A new method for application of force to cells via ferric oxide beads

 We describe a new method that uses straightforward physics to apply force to substrate-attached cells. In this method, collagen-coated magnetic ferric oxide beads attach to the dorsal surface of cells via receptors of the integrin family, and a magnetic field gradient is applied to produce a force. In this paper we present a complete characterization of the method in a configuration that is easy to use, in which a permanent magnet provides a fairly uniform gradient over a relatively large area. This allows a fairly uniform average force that can be controlled in magnitude, direction, and duration to be applied to a large number of cells. We show how to determine the applied force per cell by measuring the force per unit volume of magnetic bead, the distribution of bead diameters, and the distribution of beads per cell. We also show how to calculate the force per unit volume of bead in a three-dimensional region near the permanent magnet on the basis of field measurements, and present results for three of the magnets. An upward force applied to fibroblasts by this method produces a measurable time-dependent increase in attachment of cytoskeletal actin filaments to the force application points, and an increase in actin cross-linking. This is accompanied by an actin-dependent retraction of the force-induced upward movement of the dorsal surface of the cells. Received: 27 February 1997 / Received after revision: 10 August 1997 / Accepted: 1 September 1997     

Inhibitory effect of amiloride and gadolinium on fine afferent nerves in the rat knee: evidence of mechanogated ion channels in joints

Synovial joints are complex sensory organs which provide continuous feedback regarding position sense and degree of limb movement. The transduction mechanisms which convert mechanical forces acting on the joint into an electrochemical signal which can then be transmitted to the central nervous system are not well understood. The present investigation examined the effect of the mechanogated ion channel blockers amiloride and gadolinium on knee joint mechanosensitivity. In deeply anaesthetised rats (sodium thiopental: 100–120 mg/kg, i.p.), single unit extracellular recordings were made from knee joint group III (Aδ) and group IV (C) primary afferents in response to mechanical rotation of the joint. Afferent firing rate was measured before and after topical application of either amiloride (0.1 mM, 1 mM) or gadolinium (250 μM) onto the receptive field of the sensory unit and recording was continued every 10 min up to a total of 50 min. With normal rotation of the knee, joint mechanosensitivity was significantly reduced by both amiloride (P<0.0001; n=10–21) and gadolinium (P=0.001; n=12) and this effect was sustained throughout the recording period. This investigation provides the first in vivo electrophysiological evidence that joint mechanotransduction involves the activation of amiloride and gadolinium-sensitive mechanogated ion channels. Future studies to determine the mechanogated ion channel subtypes present in joints and the modulation of their gating properties during inflammation may yield novel approaches for the control of arthritis pain. Funding: JJMcD is funded by the Alberta Heritage Foundation for Medical Research, the Canadian Institutes for Health Research, and the Arthritis Society of Canada.     

Mechanotransduction of bone cells<Emphasis Type="Italic">in vitro</Emphasis>: Mechanobiology of bone tissue

Mechanical force plays an important role in the regulation of bone remodelling in intact bone and bone repair. In vitro, bone cells demonstrate a high responsiveness to mechanical stimuli. Much debate exists regarding the critical components in the load profile and whether different components, such as fluid shear, tension or compression, can influence cells in differing ways. During dynamic loading of intact bone, fluid is pressed through the osteocyte canaliculi, and it has been demonstrated that fluid shear stress stimulates osteocytes to produce signalling molecules. It is less clear how mechanical loads act on mature osteoblasts present on the surface of cancellous or trabecular bone. Although tissue strain and fluid shear stress both cause cell deformation, these stimuli could excite different signalling pathways. This is confirmed by our experimental findings, in human bone cells, that strain applied through the substrate and fluid flow stimulate the release of signalling molecules to varying extents. Nitric oxide and prostaglandin E₂ values increased by between two- and nine-fold after treatment with pulsating fluid flow (0.6±0.3 Pa). Cyclic strain (1000 μstrain) stimulated the release of nitric oxide two-fold, but had no effect on prostaglandin E₂. Furthermore, substrate strains enhanced the bone matrix protein collagen I two-fold, whereas fluid shear caused a 50% reduction in collagen I. The relevance of these variations is discussed in relation to bone growth and remodelling. In applications such as tissue engineering, both stimuli offer possibilities for enhancing bone cell growth in vitro.     

