Elsevier

Journal of Biomechanics

Volume 59, 5 July 2017, Pages 35-42
Journal of Biomechanics

Microfluidic co-culture platform for investigating osteocyte-osteoclast signalling during fluid shear stress mechanostimulation

https://doi.org/10.1016/j.jbiomech.2017.05.012Get rights and content

Abstract

Bone cells exist in a complex environment where they are constantly exposed to numerous dynamic biochemical and mechanical stimuli. These stimuli regulate bone cells that are involved in various bone disorders, such as osteoporosis. Knowledge of how these stimuli affect bone cells have been utilised to develop various treatments, such as pharmaceuticals, hormone therapy, and exercise. To investigate the role that bone loading has on these disorders in vitro, bone cell mechanotransduction studies are typically performed using parallel plate flow chambers (PPFC). However, these chambers do not allow for dynamic cellular interactions among different cell populations to be investigated. We present a microfluidic approach that exposes different cell populations, which are located at physiologically relevant distances within adjacent channels, to different levels of fluid shear stress, and promotes cell-cell communication between the different channels. We employed this microfluidic system to assess mechanically regulated osteocyte-osteoclast communication. Osteoclast precursors (RAW264.7 cells) responded to cytokine gradients (e.g., RANKL, OPG, PGE-2) developed by both mechanically stimulated (fOCY) and unstimulated (nOCY) osteocyte-like MLO-Y4 cells simultaneously. Specifically, we observed increased osteoclast precursor cell densities and osteoclast differentiation towards nOCY. We also used this system to show an increased mechanoresponse of osteocytes when in co-culture with osteoclasts. We envision broad applicability of the presented approach for microfluidic perfusion co-culture of multiple cell types in the presence of fluid flow stimulation, and as a tool to investigate osteocyte mechanotransduction, as well as bone metastasis extravasation. This system could also be applied to any multi-cell population cross-talk studies that are typically performed using PPFCs (e.g. endothelial cells, smooth muscle cells, and fibroblasts).

Introduction

Bone consists of a multitude of cells, including osteoclasts (bone resorption cells), osteoblasts (bone formation cells), and osteocytes (bone cells embedded in the bone matrix) (Clarke, 2008). These cells are subjected to different mechanical stimuli and biochemical signals that modulate their response. These cellular responses can regulate a variety of bone disorders, including increasing osteoclast activity associated with osteoporosis and bone metastasis. By understanding these cellular responses, it is possible to develop novel strategies for overcoming these bone disorders and pathologies (Rochefort, 2014, Zheng et al., 2013). However, recapitulating this complex environment in vitro is challenging.

One common clinical intervention for these disorders is stimulating bone through exercise (Beaton et al., 2009, Nikander et al., 2010). Osteocytes are located within interstitial spaces known as lacuna, and their processes connect through narrow channels known as canaliculi, making up the lacunar-canalicular system (LCS) (Schneider et al., 2010, You et al., 2004). When a load is applied to bone, the LCS is compressed, inducing a variety of mechanical stimuli, including fluid shear stresses (FSS) (Price et al., 2011). This results in changes in the expression of osteocyte generated signals, such as receptor activator of nuclear-κ β ligand (RANKL) (Xiong and O'Brien, 2012), osteoprotegerin (OPG) (You et al., 2008a), nitric oxide (NO) (Vatsa et al., 2007), prostaglandin E2 (PGE-2) (Kitase et al., 2010), and sclerostin (Nguyen et al., 2013). Osteocytes communicate to other cells through paracrine signaling (Schaffler and Kennedy, 2012) and gap junctions (Jiang et al., 2007). Because of their thorough distribution within bone, and their sensitivity to various mechanical stimuli, osteocytes are critical in regulating the response of bone to mechanical stimulation (Bonewald and Johnson, 2008), and likely regulate many bone pathologies (Bonewald, 2004, Zhou et al., 2016).

