CA2261567A1 - The use of proteinase inhibitors for prevention or reduction of bone resorption - Google Patents

Abstract

Bone metabolic disease is treated by inhibition of the production or action of membrane-type matrix metalloproteinase (MT-MMP) or the matrix metalloproteinase 12 (MMP-12) involved in the resorptive activity of osteoclasts. Inhibitors for MT-MMP and MMP-12 and membrane-associated metalloproteinase activity include peptides and analogues of peptides generated using a PEGA bead library, antisense nucleic acid agents and antibodies. The proteinases MT1-MMP and MMP-12 are found to be expressed in osteoclasts and may be selectively inhibited.

Description

CA 02261~67 1999-01-27 W098/04287 PCT~P97/04110 THE USE OF PROTEINASE INHIBITORS FOR PREVENTION
OR REDUC~ION O~ BONE RESORPTION

The present invention relates to the reduction of the 5 rate of bone resorption by targeting the action or ~ production of proteases.
Human bone is constantly undergoing remodelling. The fine balance between bone formation and bone resorption is regulated by local and systemic factors and by physical 0 forces acting on various cells including, in the bone environment, osteoblasts and osteoclasts. However, in several bone metabolic diseases including most importantly osteoporosis and osteolytic bone metastasis, the balance is disturbed resulting in a sustained pathological net bone 15 resorption.
Osteoporosis is a systemic skeletal disease characterised by low bone mass and microarchitectural deterioration of bone tissue, with a subsequent increase in bone fragility and susceptibility to fracture. Post-20 menopausal osteoporosis is a chronic disease which affectsmillions of women throughout the world and it has an enormous economical and social impact on society.
Reduction of bone resorption is believed to be an appropriate way to prevent and treat several metabolic bone 25 diseases, including osteoporosis and osteolytic bone metastasis. Agents such as steroid hormones ~especially oestrogen), calcitonin and bisphosphonates are able to suppress bone resorption and have been used for prevention and treatment of osteoporosis and/or osteolytic bone 30 metastasis. ~owever, these therapeutic agents fail to achieve satisfactory effects in some cases, due to subject limitation or uncertain efficacy. There is therefore need of a new prophylactic/ therapeutic method for preventing or treat ng accentuated bone resorption.
Remo~al of the mineralised osseous substance, i.e.
orgar.ic matrix embedded in deposits of calcium phosphate CONFIRMATION COPY

CA 02261~67 1999-01-27 W098/0~287 PCT~P97/04110 salts, is a complicated process. Though still a contro-versial subject, it seems probable that osteoclasts are the only cells capable of bone resorption. The progressing ~one loss in patients with osteoporosis is caused by an increase 5 in the activity of osteoclasts.
The expected life cycle of osteoclasts involve the following major phases:

1. recruitment of haematopoietic stem cells, the early precursor of osteoclasts, 2. proliferation and differentiation, 3. fusion into multinuclearity, 4. attachment to the resorptive bone surface, 5. polarisation and removal of mineralised osseous substance, and 6. death by apoptosis, necrosis or a more random process.

These phases are, however, not necessarily separate events, thus, e.g. differentiation might take place during 20 migration to the resorptive surface and fusion might take place on the bone surface. All these phases represent possibilities for intervention in order to regulate the level of bone resorption.
Traditionally, proteolytic enzymes have been known to 25 play a role in degradation of the organic matrix of bone.
The knowledge about proteolytic enzymes involved in bone resorption mainly comes from in vitro and in vivo studies of the effects of natural and particularly synthetic enzyme inhibitors. ~urthermore, histochemical and immuno-30 cytochemical characterisation of enzymes in bone cells andtissues as well as more recently identification of enzyme-encoding mRNA in osteoclasts and other bone cells has increased the information about proteolytic enzymes involved ~n bone resorption. The proteolytic enzymes of major 3~ relevance to osteoclastic bone resorption seem to be members of the families of cysteine proteinases and matrix metalloproteinases ~MMPs).

