Reprinted from R.K. Nalla et al. “Effect of aging on the toughness of human cortical bone:evaluation by R-curves” Bone 35 (2004) 1240–1246, Copyright 2004, with permission from Elsevier; c 2D cross-sectional images of mouse femurs obtained with μCT. Top rows show the femoral cortical bone in both male and female mice of different ages (6, 12, 18, and 22 months respectively). Bottom rows, instead, show femoral trabecular bone in male and female mice of different ages (6, 12, 18, and 22 months respectively). Scale bars are 1 mm. Figure adapted from ref. ; d Mechanical properties of human cortical bone as a function of age obtained with three-point bending mechanical tests. (I) Stress–strain curves and (II) fracture–toughness R-curves show a decline of both strength and crack toughness with advancing age. Figure reproduced from ref. .