Most in vitro osteocyte mechanotransduction and cell regulation studies are performed using parallel plate flow chambers (PPFC), where osteocyte-like cells are subjected to FSS, and conditioned medium is applied to different cells (Cheung et al., 2011, You et al., 2008a). However, this methodology lacks dynamic and real-time biochemical signaling between the cells involved, the cell signaling is unidirectional, and it ignores interactions of low half-life signals, such as NO (Clancy et al., 1990). Additionally, these macro-scaled chambers prevent the use of primary osteocytes (Kato et al., 1997) due to the large cell numbers required. Various microfluidic platforms have been developed to study the mechanical stimulation of bone cells. These include a multi-shear platform to investigate osteoblast calcium dynamics (Kou et al., 2011), a magnetic stimulator to promote osteoblast proliferation (Song et al., 2010), as well as a model of the LCS (You et al., 2008b). However, none of these platforms address the limitations of cellular cross-talk that currently exists with PPFCs.

To mitigate these constraints, different co-culture systems have been developed to investigate a variety of intercellular signaling pathways (Miki et al., 2012); the most common being transwell inserts. However, transwells have limited capabilities to apply FSS to cells. Specifically, rotating disks have been used to apply flow to transwells, resulting in radially dependent FSS (Taylor et al., 2007). As well, flow can only be applied to the population of cells seeded on the top of the transwell. Various microfluidic systems have also been developed to perform flow based co-culture. These devices have one cell type embedded in a gel (Chen et al., 2013, Sellgren et al., 2015), use model extracellular matrix (ECM) (Jeon et al., 2013), or use porous membranes to separate different cell populations (Booth and Kim, 2012). However, these techniques reduce signal transport between the cell populations separated by the gel (Amsden, 1998), and can be complex to fabricate for biologically focused labs.

In this work, we present a gel-free microfluidic co-culture system. High resistance side channels allow this system to apply isolated stimulatory flow (>0.5 Pa) (Li et al., 2012) while promoting signaling between different cell populations. Our device is validated through analytical modeling and experimental measurements. Finally, this device was used to investigate osteoclast precursor responses to signals produced by mechanically stimulated (fOCY) or unstimulated (nOCY) osteocytes, as well as the regulation of osteocyte mechanosensitivity by osteoclasts.

Section snippets

Device development

The device models the bone microenvironment in terms of developing cell signaling gradients and fluid flow stimulation (Fig. 1A). The device consists of either 2 or 3 parallel cell culture channels (Length: 1.6 cm, Width: 1 mm, Height: 60 μm) depending on the experiment being performed. Each channel is separated by high resistance side channels (Length: 200 μm, Width: 20 μm, Height: 60 μm) at a physiologically relevant distance (Pfeiffer, 1998) to minimize convection between channels. The device

Fluid shear stresses in co-culture system

In the full experimental setup, a 2 Pa flow is applied to the device. Experimentally, a FSS of 1.65 Pa, 0.28 Pa, and 0.07 Pa was measured in the fOCY, OCL, and nOCY channels respectively (Fig. 2). This result is similar to our analytical prediction of a 1.54 Pa FSS in the fOCY channel. We took the threshold mechanical stimulation to be 0.5 Pa, as this was the shear stress level where Cox-2 and RANKL/OPG mRNA of MLO-Y4 cells was not significantly altered relative to static (Li et al., 2012). This

Discussion

Current PPFCs are limited when investigating mechanically regulated cell interactions. Specifically, these systems lack direct and dynamic cell signaling, allow only unidirectional signaling, and omit low half-life signal interactions. Furthermore, these macro-scaled platforms are not applicable for primary osteocyte studies due to the large number of cells required. There have been recent developments in flow based co-culture systems that use gels to isolate fluid stimulation, but these

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

We would like to thank Junyi (Danny) Cen for performing the ELISA experiments presented in the supplement. We also acknowledge postgraduate scholarships provided by the Natural Sciences and Engineering Research Council (NSERC), the NSERC CREATE Microfluidic Applications and Training in Cardiovascular Health program, the Toronto Musculoskeletal Centre, and Barbara and Frank Milligan. Funding for this research was provided by the Canadian Institutes of Health Research (fund #282723). Microfluidic

References (54)

  • Y. Takeshita et al.

