Encapsulated chondrocyte response in a pulsatile flow bioreactor

Abstract

We have developed a bioreactor-based millifluidic technique that allows for dynamic culture conditions and measurement of the fluid flow impinging upon a three-dimensional tissue engineering scaffold. Chondrocytes in scaffolds have been shown to require mechanical stimulation to produce an extracellular matrix that resembles native cartilage. This study investigates the effect of pulsatile flow on chondrocyte response in a model poly(ethylene glycol) dimethacrylate hydrogel. Bovine chondrocytes were encapsulated in the hydrogel and cultured for 7, 14 and 21 days at pulsatile flow frequencies of 0.5 Hz (15 ml/min) and 1.5 Hz (17 ml/min). The scaffolds cultured under dynamic conditions were compared to those cultured under static (non-flow) conditions. Quantitative real-time reverse transcription polymerase chain reaction was used to quantify collagen type I, collagen type II and aggrecan gene copy numbers as markers for chondrocyte phenotypic expression. Histological sections stained with hematoxylin & eosin, and Alcian blue confirmed chondrocyte morphology and matrix formation. Interestingly, regulation of the collagen type II gene was particularly sensitive to the flow conditions. The understanding of the cell response to encapsulation and flow could be used to identify the appropriate culture conditions necessary to design and develop hydrogel carriers to promote the formation of extracellular matrix as well as to further our knowledge of chondrocyte mechanobiology.

Introduction

Regeneration of cartilage using tissue engineering technologies has been an active area of biomedical research for the repair of degenerative and injured cartilage conditions. Various research groups have shown that mechanical stimulation of chondrocytes in three-dimensional (3-D) matrices help develop and maintain the proper cell phenotype [1], [2], [3], [4], [5], [6], [7]. For connective tissues such as cartilage, the proper phenotype was generally defined by the type and amount of extracellular matrix (ECM) produced. Articular cartilage is largely composed of collagen type II and proteoglycans such as aggrecan. Collagen type II comprises approximately 90% of all collagen in human articular cartilage, and aggrecan comprises approximately 80–90% of all proteoglycans in human articular cartilage matrix [8]. Tissue-engineered cartilage containing similar amounts of these matrix components may have the necessary mechanical properties to serve as a replacement for damaged cartilage.

One challenge for engineering new cartilage is determining the appropriate culture conditions that promote the formation of functionally equivalent tissues [9], [10], [11]. Two popular approaches to provide the necessary mechanical signals for promoting chondrocyte phenotype are direct mechanical stimulation and stimulation by fluid stress. There is a large body of work on direct mechanical stimulation of cartilage explants and tissue-engineered constructs, many of which use the amplitude and frequency of compressive or shear loading as tissue culture variables. It has been observed that static compression up to 60% strain reduced the ECM produced by chondrocytes in alginate gels or cartilage explants, but cyclic strains increased the [³⁵S]-sulfate (proteoglycan content) and [³H]-proline (collagen content) incorporation [12]. The same study also found significant frequency dependence in the chondrocyte response, where the ECM production monotonically increased as the rate of compression increased from 0.01 to 1 Hz. Other studies also observed a strong dependence in the chondrocyte response on the magnitude of the static strain over which the oscillatory compression was superimposed [13].

In addition to direct mechanical stimulation, many bioreactors employ fluid stresses to promote the development of proper phenotype [14], [15], [16], [17], [18]. Comparing scaffolds cultured under static conditions with those cultured in mixed-flask and rotating-vessel bioreactors, the tissues developed in the rotating vessel bioreactor are superior [19]. Both the glycosaminoglycan (GAG) and collagen production in constructs cultured in the rotating vessel were significantly higher than those from the mixed flasks. The difference in bioreactor results has been attributed to the flow patterns; mixed flasks subject the cell–scaffold constructs to a turbulent flow while rotating vessels subject the samples to a laminar flow. Tissues from rotating vessels have concomitantly higher equilibrium modulus and dynamic stiffness, even approaching that of native cartilage [20].

The challenge with designing rotating vessel bioreactors is that samples must be dynamically suspended in a laminar rotational field that balances centrifugal and buoyant forces [21]. Chondrocytes respond sensitively to the hydrodynamic conditions, but the vessel rotation rate must be set to keep the construct in a constant state of freefall. Thus, the culture conditions are dictated in part by engineering constraints instead of entirely by what would be optimal for promoting tissue development. While it is possible to vary the vessel design to match the chondrocyte requirements, having a model bioreactor system to test the effects of flow and soluble factors on cell–scaffold constructs could be quite valuable.

In this paper, we use millifluidic techniques to build a perfusion bioreactor capable of applying fluid flows to chondrocytes embedded in photopolymerized hydrogels. Millifluidic techniques are well suited to bioreactor design because the fabrication is straightforward. We hypothesize that mechanotransduction of chondrocytes is a key variable that can be used to regulate the gene expression and ECM production by applying fluid stress to chondrocytes encapsulated in poly(ethylene glycol) dimethacrylate (PEGDM) gels.