Introduction and Objective

Bone remodelling is a continuous process whereby osteocytes regulate the activity of osteoblasts and osteoclasts to repair loading-induced microdamage. While many in vitro studies have established the role of paracrine factors (e.g., RANKL/OPG) and cellular pathways involved in bone homeostasis, these techniques are generally limited to two-dimensional cell culture, which neglects the role of the native extracellular matrix in maintaining the phenotype of osteocyte. Recently, ex vivo models have been used to understand cell physiology and mechanobiology in the presence of the native matrix. Such approaches could be applicable to study the mechanisms of bone repair, whilst also enabling exploration of biomechanical cues. However, to date an ex vivo model of bone remodelling in cortical bone has not been developed. In this study, the objective was to develop an ex vivo model where cortical bone was subjected to cyclic strains to study the remodelling of bone.

While the phenomena of bone adaption to mechanical loading has been long observed, the mechanisms governing bone mechanotransduction during health and disease are not well understood. Our multidisciplinary experimental and computational research strives to enhance understanding of bone mechanobiology, and in particular how this process is affected at the onset of osteoporosis. We have provided an enhanced understanding of bone cell mechanosensation. We have characterised the local mechanical environment of MSCs, osteoblasts and osteocytes in vivo. Most importantly, we have discovered that the matrix composition, expression of mechanosensors and the mechanical environment of osteocytes is altered during osteoporosis. Interestingly, a mechanobiological response restores the homeostatic mechanical environment of the cells in the longer term. Our recent in vitro studies have revealed that estrogen withdrawal from bone cells alters calcium signalling, mineralisation, biochemical responses and osteogenic gene expression when these cells are exposed to an applied fluid shear stress. Our ongoing research is investigating mechanobiology-based therapeutic approaches for treatment of bone pathologies, by (1) targeting mechanoregulatory signalling pathways and (2) developing in vitro tissue regeneration strategies that seek to optimise the mechanical environment (through matrix stiffness, bioreactors) to stimulate osteogenesis.

3D cell culture studies more accurately represent the complex in vivo mechanical environment of human bone and are, thus, superior to 2D studies when testing the efficacy of osteoporosis therapies. As such, the objective of this study was to use a 3D model to investigate the effect of sclerostin antibodies. Sclerostin is a protein, which inhibits osteoblasts and is downregulated under mechanical stimulation. It is not yet known how expression of sclerostin mediates the site-specific and temporal changes in mineralisation. To address this, we developed a 3D cellular niche of MC3T3 osteoblasts encapsulated within gelatin and applied mechanical loading to the constructs using a custom-designed compression bioreactor system (0.5% strain at 0.5 Hz, 1 hr/day) (VizStim) under continuous perfusion of cell culture media. Osteoblasts were pretreated with estrogen for 14 days, followed by estrogen withdrawal (EW) to simulate postmenopausal conditions. 3D constructs were successfully fabricated and actin staining revealed the formation of dendritic cells under both static and stimulated conditions indicative of osteocyte-like cells. Under static conditions, estrogen treatment enhanced production of calcium by osteoblasts when compared to the same cells cultured under estrogen deficient conditions. Overall, preliminary results propose a link between mechanical stimulation, estrogen deficiency and mineral production by osteoblasts. Ongoing studies are comparing the static and stimulated groups after a longer culture period of 21 days using sclerostin antibodies. This research aims to deliver further understanding of the mechanical regulation of bone formation, and will inform novel approaches for regeneration of bone tissue and treatment of osteoporosis.

Although osteoporosis reduces overall bone mass causing bone fragility, our recent studies have shown that bone tissue composition is altered at the microscopic level, which is undetectable by conventional diagnostic techniques (DEXA) but may contribute to bone fracture. However, the time sequence of changes in bone microarchitecture, mechanical environment and mineral distribution are not yet fully understood. This study quantified the longitudinal effects of estrogen deficiency on the trabecular microarchitecture and mineral distribution in the tibia of Female Wistar rats (6 months) that underwent ovariectomy (OVX, n=10) or sham surgery (SHAM, n=10). Weekly micro-CT scans of the proximal tibia were conducted at 15µm resolution for the first month of estrogen deficiency. Morphometric analysis was conducted to characterise the trabecular bone microarchitecture. The bone mineral composition was characterised with analysis of bone mineral density distributions (BMDD). There was significantly reduced trabecular bone volume fraction at 2 weeks in OVX rats compared to controls (p<0.01). There was no difference in mineral distribution between the OVX and control animals. This study provides the first evidence in uncovering the temporal nature of changes in bone microarchitecture and mineral distribution, showing that structure changes before composition. In-vivo µCT analysis for later time points (week 8, 14 and 34) is ongoing to comprehensively examine these longitudinal compositional changes. Moreover, we are conducting ex-vivo mechanical analysis (nanoindentation), and together these will uncover the time-sequence and respective contribution of changes in bone mass and composition to the integrity of the bone tissue at these stages of estrogen deficiency.