    An in vitro blood–brain barrier model combining shear stress and endothelial cell/astrocyte co-culture

    J. Neurosci. Methods

    (2014)
  • R.F. van Oers et al.

    A unified theory for osteonal and hemi-osteonal remodeling

    Bone

    (2008)
  • A. Vatsa et al.

    Extracellular NO signalling from a mechanically stimulated osteocyte

    J. Biomech.

    (2007)
  • L. You et al.

    Osteocytes as mechanosensors in the inhibition of bone resorption due to mechanical loading

    Bone

    (2008)
  • J.-N. Zhang et al.

    The role of the sphingosine-1-phosphate signaling pathway in osteocyte mechanotransduction

    Bone

    (2015)
  • S. Zhang et al.

    The affinity of human RANK binding to its ligand RANKL

    Arch. Biochem. Biophys.

    (2009)
  • Y. Zheng et al.

    The role of the bone microenvironment in skeletal metastasis

    J. Bone Oncol.

    (2013)
  • S.A. Al-Dujaili et al.

    Apoptotic osteocytes regulate osteoclast precursor recruitment and differentiation in vitro

    J. Cell. Biochem.

    (2011)
  • B. Amsden

    Solute diffusion within hydrogels. Mechanisms and models

    Macromolecules

    (1998)
  • G.J. Atkins et al.

    Sclerostin is a locally acting regulator of late-osteoblast/preosteocyte differentiation and regulates mineralization through a MEPE-ASARM-dependent mechanism

    J. Bone Miner. Res.

    (2011)
  • R. Beaton et al.

    Effects of exercise intervention on persons with metastatic cancer: a systematic review

    Physiother. Canada

    (2009)
  • D.M. Biskobing et al.

    Characterization of MCSF-induced proliferation and subsequent osteoclast formation in murine marrow culture

    J. Bone Miner. Res.

    (1995)
  • L.F. Bonewald

    Osteocyte biology: its implications for osteoporosis

    J. Musculoskelet. Neuronal Interact.

    (2004)
  • R. Booth et al.

    Characterization of a microfluidic in vitro model of the blood-brain barrier (μBBB)

    Lab Chip

    (2012)
  • M.B. Chen et al.

    A 3D microfluidic platform incorporating methacrylated gelatin hydrogels to study physiological cardiovascular cell-cell interactions

    Lab Chip

    (2013)
  • W.-Y. Cheung et al.

    Osteocyte apoptosis is mechanically regulated and induces angiogenesis in vitro

    J. Orthop. Res.

    (2011)
  • B. Clarke

    Normal bone anatomy and physiology

    Clin. J. Am. Soc. Nephrol.

    (2008)
  • Cited by (56)

    • Bone-on-a-chip platforms and integrated biosensors: Towards advanced in vitro bone models with real-time biosensing

      2023, Biosensors and Bioelectronics
      Citation Excerpt :

      RAW264.7 derived osteoclasts and osteocytes were located at physiologically relevant distances within adjacent channels, and osteocytes were mechanically stimulated (Fig. 1a). The results showed a decreased osteoclasts density and differentiation toward the stimulated channels due to decreased RANKL expression and increased OPG expression in stimulated MLO-Y4 cells (Middleton et al., 2017). Moreover, shear stresses have been reported to affect the metabolic activities of bone cells and cell-cell interactions in bone development.

    • Developments of microfluidics for orthopedic applications: A review

      2023, Smart Materials in Medicine
      Citation Excerpt :

      Further, a device capable of simultaneously culturing two kinds of cells confirmed that mesenchymal stem cells are more sensitive than adipose-derived stem cells under the same mechanical stimulation, exhibiting higher osteogenic differentiation, which provides valuable guidance on how to select seed stem cells. A typical parallel plate flow chamber structure is designed to provide multiple identical flow paths, enabling different shear forces to be cultured for different cells while also observing the interaction between cells [69]. This makes the single cell culture method more diverse and closer to the complex environment in which multiple tissues interact in the body [70].

    View all citing articles on Scopus
    View full text