Structure‐activity studies of the s‐echistatin inhibition of bone resorption

Synthetic Arg‐Gly‐Asp (RGD)‐containing peptides were examined in bone resorption or attachment and detachment assays with isolated mammalian osteoclasts in an effort to elucidate the mechanistic and structural basis for the inhibition of bone resorption by s‐echistatin. Bone resorption was the process most sensitive to inhibition by s‐echistatin, with IC 50 = 0.3 nM; inhibition of attachment to bone or detachment (lamellipodial retraction) was 30‐ to 70‐fold less sensitive, with IC 50 = 10 or 20 nM, respectively. Single amino acid substitutions within the 49‐residue sequence of s‐echistatin showed that although the efficacy of s‐echistatin is dependent on the Arg 24 ‐Gly 25 ‐Asp 26 sequence, additional residues, including Asp 27 , Met 28 , and Cys 39 , are also critical for potent inhibition of the resorbing activity of isolated rat osteoclasts. Because of the identification of the a v β 3 as the primary integrin on rat osteoclasts interacting the RGD peptides (Helfrich et al.), we examined the possibility of modeling bone resorption with other β 3 ‐mediated processes. Specifically, a v β 3 endothelial cell (human or rat) attachment to vitronectin and a IIb β 3 platelet aggregation were compared with bone resorption for sensitivity to s‐echistatin analogs, linear RGD peptides, and cyclic RGD peptides. Essentially no similarity in sensitivity to RGD peptides were observed between bone resorption, platelet aggregation, or endothelial cell attachment. Because rat osteoclasts and human giant cell tumors (osteoclastomas) shared similar sensitivity to s‐echistatin and rat and human endothelial cells showed a similar sensitivity profile to RGD peptides, the dissimilarity of bone resorption to other β 3 ‐mediated processes cannot be explained in terms of species differences. These data show that the osteoclast a v β 3 integrin is functionally distinct, despite epitopic similarity to the vitronectin receptor on other cells. Laser scanning confocal and scanning electron micrographs of rat osteoclasts on bone supported the hypothesis that resorption differs from the processes of attachment or detachment from substrates and suggest that inhibition of a minority of integrin‐bone interactions by s‐echistatin is sufficient to inhibit bone resorption. s‐Echistatin was also found to inhibit bone resorption in an organ culture system of neonatal mouse calvariae with IC 50 = 1 μM. Taken together, these data show that although it may be possible to target the osteoclast a v β 3 integrin for the development of pharmaceuticals, s‐echistatin itself is not useful in this capacity.