流体剪切力在骨生长、重建中的重要作用

骨组织具有最优化结构,是一个典型的结构-功能受生物力学控制的例子。力学因素在骨的生长、重建和成形中起着十分重要的作用,但力学刺激的本质形式至今无法确认。骨组织存在多孔结构,力学负荷引起的变形会促使流体流动对骨细胞产生作用,离体实验也证实骨细胞能对流体产生响应,因此流体剪切力是研究骨组织受力学调控时考虑的一个重要方面。现有的研究主要体现在流体的有效作用形式,细胞的生物学反应,以及流体剪切力在细胞中的力学转导;各个方面的研究都提示,流体剪切力至少部分的参与了骨组织内的力转导。作者就这方面的研究进展作一综述。     

Modeling Deformation-Induced Fluid Flow in Cortical Bone’s Canalicular–Lacunar System

To explore the potential role that load-induced fluid flow plays as a mechano–transduction mechanism in bone adaptation, a lacunar–canalicular scale bone poroelasticity model is developed and implemented. The model uses micromechanics to homogenize the pericanalicular bone matrix, a system of straight circular cylinders in the bone matrix through which bone fluids can flow, as a locally anisotropic poroelastic medium. In this work, a simplified two-dimensional model of a periodic array of lacunae and their surrounding systems of canaliculi is used to quantify local fluid flow characteristics in the vicinity of a single lacuna. When the cortical bone model is loaded, microscale stress, and strain concentrations occur in the vicinity of individual lacunae and give rise to microscale spatial variations in the pore fluid pressure field. Furthermore, loading of the bone matrix containing canaliculi generates fluid pressures in the contained fluids. Consequently, loading of cortical bone induces fluid flow in the canaliculi and exchange of fluid between canaliculi and lacunae. For realistic bone morphology parameters, and a range of loading frequencies, fluid pressures and fluid–solid drag forces in the canalicular bone are computed and the associated energy dissipation in the models compared to that measured in physical in vitro experiments on human cortical bone. The proposed model indicates that deformation-induced fluid pressures in the lacunar–canalicular system have relaxation times on the order of milliseconds as opposed to the much shorter times (hundredths of milliseconds) associated with deformation-induced pressures in the Haversian system.     

Finite-Element Stress Analysis of a Multicomponent Model of Sheared and Focally-Adhered Endothelial Cells

Hemodynamic forces applied at the apical surface of vascular endothelial cells may be redistributed to and amplified at remote intracellular organelles and protein complexes where they are transduced to biochemical signals. In this study we sought to quantify the effects of cellular material inhomogeneities and discrete attachment points on intracellular stresses resulting from physiological fluid flow. Steady-state shear- and magnetic bead-induced stress, strain, and displacement distributions were determined from finite-element stress analysis of a cell-specific, multicomponent elastic continuum model developed from multimodal fluorescence images of confluent endothelial cell (EC) monolayers and their nuclei. Focal adhesion locations and areas were determined from quantitative total internal reflection fluorescence microscopy and verified using green fluorescence protein–focal adhesion kinase (GFP–FAK). The model predicts that shear stress induces small heterogeneous deformations of the endothelial cell cytoplasm on the order of <100 nm. However, strain and stress were amplified 10–100-fold over apical values in and around the high-modulus nucleus and near focal adhesions (FAs) and stress distributions depended on flow direction. The presence of a 0.4 μm glycocalyx was predicted to increase intracellular stresses by ∼2-fold. The model of magnetic bead twisting rheometry also predicted heterogeneous stress, strain, and displacement fields resulting from material heterogeneities and FAs. Thus, large differences in moduli between the nucleus and cytoplasm and the juxtaposition of constrained regions (e.g. FAs) and unattached regions provide two mechanisms of stress amplification in sheared endothelial cells. Such phenomena may play a role in subcellular localization of early mechanotransduction events.     

