A mini review: proteoglycan aggregate profiles in the Pond-Nuki dog model of osteoarthritis and in canine disuse atrophy.

The Pond-Nuki dog model of osteoarthritis has characteristics which seem to mimic the human disease in early stages, particularly with respect to progressive changes in the cartilage matrix. Aggregating proteoglycans were studied using novel extraction and ultracentrifugation methods designed to separate very large macromolecules. With these methods two large peaks of proteoglycan (PG) aggregates (PGA-1 and PGA-2) were separated in preparative amounts and were shown to have unequivocal differences in composition in many respects. The profiles of these peaks have been studied as a function of joint location, topographic site, cartilage layer, presence of cartilage atrophy versus osteoarthritis, as well as treatment of the animals with various agents. Both link protein (essential for forming link-protein stabilized aggregates) and hyaluronate are required to regenerate normal aggregate profiles from the deficient aggregate fractions obtained from osteoarthritic cartilage. Canine proteoglycan link-stabilized aggregates (PGA-2) are confined to the middle and deep zone of cartilage. We believe that their reduction or elimination in the Pond-Nuki model results from a disturbance or loss of functional link protein (and hyaluronate), thereby weakening the middle and deep cartilage layers.