Bone, and especially the subchondral bone plate, is involved in the pathogenesis of osteoarthritis (OA). OA bone tissue is sclerotic yet undermineralized indicating abnormal bone cell metabolism. Studies in both human and animal models of OA support the concept that bone sclerosis could precede cartilage degradation and loss. Clinical studies show that the indices of bone resorption and formation are increased in OA patients. A working hypothesis of the sequence of changes leading to OA holds that enhanced bone remodeling is the initiating event triggering cartilage damage. The attempt to repair the damaged cartilage then leads to a number of biochemical adaptations in bone and cartilage. In bone, this repair attempt modifies insulin-like growth factor 1 (IGF-1), IGF binding proteins (IGFBPs), and transforming growth factor-beta (TGF-beta), and alters the urokinase plasminogen activator (uPA)/plasmin system. In the cartilage, it also modifies IGF-1/IGFBP levels and the uPA/plasmin system. However, bone changes may overwhelm the attempts to repair cartilage, and lead to further sclerosis and damage. Some of these specific pathways have been investigated, and indeed are modified in OA subchondral osteoblasts. Thus, subchondral bone sclerosis in OA may be due to abnormal osteoblasts characterized by increased metabolic activities that result in an increase in osteoid matrix that is undermineralized. The exact role played by cytokines and prostaglandins remains controversial. However, restraining collagen deposition and mineral removal, and/or improving mineral deposition, could provide a better, more mineralized, bone matrix in OA patients.