Modulation of ATP/ADP Concentration at the Endothelial Surface by Shear Stress: Effect of Flow Recirculation

The extracellular presence of the adenine nucleotides ATP and ADP induces calcium mobilization in vascular endothelial cells (ECs). ATP/ADP concentration at the EC surface is determined by a balance of convective-diffusive transport to and from the EC surface, hydrolysis by ectonucleotidases at the cell surface, and flow-induced ATP release from ECs. Our previous numerical simulations in a parallel plate geometry had demonstrated that flow-induced ATP release has a profound effect on nucleotide concentration at the EC surface. In the present study, we have extended the modeling to probe the impact of flow separation and recirculation downstream of a backward facing step (BFS) on ATP/ADP concentration at the EC surface. The results show that for both steady and pulsatile flow over a wide range of wall shear stresses, the ATP + ADP concentration at the EC surface is considerably lower within the flow recirculation region than in areas of undisturbed flow outside the recirculation zone. Pulsatile flow also leads to sharp temporal gradients in nucleotide concentration. If confirmed experimentally, the present findings suggest that disturbed and undisturbed flow may affect EC calcium mobilization differently. Such differences might, in turn, contribute to the observed endothelial dysfunction in regions of disturbed flow.     

Collagen-Dependent Neurite Outgrowth and Response to Dynamic Deformation in Three-Dimensional Neuronal Cultures

In vitro models of brain injury that use thick 3-D cultures and control extracellular matrix constituents allow evaluation of cell–matrix interactions in a more physiologically relevant configuration than traditional 2-D cultures. We have developed a 3-D cell culture system consisting of primary rat cortical neurons distributed throughout thick (>500 μm) gels consisting of type IV collagen (Col) conjugated to agarose. Neuronal viability and neurite outgrowth were examined for a range of agarose (AG) percentages (1.0–3.0%) and initial collagen concentrations ([Col]_i; 0–600 μg/mL). In unmodified AG, 1.5% gels supported viable cultures with significant neurite outgrowth, which was not found at lower (≤1.0%) concentrations. Varying [Col]_i in 1.25% AG revealed the formation of dense, 3-D neurite networks at [Col]_i of 300 μg/mL, while neurons in unmodified AG and at higher [Col]_i (600 μg/mL) exhibited significantly less neurite outgrowth; although, neuronal survival did not vary with [Col]_i. The effect of [Col]_i on acute neuronal response following high magnitude, high rate shear deformation (0.50 strain, 30 s⁻¹ strain rate) was evaluated in 1.5% AG for [Col]_i of 30, 150, and 300 μg/mL, which supported cultures with similar baseline viability and neurite outgrowth. Conjugation of Col to AG also increased the complex modulus of the hydrogel. Following high rate deformation, neuronal viability significantly decreased with increasing [Col]_i, implicating cell–matrix adhesions in acute mechanotransduction events associated with traumatic loading. These results suggest interrelated roles for matrix mechanical properties and receptor-mediated cell–matrix interactions in neuronal viability, neurite outgrowth, and transduction of high rate deformation. This model system may be further exploited for the elucidation of mechanotransduction mechanisms and cellular pathology following mechanical insult. D. Kacy Cullen and M. Christian Lessing contributed equally to this work.     

Nano-topography: Quicksand for cell cycle progression?

The 3-D spatial and mechanical features of nano-topography can create alternative environments, which influence cellular response. In this paper, murine fibroblast cells were grown on surfaces characterized by protruding nanotubes. Cells cultured on such nano-structured surface exhibit stronger cellular adhesion compared to control groups, but despite the fact that stronger adhesion is generally believed to promote cell cycle progression, the time cells spend in G1 phase is doubled. This apparent contradiction is solved by confocal microscopy analysis, which shows that the nano-topography inhibits actin stress fiber formation. In turn, this impairs RhoA activation, which is required to suppress the inhibition of cell cycle progression imposed by p21/p27. This finding suggests that the generation of stress fibers, required to impose the homeostatic intracellular tension, rather than cell adhesion/spreading is the limiting factor for cell cycle progression. Indeed, nano-topography could represent a unique tool to inhibit proliferation in adherent well-spread cells.