Mechanical bioreactors for bone tissue engineering

Abstract

Static in vitro culture for cell monolayer and small explants has been regularly employed for many decades. These protocols have provided adequate nutrients and oxygenation to the cells or tissue by incubation in a temperature controlled and CO2/bicarbonate buffered environment. Scale up to increase the number of cells or size of a cell seeded scaffold leads to limitations in the diffusion of nutrients and waste products throughout the tissue or engineered construct. This often allows the formation of a construct that has cell viability and proliferation at the periphery but a cell seeded scaffold core that is necrotic. Bioreactors can control the environment for tissue engineering and conditioning such that these diffusion limitations are reduced. The biochemical environment in a bioreactor can be controlled by allowing the transport of nutrients, (e.g. glucose and dissolved oxygen) to the cells and the removal of degradation products away from the cells seeded throughout the construct. Factors such as pH, growth factors and other cell signalling molecules available to the construct can also be monitored. Figure 1 demonstrates some of these criteria that is needed to be considered when designing a bioreactor for bone tissue engineering. (Figure presented) Developing large-scale cell bioreactors for scale up has been widely applied in the cell recombinant technology processing industry. In these reactors, much work has been carried out into improving cell nutrient supply and viability when culturing large numbers of cells in mass culture. These circulation or mixing strategies improve mass transfer within the cultures but are not specifically designed to provide physiologically relevant levels of strain or mechanical mixing. These types of production units may meet the requirements for cell proliferation required as a source of high cell numbers for stem cell therapies without delivery vehicles or implant tissue production. In contrast for tissue engineering where the ultimate aim is an 'off the shelf' tissue, bioreactors aim to meet two requirements, firstly, to improve mass transfer as described above potentially through mixing or perfusion strategies and secondly, to apply physiologically relevant loads to the tissues which enables constructs to be conditioned according to the implantation environment and matrix production to be accelerated in vitro, thus reducing production time. This is particularly relevant to bone tissue engineering where load bearing is a major requirement for any replacement tissue. Studies in vivo have demonstrated that the biomechanical profile on the skeleton is made up of a number of components. Long bones such as the femur are not entirely straight but curvilinear in nature, therefore the application of force induces both localised compression and bending (Rubin and Lanyon, 1982). Bending induces both compression and tension forces on the opposing surfaces of a long bone with the neutral axis between the two found to experience rotational force (Jones et al., 1995). The nature of the applied load is important in determining any possible response, e.g. dynamic loading of approximately 1000 □icrostrain in vivo is known to induce bone formation (Rubin and Lanyon, 1984; Rubin and Lanyon, 1985; Turner et al., 1994a) which is absent when bones are subjected to static loading (Turner et al., 1994b; Lanyon and Rubin, 1984). The major parameters which define the effectiveness of a particular loading regime in stimulation of an osteogenic response are the strain magnitude, strain rate and frequency (Lanyon and Rubin, 1984; Rubin and Lanyon, 1984; Rubin and Lanyon, 1985; Rubin & McLeod, 1994; Turner 1998). The duration of the period of mechanical loading is also known to have an effect, but extended periods of loading are known to have a diminishing effect upon load induction (Rubin and Lanyon, 1984), as the bone formation response becomes saturated (Turner, 1998). The sensitivity of various bone cell types to minute changes in perceived strain is very high with deformations of 0.1% (1000 □icrostrain) sufficient to induce a response in vivo (Rubin & Lanyon, 1984) and at considerably lower strains in vitro (Thomas and El Haj, 1996; Salter et al., 1997, Walker, 1999). As not all of the cells within the bone experience the peak strain, then any response must be co-ordinated by intercellular signalling mechanisms. In bone fracture repair, the type and magnitude of loading (compressive, tensile, hydrostatic pressure and strain) seem to control which type of tissue will be formed from stem cells sources such as bone marrow in the callus tissue, i.e. fibrous tissue, fibrocartilagenous tissues or intramembranous bone (Claes and Heigele, 1999). High strains (5-15% deformation) have been proposed to induce connective tissue or fibrocartilage, wheras low strains (0.15-0.3%) of compression of tension have been proposed to induce intramembranous bone formation. With this in mind, many investigators have studied the effects of differing components of load, shear, tensile and compressive forces on bone cells in 2D and organ culture. The biomechanical environment can be controlled in many ways. Many different stages of bone cell differentiation are responsive to these types of mechanical forces. Cell activity (such as proliferation rate, differentiation or matrix production) has been shown to be influenced by the application of such forces. Shear stress can be applied to the cells via altering the flow rate and pattern of the culture media in the bioreactor. It is also possible to deliver tensile or compressive forces to cellular constructs in a bioreactor via a number of strategies. In this chapter, we outline the different methods of applying mechanical forces to cells using bioreactors and also describe the effects that these forces have on cell activity. © 2005 Springer.