Recent studies have shown that bone mineral distribution is more heterogeneous in bone tissue from an animal model of osteoporosis and osteoporotic human vertebral trabeculae. These tissue alterations may play a role in bone fragility seen in osteoporosis, albeit that they are not detectable by current diagnostic techniques (dual-energy X-ray absorptiometry, DXA). Type II Diabetes Mellitus (T2DM) also increases a patient's fracture risk beyond what can be explained or diagnosed by DXA, and is associated with impaired bone cell function, compromised collagen structure and reduced mechanical properties. However, it is not currently known whether osteoporosis or T2DM leads to an increased mineral heterogeneity in the femoral head of humans, a common osteoporotic fracture site. In this study, we examine bone microarchitecture, mineralisation and mechanical properties of trabecular bone from osteoarthritic, diabetic and osteoporotic patients. We report that while osteoporotic trabecular bone has significantly deteriorated mechanical properties and microarchitecture compared to the other groups, there is also a significant increase in mean mineral content. Moreover, the heterogeneity of the mineral content in osteoporotic bone is significantly higher than osteoarthritic (+35%) and diabetic (+13%) groups. We propose that the compromised architecture following bone loss at the onset of osteoporosis alters the mechanical environment, which initiates compensatory changes in mineral content. We show for the first time that trabecular bone mineralisation is significantly more heterogeneous (+20%) in T2DM compared to osteoarthritic controls. Interestingly, bone microarchitecture and mechanical properties are not significantly different between diabetic and osteoarthritic groups despite this increase in mineral heterogeneity.

The expression of the mechanosensor, integrin αvβ3, is reduced in osteoporotic bone cells compared to controls. MLO-Y4 osteocytes experience altered mechanotransduction under estrogen deficiency and it is unknown whether this is associated with defective αvβ3 expression or signalling. The objectives of this study are to (1) investigate αvβ3 expression and spatial organisation in osteocytes during estrogen deficiency, and (2) establish whether altered responses of osteocytes under estrogen deficiency correlate to defective αvβ3 expression and functionality. MLO-Y4 cells were cultured as follows: Ctrl (no added estradiol), E+ (10nM 17β-estradiol for 5 days), and Ew (10nM 17β-estradiol for 3 days and withdrawal for 2 days). Cells were cultured with/without 0.5µM IntegriSense750 (αvβ3 antagonist). Laminar oscillatory fluid flow of 1Pa at 0.5Hz was applied for 1hr. αvβ3 content was quantified using an ELISA. The location and quantity of αvβ3 and focal-adhesions was determined by immunocytochemistry. Estrogen withdrawal under static conditions led to lower cell and focal-adhesion area (p<0.05), compared to E+ cells. Fluid flow led to higher αvβ3 content (p<0.05) in all groups, compared to static counterparts, with αvβ3 blocking altering this response. Fluid flow on Ew cells had the highest αvβ3 levels (p<0.05), but αvβ3 did not localise at focal-adhesions sites. Cell morphologies were similar after treatment with the αvβ3 antagonist to the Ew group. These results suggest there are fewer functional focal-adhesion sites at which αvβ3 integrins localise to facilitate mechanotransduction. To further understand these results, we are analysing osteocyte mechanotransduction by quantifying PGE2 and gene expression (COX-2, RANKL, OPG, SOST).

Osteoporosis has long been associated with weak bones but recent studies have shown that bone tissue mineral becomes more heterogeneous and the expression of mechanosensors are altered during estrogen deficiency in an animal model of osteoporosis. However, whether these changes occur as a primary response to estrogen deficiency is unknown. In this study we investigate whether matrix production and mineralisation by mechanically-stimulated osteoblasts are impaired as a direct consequence of estrogen depletion. Osteoblast-like MC3T3-E1 cells were cultured for 14 days with 10⁻⁸M of 17β-estradiol and subsequently cultured with osteogenic media only, or supplemented with estrogen or an estrogen antagonist (Fulvestrant, 10⁻⁷M). Physiological shear stress (1Pa) was applied using an orbital shaker (290rpm, 40min/day), which allows long-term culture and induces oscillatory flow on cells. Osteoblasts phenotype, extracellular matrix (ECM), mineralisation and mechanosensors were tracked by qRT-PCR (Runx2, Col1a1, Col1a2, Cox2, Bglap2, FN1), by biochemical assays (ALP activity, DNA and calcium content), by immunostaining (integrin α_v, BSP2, fibronectin) and by labelling with calcein the calcium. The results of this study demonstrate that after 7 days, estrogen depleted cells had less integrin α_v mechanosensors compared to those that received continuous estrogen treatment. By 14 days the ECM formation (calcium, fibronectin) by osteoblasts was altered under estrogen depletion, when compared to cells that were cultured continuously with estrogen. This study provides evidence of changes in osteoblast behaviour under estrogen depletion, which might explain the alteration in tissue mineral content and the decrease of integrins observed previously in ovariectomized rats in vivo.