Dynamic Loading of Porous Gels and Cartilage Can Induce Active Solute Transport

Albro, M B; Li, R; Yeager, K ; Hung, C L; Ateshian, G A Columbia University, New York, NY [email protected] INTRODUCTION: Transport pathways play a key role in maintaining cellular metabolic activity. Efforts to maintain or enhance the transport of nutrients may prove beneficial to native and engineered tissue, particularly in avascular tissues such as articular cartilage. Various studies have investigated the potential of dynamic mechanical loading to increase the uptake and desorption rate of solutes in articular cartilage [1-3]. However, recently a novel concept has been theoretically suggested that such dynamic loading of porous deformable media, may additionally yield higher steady state concentrations of solutes, beyond those achieved by passive diffusion [4]. To investigate this theoretical prediction, this experimental study implements dynamic loading on a dextran agarose model transport system, which is representative of some cartilage tissue engineering studies [5], as well as on native cartilage. METHODS: Materials: Agarose gels of 7% and 9% w/v were prepared with type VII 2-hydroxyethyl agarose powder (Sigma, St Louis MO) added to phosphate buffered saline (PBS) and heated on a hotplate for 30 minutes. The liquid agarose was then cast between glass slides where it was allowed to gel for 45 minutes before being transferred to a PBS bath. Tracer solutions of FITC conjugated dextran (70kDa, Invitrogen) were prepared in PBS at 0.5 mg/mL. After casting, agarose discs (∅3.85mm×2.3mm) were incubated in dextran solution for 24 hours. Middle zone cartilage discs (∅4mm×2.5mm) were harvested from the femoral condyles of a bovine calf. Agarose loading: After incubation, 7% and 9% agarose gels were subjected to a dynamic loading (DL) routine while exposed to dextran solution. Discs were placed in unconfined compression and loaded with a dynamic compressive strain (±5% amplitude) superposed over a static strain offset (15%) at a frequency of 1 Hz. Concurrently, a second control group of discs remained in a petri dish filled with dextran solution under unloaded passive diffusion (PD) conditions. After 5, 25, and 45 hours of loading, discs from each group (n=6) were removed, sectioned into quadrants and placed in a PBS desorption solution for 48 hours. Interstitial solute content was subsequently determined with a standard fluorescent assay and expressed as ĉ , the concentration of dextran in the gel normalized to the concentration of the bathing solution. Agarose recovery: After 45 hours of loading, one group of 9% discs was returned to a dextran filled petri dish and maintained unloaded to investigate whether the prior effect of loading on solute uptake was reversible. At four time points during this recovery period, discs (n=6) were removed and assayed for their solute content. Solute desorption: Pre-incubated 7% agarose discs were exposed to a solute-free PBS solution to induce dextran desorption from the gel. One group was subjected to the aforementioned dynamic loading routine while another was held under a static 15% compressive strain. Discs from each group (n=6) were assayed for their dextran concentration at 5, 25, 45, and 120 hours. Cartilage loading: While exposed to dextran solution in unconfined compression, cartilage discs (n=8) were dynamically loaded (±20% amplitude) at 0.2 Hz frequency with no static offset. This configuration, which intentionally prescribes platen lift-off on the upstroke mimics the physiologic loading regimen cartilage experiences under sliding reciprocation. RESULTS: For cartilage and agarose gels, the concentration ratio under passive diffusion, ĉ PD was observed to be less than unity (Fig. 1). For cartilage and both gels, dynamic loading significantly increased the dextran concentration ratio, ĉ DL , relative to ĉ PD , as time increased (p<0.001); the enhancement ratio was ĉ DL ĉ PD =3.5 ± 1.3 for 7%, ĉ DL ĉ PD =6.4 ±