AN ABSOLUTE or relative increase in bone resorption by osteoclasts results in the bone loss that characterizes osteoporosis. The recent discovery of osteoprotegerinligand (OPG-L) and its decoy receptor, osteoprotegerin, has altered our thoughts on how osteoclasts are formed, and possibly deferred potentially significant efforts to understand the function of these cells in bone resorption. From a therapeutic standpoint, however, it is perhaps equally efficient to inhibit resorption by mature osteoclasts as it is to prevent osteoclast formation. In this respect, the elegant study by Stroup et al. published in this volume revives our interest in that aspect of osteoclast biology which delves into fundamental mechanisms of matrix degradation by proteolytic enzymes. The authors report that a novel cathepsin K inhibitor, SB-357114, when administered to the hypogonadal nonhuman primate, Cynomolgus macaque, produces an acute, profound, and lasting suppression in biochemical markers of bone resorption. The Ki for SB357114 is in the low nanomolar range, well within its circulating micromolar concentrations. The drug’s in vitro inhibitory potency is also significantly higher, implying specificity for human cathepsin K compared with other human cathepsins, namely cathepsins S, B, and L, or indeed rat cathepsin K. Cathepsin K is a member of the cysteine protease family that, unlike other cathepsins, has the unique ability to cleave both helical and telopeptide regions of collagen 1, the major type of collagen in bone. Pharmacologic studies conducted in the 1980s formed the basis of our current thinking that a resorption hemivacuole, the unit of osteoclastic activity, provides the sealed microenvironment within which a low pH permits the action of acid-optimal enzymes, such as cathepsins, to initiate matrix cleavage. Inhibitors of cathepsins B and L were developed thereafter; however, these inhibitors had modest or no effects on bone resorption. It was not until 1996 that Gelb et al. discovered that pycnodysostosis, an autosomal recessive disease characterized by osteopetrosis and short stature was the result of cathepsin K gene mutations. Several other groups have since then identified nonsense, missense, and stop codon mutations in the gene. However, short of its complete absence, there is no phenotypic manifestation. Indeed, even a 50 to 80% reduction in cathepsin K expression noted in a family displaying compound heterozygosity causes no bone disease. Nevertheless, the requirement of cathepsin K in bone resorption is underscored by the key observation that the cathepsin K null mouse manifests osteopetrosis, characterized by dysfunctional matrix digestion, expectedly without a mineralization defect. Moreover, osteoclasts isolated from cathepsin K null mice do not effectively resorb bone in the pit assay. Thus, cathepsin K joins the growing list of molecules, including PU1, c-fos, NFkB, TRAF-6, OPG-L, c-src, PI-3-kinase, H-ATPase, and b3, that result in osteopetrosis when their respective genes are deleted. The bone phenotype displayed by cathepsin K null mice suggests strongly that redundant mechanisms involving either compensation by, or up-regulation of, other cathepsins is incomplete, implicating cathepsin K as a predominant effector of bone resorption and thus a viable therapeutic target (with some exceptions as noted below). This is consistent with abundant cathepsin K messenger RNA (mRNA) and protein expression in human and murine osteoclasts where cathepsins B, S, and L are not detected. The absence of a bone phenotype in mice deficient in cathepsin L also substantiates a minimal, if any, role of this alternate cathep-