Unlike other synthetic or physiological inhibitors for matrix metalloproteinases (MMPs), the ␤-amyloid precursor proteinderived inhibitory peptide (APP-IP) having an ISYGNDALMP sequence has a high selectivity toward MMP-2. Our previous study identified amino acid residues of MMP-2 essential for its selective inhibition by APP-IP and demonstrated that the N to C direction of the decapeptide inhibitor relative to the substratebinding cleft of MMP-2 is opposite that of substrate. However, detailed interactions between the two molecules remained to be clarified. Here, we determined the crystal structure of the catalytic domain of MMP-2 in complex with APP-IP. We found that APP-IP in the complex is indeed embedded into the substratebinding cleft of the catalytic domain in the N to C direction opposite that of substrate. With the crystal structure, it was first clarified that the aromatic side chain of Tyr 3 of the inhibitor is accommodated into the S1 pocket of the protease, and the carboxylate group of Asp 6 of APP-IP coordinates bidentately to the catalytic zinc of the enzyme. The Ala 7 to Pro 10 and Tyr 3 to Ile 1 strands of the inhibitor extend into the nonprime and the prime sides of the cleft, respectively. Therefore, the decapeptide inhibitor has long range contact with the substrate-binding cleft of the protease. This mode of interaction is probably essential for the high MMP-2 selectivity of the inhibitor because MMPs share a common architecture in the vicinity of the catalytic center, but whole structures of their substrate-binding clefts have sufficient variety for the inhibitor to distinguish MMP-2 from other MMPs.The matrix metalloproteinases (MMPs) 2 comprise a family of zinc-dependent endopeptidases capable of degrading protein components of the extracellular matrix and play pivotal roles in tissue remodeling under physiological and pathological conditions such as morphogenesis, angiogenesis, tissue repair, and tumor invasion (1-4). The association of MMPs with tumor invasion and metastasis has suggested that these proteases represent attractive target for the development of antitumor therapies. To date, a large number of MMP inhibitors based on hydroxamic acid derivatives or other synthetic inhibitors have been designed (5-8). However, none of them has been developed successfully as anti-tumor drugs mainly because of deleterious side effects; the broad specificity of the MMP inhibitors must be a stiff obstacle for developing safe and effective drugs. A common architecture of catalytic sites of MMPs probably relates to the broad specificity of the inhibitors. Moreover, recent studies (2, 9) suggest that some members of MMPs have anti-tumorigenic and anti-metastatic functions, and inhibition of their activities by broad spectrum MMP inhibitors therefore offset anti-tumor effects of the inhibitors, and even worse, stimulate tumor growth and metastasis.

Background: Lack of enzyme selectivity has been a major obstacle in developing MMP inhibitors into drugs. Results: A fusion protein consisting of APP-IP and TIMP-2 has a highly selective inhibitory activity against MMP-2. Conclusion: The exosite-assisted inhibitory mechanism makes APP-IP-TIMP-2 a highly selective inhibitor. Significance: APP-IP-TIMP-2 may be developed as an effective drug for treatment of diseases in which MMP-2 activity is involved.