A micro-multiphysics in silico model for single-cell mechanomics of bone in the aging mouse

Abstract

Osteoporosis is one of the most prevalent degenerative diseases, and it is characterized by a reduction in bone mass and an increase in the risk of bone fractures. It emerges with aging in the human population, and the current treatments still need to be improved. The development of computational tools that can analyze the mechanobiology of bone remodeling at the tissue, cellular, and protein levels in the trabecular microarchitecture is suitable for quantifying and identifying potential targets for novel therapies. For this purpose, a first version of an in silico model for simulating homeostatic trabecular bone remodeling needs to be established as the basis for further developments. To this end, the comparison of the bone morphometry of the in silico data against the corresponding in vivo data is an established approach for determining which bone microarchitectural properties are better captured. For the first time, cellular and protein information can be established in the model and used for testing different pathophysiological conditions. In this way, it is possible to test hypotheses about the rules of the model and to explore the parameter space. In this thesis, we aimed to investigate these aspects via three aims: (i) to establish a micro-multiphysics agent-based (micro-MPA) in silico model of trabecular bone remodeling for simulating homeostatic bone remodeling using single-cell mechanomics, i.e., the response to the mechanical signal in terms of cytokine production and bone remodeling; ii) to create a clustering technique for analyzing single-cell mechanomics data from micro-MPA simulations of bone remodeling; iii) to determine the tissue mechanoregulation and cellular properties that are altered in prematurely aged. To achieve these aims, micro-computed tomography (micro-CT) images and micro-FE were employed to determine the mechanical environment to be used in the in silico model and in the mechanoregulation analysis. The mechanoregulation analysis was applied to determine how the mechanical environment is translated into bone formation, resorption, and quiescence in young and prematurely aged mice under mechanical control and loading conditions. In the first part of the thesis, a computational model for simulating bone remodeling was developed. The micro-MPA model was developed and specifically calibrated to simulate bone remodeling in the trabecular region of the sixth caudal vertebra (CV6) of PolgA(D257A/D257A) mice, a mouse model of premature aging. For this purpose, the model employed the time-lapsed micro-CT images acquired in vivo from a group of mice with a sham-loading regime and used them as input to the simulations. Micro-FE models were created using the micro-CT images to get insight into the mechanical environment. The single-cell mechanomics was defined for osteoblasts, osteoclasts, and osteocytes. The trabecular region was simulated to undergo homeostatic bone remodeling, and the in silico data were analyzed and compared against the in vivo data using static and dynamic bone morphometry. For the first time, a micro-MPA model was used to simulate trabecular bone remodeling using in vivo data and employing single-cell mechanomics. Additionally, a sensitivity analysis was performed by simulating bone remodeling with changes in the single-cell mechanomics of the osteocytes. Specifically, the production values of osteoprotegerin (OPG), sclerostin (Scl), and receptor activator of nuclear factor kappaB ligand (RANKL) varied from the values used in the simulations of homeostatic remodeling. The static and dynamic bone morphometry of these simulations was then computed and compared. This approach showed the capability of such a micro-MPA model to investigate the effects of the modeled signaling pathways in regulating bone remodeling. The simulations of homeostatic bone remodeling were further analyzed by selecting a representative sample of simulated bone remodeling and performing clustering of osteocytes from a cross-section of the trabecular region. The clustering was performed by using the protein concentration values where the osteocytes resided. Additionally, it was possible to define the closest remodeling events to the osteocytes and their local mechanical signal from the micro-FE analysis of the micro-MPA simulation. This study confirmed the relevant role of the osteocytes in orchestrating the response to the mechanical signal through their signaling pathways. To address the third aim of the thesis, the previously time-lapsed micro-CT data of four groups of PolgA mice were analyzed with respect to their mechanoregulatory capacities. Mechanoregulation, or how mechanics drive bone formation, resorption, and quiescence, was studied in young and prematurely aged mice under mechanical control or a loading regime. Additionally, these groups were simulated using the single-cell mechanomics micro-MPA model. In the simulations, the parameters of the cellular characteristics were altered to reflect the physiological conditions of the mice. From this study, quiescence resulted in being a more frequent event in aged mice, and the osteocytes’, osteoblasts’, and osteoclasts’ activities were altered with a reduction in the anabolic and anti-catabolic responses and an increase in the catabolic and anti-anabolic responses. In summary, the trabecular environment of the CV6 was extensively investigated with in silico tools. The development of the micro-MPA model allowed the simulation of signaling pathways and bone cells and the analysis of single-cell mechanomics. This model showed that it is possible to model how different bone cells can interact with each other, how they move on the bone surface, and how they remodel bone. Single-cell mechanomics allowed the inspection of the influence of the signaling pathways on bone remodeling across various tissue scales. This approach is a first step towards a more integrated analysis of the cells in their mechanical and biological environment. Lastly, the computational analysis of disrupted bone remodeling in prematurely aged mice showed how it is possible to give insights into the cellular mechanisms and to give perspectives on potential future directions of investigation for novel experiments and treatments.