A Bioreactor System for Magnetic Resonance Microimaging and Spectroscopy of Chondrocytes and Tissue Engineered Cartilage Osteoarthritis is one of the leading causes of joint pathology in the older population. One approach to control of this disease is the use of chondrocyte transplantation, with eventual elaboration of neocartilage from the transplanted cells. Studies in monolayer chondrocyte culture have been critical in the development of these techniques. However, the monolayer system does not reflect the true three-dimensional nature of the developing chondrocyte-cartilage system. The ideal in vitro system for investigating the regulation of cartilage formation and maintenance would allow for three dimensional tissue growth, a wide range of biochemical interventions, and non-destructive evaluation. We have accordingly investigated neocartilage tissue grown in a hollow fiber bioreactor (HFBR) system which meets these criteria and which is MRI microscopy-compatible. We have carried out correlations between structures visible under various MRI contrast modalities and cartilage microstructure. In cartilage formed by cells from whole chick sterna, MRI microscopy revealed the development of stromal layers between the elaborated matrix centered about each hollow fiber. Density images show decreased mobile water content in these layers. There is little additional density variation within the developed neocartilage. Further insight into water content and mobility of the stromal layers and of the more homogeneous neocartilage is gained from T1 and T2 contrast images. Just outside the fiber walls, we find high proton density with relatively low mobility. Mobility increases with distance from the hollow fibers within the growth units, corresponding to differences in cell size and density. In magnetization transfer contrast images, we find that the lowest km values correspond to areas of high proteoglycan concentration, found in the mid-regions of the growth units. In contrast, the stromal layers and the regions around the fibers, which are relatively proteoglycan poor, show the highest km values, potentially indicating greater collagen-water interactions. Temporal development of the tissue has also been investigated. Tissue volume and cellularity increased during development over four weeks; this was accompanied by changes in MR properties, including relaxation times and water diffusion. The increase in collagen content as measured by biochemical assays was particularly striking, and was readily observable by magnetization transfer maps of the tissue. Extensions of this imaging work have tested the hypothesis that MRI results correlate with the biochemical composition of cartilage matrix, and can therefore be used to evaluate natural tissue development and the effects of biologic interventions. This was investigated by forming detailed correlations between tissue hydration, GAG concentration and collagen concentration, as determined biochemically, and MRI determined relaxation times, magnetization transfer rates, and water diffusion coefficient. Besides natural development, the response of the tissue to administration of retinoic acid, interleukin-1b, and daily dosing with ascorbic acid was studied. Overall, we find that MRI-derived parameters can be used to assess the composition of cartilage forming in the bioreactor system. Osteoarthritis is, in part, characterized by an imbalance between catabolic and anabolic processes in articular cartilage. Therefore, any analgesic or antiinflammatory agent which disrupts the already-limited repair potential of cartilage may be sub-optimal for therapeutics. Previous cell culture studies have demonstrated that NSAIDs may have significant catabolic effects on developing cartilage matrix; however, these two-dimensional culture experiments necessarily did not incorporate the diffusion of the agents into tissue, such as occurs in articular cartilage. These studies also did not permit non-invasive longitudinal measures of matrix growth. Therefore, we are examining the effects of high therapeutic doses of aspirin and ibuprofen on the development of cartilage in the HFBR. We find that ibuprofen greatly decreases the amount of tissue growth and degrades its biomechanical properties. This corresponds to a substantial decrease in matrix fixed charge density, reflecting decreased proteoglycan content. Our preliminary results indicate that both aspirin and ibuprofen adversely impact the main constituents of extracellular matrix in a three-dimensional cartilage culture system. We have used 31P NMR spectroscopy to assess the metabolic stability of the developing cartilage over time, and to gain insight into metabolic adaptations as chondrocytes mature. We have established the presence of NMR-visible phosphocreatine in this system and have found a significant decrease in intracellular pH during early development of the tissue. We have also characterized the T1 relaxation times of g-ATP, phosphocreatine, and inorganic phosphate (Pi) in the developing cartilage. Over a four week period of development, the T1 of Pi in cartilage derived from chondrocytes harvested from the proximal sternum of chick embryos increased significantly. These results are consistent with decreasing molecular mobility in the proximal sternum tissue, which may be related to the fact that in vivo, the proximal sternum ossifies, while the distal sternum remains cartilaginous. Further metabolic work on engineered cartilage tissue has centered on the development and application of a new electron paramagnetic resonance (EPR) imaging technique, EPROM (EPR oxygen mapping) permitting direct visualization of tissue oxygen concentration. EPROM relies upon the well-known fact that the linewidths of many EPR spin probes increase monotonically with oxygen content. We have applied EPROM to study oxygen utilization by chondrocytes in cartilage matrix. We found that addition of cyanide, an inhibitor of the mitochondrial electron transport chain, increased oxygen levels as compared to control tissue. This directly demonstrates utilization of aerobic metabolism in our developing cartilage system. An ongoing extension of these studies addresses the question of whether MRI determination of fixed charge density (FCD) in hyaline cartilage under non-steady-state situations, such as during histogenesis of cartilage or during rapid matrix turnover, correlates with tissue biomechanics. This question is of particular relevance for tissue growth and repair. These experiments will potentially have a large impact on the treatment of osteoarthritis by offering a new modality for monitoring cartilage tissue engineering protocols, therapies based on implantation of autologous chondrocytes, and the efficacy of pharmacological therapies. To purse this, we are first utilizing proton MRI microscopy to determine matrix FCD during the natural development of cartilage in the HFBR and to correlate it with biomechanical and biochemical properties of the tissue. The data we have obtained has shown the promise of our overall approach, with decreased FCD as measured noninvasively by MRI showing a correlation with decreased tissue mechanical stiffness. This is the first study of the relationship between matrix properties, biomechanics, and MRI-derived tissue biomechanical properties in engineered tissue.
Showing the most recent 10 out of 36 publications
