On the development of the patella.

The current view of skeletal patterning fails to explain the formation of sesamoid bones. These small bones, which facilitate musculoskeletal function, are exceptionally embedded within tendons. Although their structural design has long puzzled researchers, only a limited model for sesamoid bone development has emerged. To date, sesamoids are thought to develop inside tendons in response to mechanical signals from the attaching muscles. However, this widely accepted model has lacked substantiation. Here, we show that, contrary to the current view, in the mouse embryo the patella initially develops as a bony process at the anteriodistal surface of the femur. Later, the patella is separated from the femur by a joint formation process that is regulated by mechanical load. Concurrently, the patella becomes superficially embedded within the quadriceps tendon. At the cellular level, we show that, similar to bone eminences, the patella is formed secondarily by a distinct pool of Sox9- and Scx-positive progenitor cells. Finally, we show that TGFß signaling is necessary for the specification of patella progenitors, whereas the BMP4 pathway is required for their differentiation. These findings establish an alternative model for patella development and provide the mechanical and molecular mechanisms that underlie this process. More broadly, our finding that activation of a joint formation program can be used to switch between the formation of bony processes and of new auxiliary bones provides a new perspective on plasticity during skeletal patterning and evolution.

Irreversible changes in glycosaminoglycan composition of anatomically intact bovine articular cartilage induced by intermittent loading.

The changes in matrix composition induced by I MPa intermittent (0.2 Hz) loading of anatomically intact bovine articular cartilage in vitro are studied. The kinetics of chondrocyte response was determined in experiments where sesamoid bones of adult cows were loaded for 0, 3, 5 and 7 days. The reversibility of the induced changes were studied in sesamoid bones that were loaded for 5 days and subsequently cultured without loading for another 1, 2 and 3 weeks. Water content was not affected by loading, nor by subsequent unloaded culture. Glycosaminoglycan content was not affected by loading, nor by subsequent unloaded culture for another 2 weeks. However, after 3 weeks of culture following 5 days of loading, a significant decrease in glycosaminoglycan content was found. Glycosaminoglycan synthesis was already increased after 3 days of loading. Loading for longer periods (5 and 7 days) further increased the glycosaminoglycan synthesis rate. Glycosaminoglycan synthesis decreased during the first week of subsequent culture without loading. In the third week it dropped to a lower level than observed for the control. The amount of 3-B-3(-) epitope was increased and the length of newly synthesized glycosaminoglycans was decreased by intermittent loading. Subsequent unloaded culture did not further increase the amount of 3-B-3(-) epitope. However, after 3 weeks of unloaded culture the newly synthesized glycosaminoglycans were as long as those synthesized in the control. The results suggest that the changes in glycosaminoglycan chain length were reversible. However, the overall chondrocyte response to our loading regime seems to be detrimental to the tissue: expression of 3-B-3(-) epitope induced by loading together with the drop in glycosaminoglycan content and synthesis observed at the end of the total culture period indicate that the cartilage is irreversibly damaged.