MMP-9 antisense oligodeoxynucleotide exerts an inhibitory effect on osteoclastic bone resorption by suppressing cell migration Osamu Ishibashi a,1 , Satoru Niwa b,1 , Keiichi Kadoyama c , Takashi Inui d, ⁎ a Division of Molecular and Cellular Pharmacology, Niigata University Graduate School of Dental and Medical Sciences, 2-5274 Gakkocho-dori, Niigata 951-8514, Japan b Novartis Pharma Co. Ltd., 8 Okubo, Tsukuba 300-2611, Japan c Division of Molecular Regenerative Medicine, Department of Biochemistry and Molecular Biology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan d Department of Molecular Biology and Cell Informatics, Graduate School of Life and Environmental Sciences, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan Received 25 November 2005; accepted 24 May 2006 Abstract We have previously shown that matrix metalloproteinases (MMPs) play a role in osteoclastic bone resorption by facilitating migration of osteoclastic cells toward bone surface through matrices. Of MMPs identified so far, MMP-9 is likely the most important proteinase for the action, since osteoclasts express this enzyme at a tremendously high level. However, no direct evidence has been provided to demonstrate its contribution to bone resorption. In this study, to address this point, we used an MMP-9 antisense phosphothiorate oligodeoxynucleotide (S-ODN), which was shown to inhibit the protein synthesis of MMP-9 efficiently. We demonstrated that the antisense S-ODN inhibited osteoclastic pit formation on matrigel-coated dentine slices in a concentration-dependent manner with a maximum reduction of total pit volume by 53% at 10 μM. These results, taken together, suggest that MMP-9 is involved in osteoclastic bone resorption process possibly by facilitating migration of osteoclasts through proteoglican-rich matrices. © 2006 Elsevier Inc. All rights reserved. Keywords: Matrix metalloproteinase (MMP); MMP-9; Osteoclast; Bone resorption; Antisense; Basement membrane; Migration; Pit volume Introduction Bone matrix is composed of inorganic and organic compo- nents, which are represented by hydroxyapatite and type I colla- gen, respectively. Degradation of these components is a primary process of bone resorption. The degradation of the organic bone matrices is conducted by highly specialized hematopoietic cells designated osteoclasts, which produce and secrete mainly two types of proteinases, i.e., matrix metalloproteinases (MMPs) and lysosomal cysteine proteinases (LCPs). Previously, our in vitro study using selective inhibitors revealed that both MMPs and LCPs are necessary for osteoclastic bone resorption, and that they contribute to the process through distinct modes of action (Inui et al., 1999). Briefly, we demonstrated that MMPs are involved in migration of osteoclasts to resorption sites through matrices, whereas LCPs play a role in direct degradation of bone collagen. Consistently, Engsig et al., 2000 have reported that MMP-9 expressed in osteoclasts at an early stage of bone development is necessary for recruitment of the cells through non-mineralized cartilage matrices. As for LCPs, we showed that cathepsin K plays a central role in osteoclastic bone matrix degradation, by the use of cathepsin K-specific antisense S-ODN, whereas contribution of other LCPs, such as cathepsins B and L, to the process is almost negligible (Inui et al., 1997). This result is also supported by several biochemical and histological studies (Drake et al., 1996; Littlewood-Evans et al., 1997; Ishibashi et al., 2001). On the other hand, an issue as to which MMPs play the most important role in the osteoclast migration has not been conclusive yet. Several lines of studies have provided evidence that MMP-9, Life Sciences 79 (2006) 1657 – 1660 www.elsevier.com/locate/lifescie ⁎ Corresponding author. Tel.: +81 72 254 9473; fax: +81 72 254 9474. E-mail address: inuit@bioinfo.osakafu-u.ac.jp (T. Inui). 1 Equally contributed to this study. 0024-3205/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2006.05.024