The use of computational fluid dynamic models for the optimization of cell seeding processes

Abstract

The seeding of a porous scaffold with stem cells is a fundamental step in engineering sizeable tissue constructs that are clinically viable. However, a key problem often encountered is inhomogeneous seeding of the cells particularly when the cells are delivered through the thickness of the scaffold. The objective of this study was to establish the quantitative relationships between the cell seeding efficiency and the initial vacuum pressure in a compact perfusion seeding device that uses the effect of differential pressure induced by vacuum to seed cells on a porous scaffold. A transient CFD solution of the fluid flow in the device was used to optimize the initial vacuum pressure for efficient cell seeding. Results indicate that the optimal initial vacuum pressure for homogenous cell seeding is approximately −20 kPa for the seeding device. This study presents a 3-D computational model that can be employed in designing and optimizing cell seeding techniques and corresponding technology.

Introduction

A key problem, often encountered in tissue engineering of sizable construct is an inhomogeneous loading or seeding of cells on scaffold. Cell seeding is the process of incorporating cells into or unto the scaffold prior to culture or implantation. An efficient seeding technique will minimize cell injury, reduce cell seeding time, seed cells uniformly on the scaffold, be highly reproducible and be user independent [1]. Cells can be seeded on the scaffold by lining the cells on the surface of the scaffold (surface seeding) or by delivering the cells through the thickness of the scaffold (bulk seeding). Surface seeding is by far the most commonly used seeding method since bulk seeding is difficult to achieve in a controllable manner. Furthermore, the use of cell seeding devices has proven to be challenging because it involves mechanical forces that may result in shear-mediated lysis or triggering apoptotic pathways [2], [3]. Martin and Wendt in a review paper reported various methods of cell seeding namely; static, stirred flask and perfusion method [4]. In Static method the cells are spread on the surface of the scaffold with the aid of a micropipette. This method can readily be used for any cell type and scaffold configuration. However, many studies have reported low seeding efficiencies [5], [6], [7], [8], [9]. In the stirred flask method, scaffolds are suspended in a well-mixed spinner flask filled with the cell suspension. The mixing provides a relative velocity between the suspended cells and the scaffold. Thus, the cells are transported to and into the scaffold by convection [6], [10], [11], [12], [13]. A problem often encountered in stirred flask cell seeding is non-uniform distributions of cells with a higher density of the cells lining the scaffold surface. For the direct perfusion method, the cells suspension flows directly through the scaffold thus depositing cells directly into the scaffold pores. Comparison of these three methods suggested that the direct perfusion method produced the highest seeding efficiency [13]. It is therefore worthwhile to pursue further advances in the direct perfusion method of seeding.

Over the past few years, different perfusion seeding devices have been designed by researchers. As a fundamental element, these devices are designed to better exploit the improved cell suspension transport into the scaffold obtainable in the perfusion technique. Wendt and Marsano [13] employed the use of oscillatory motion of fluid in a U-shaped tube to force the medium with the cell suspension through the scaffold. Though higher seeding efficiency was obtained in this approach, but the use of this technique in a surgical environment that requires timely seeding of the cells into the porous scaffold is questionable. In another approach, Soletti and Nieponice [1] developed a seeding device that uses the synergistic effects of vacuum, centrifugal force and fluid flow to seed the scaffold. However, this technique is limited to hollow tubular tissue construct. Thus, Govil et al. [14] developed a novel compact perfusion seeding device that utilizes the vacuum induced suction effect and the corresponding pressure differential to initiate the flow of the cell suspension into a cavity that holds the scaffold. During this process, the cell suspension soaks the scaffold thereby depositing the cells uniformly in the scaffold.

Computational fluid dynamics (CFD) is a set of numerical techniques applied to the fluid conservation equations in order to obtain approximate solutions to the fluid flow as well as heat and mass transfers. For instance, CFD methods are able to solve the Navier–Stokes equations to determine entire fluid flow fields such as velocity, pressure, density, temperature etc. CFD modeling provides a detailed, efficient and non-destructive tool to theoretically evaluate and characterize large numbers of parameters that influence cells, tissue and organ in the context of tissue engineering without having to perform numerous experiments. In recent years, researchers have used computational fluid dynamics to characterize 3-D flows in tissue engineering bioreactors with different configurations. For example, CFD simulations have successfully been used to quantify shear stresses acting inside microstructures [15], [16], [17]. In another example, CFD simulations were used to numerically characterize fluid flow within a spinner flask under operating conditions used in cartilage tissue engineering [18]. Similarly, several groups have used computational simulations to demonstrate how scaffold morphology influences hydrodynamic shear stresses and nutrients concentration restrictions imposed in cells within construct [19]. Hence, CFD simulations have become a tool for tissue engineers to understand the influence of fluid flow and transport on cell function without having to perform many and expensive bioreactor experiments. Consequently, CFD simulation provides significant insight into the design and optimization of cell seeding devices while simultaneously saving time and resources. It is noteworthy, however, that little work has been done in applying CFD modeling to design and optimize cell seeding techniques and devices in tissue engineering. Till now, a transient CFD simulation of a seeding technique and device that allows the manufacture of reproducible and reliable engineered constructs and able to reach general clinical application is lacking.

Therefore, the aim of this study is to introduce a time dependent computational fluid dynamics model to characterize the fluid flow in the compact seeding device developed by Govil et al. [14] and subsequently optimize the device for efficient cell seeding.