W098l04287 PCT~P97/04110 The use of proteinase inhibitors in disease control has been suggested in several scientific publications and in patents and patent applications. For MMP inhibitors the maln focus has been the potential of inhibitors in treatment 5 of cancer and tumour metastasis, but also diseases such as arthritis, ulcers, periodontal and bone diseases, ~IV
infection, corneal and other eye diseases, diabetes and myocardial infarction have been the target of these speculations and ensuing early experiments (reviewed by 10 ~3irkedal-Hansen et al, 1993 2).
In some particular cases, however, the studies have been emphatic leading to particularly important conclusions and products of relevance to the use of proteinase inhibitors in disease control. Selected peptidyl 15 derlvatives were shown to be effective inhibitors of metalloproteinases reaching Ki-values down to 5 pM for MMP-2 by kinetic studies based on a fluorogenic synthetic peptide substrate incubated with MMP-l, -2 or -3 and the substances were orally active and non-toxic in mice at suitable doses 20 (W094/25434).
Membrane-type matrix metalloproteinases (MT-MMPs) were originally identified in cancer cells and have been implicated with the migration of these cells (Sato et al l99413). Based on this disclosure, its seems that the use of 25 MT-MMP inhibitors will be appropriate for the reduction of the spread of tumours. No studies have, however, yet described inhibitors of MT-MMPs and thus no data are available on the use of MT-MMP inhibitors as agents in the treatment of diseases. From the usually low selectivity of 30 synthetic MMP-inhibitors, it seems probable that some established MMP-inhibitors will inhibit MT-MMPs.
Furthermore, cDNA encoding MTl-MMP (also referred to in the literature as MT-MMP-l and as MMP-14) as well as anti-MTl-MMP antibodies have been suggested, though rather 35 unspecifically, as useful for applicatlon not only in the diagnostic area but also in other medical fields (FP-A-0685557 and W095/25171).

CA 0226l~67 1999-01-27 7 PCT~EP97/04110 The inhibition of cathepsins is considered another possible way of reducing bone resorption by using proteinase inhibitors. Several cathepsins are produced by osteoclasts and though still somewhat controversial, they are apparently 5 lnvolved in the degradation of organic matrix in the acidic environment of the sub-osteoclastic resorption zone.
Recently a novel cathepsin named cathepsin K, cathepsin O or OC2 was cloned from osteoclasts and osteoclast-like cells by several independent groups. It was suggested that lO development of antisense probes or synthetic inhibitors to this proteinase could be of value in the treatment of several diseases including osteoporosis. For cathepsin L
several compounds have been produced for use as specific inhibitors in the treatment and prevention of osteoporosis 5 (EP-A-0611756).
The general use of hybrid molecules for conferring specificity to cell- and tissue-interacting agents has been proposed in several modifications including hybrids consisting of three parts including not only a cell-binding 20 ligand and a chemical entity to be introduced into the target cell but also an intermediate part constituting a translocation domain for enabling the entrance of the chemical entity into the cell (WO91/0987). Another approach to resist clearance and degradation and ease the uptake in 25 cells of peptides and proteinase inhibitors is by administering them as lipid conjugates (W093/01828).
Speculations about the biological roles of osteoclastic proteinases have been almost entirely focused on their potential ability as mediators of degradation of organic 30 bone matrix in the sub-osteoclastic resorptive zone.
However, our recent findings have shown that proteolytic enzymes are also very important for the migration and attachment of osteoclasts to the resorptive surface (Blavier ~ Delaissé, 19953). Furthermore, the proteinase-3~ dependent migration of immature osteoclasts seems to beassociated with the maturation into active bone-resorbing T

W 098/04287 PCT~EPg7/04110 osteoclasts as well as of importance for the events leading to fusion into multi-nuclearity, i.e. osteoclast differentiation processes.
Being an earlier phase of the osteoclast life cycle, 5 interference by an inhibitor of a proteolytic enzyme involved in osteoclast migration and/or attachment might be more effective than inhibition of an enzyme involved directly in the resorptive process. This type of interference will also be easier to accomplish since the lo secreted enzymes of the migrating cells are not protected from inhibition as they are when secreted into the tightly sealed resorption zone which is formed when the active polarised osteoclasts attach to bone.
We have now discovered that an MT-MMP closely related lS to or identical to MT1-MMP, previously identified in cancer cells not related to bone, is expressed by osteoclasts. It may be expected that this osteoclast MT1-MMP plays an important role in the action of osteoclasts, probably being implicated in their migration to their site of action at 20 which to degrade bone (see Examples 1, 2, 3-2 and 3-3 and Figures 1 to 3). This finding indicates that also other membrane-associated metalloproteinases such as other MT-MMPs or members belonging to families of non-matrix type of membrane metalloproteinases (e.g. meltrins and "A
25 disintegrin and metalloproteinase"'s, ADAMs)) could be produced by osteoclasts.
Furthermore, we have identified and characterised the full length gene and the encoded protein of osteoclast metalloelastase MMP-12, a proteinase hitherto believed to be 30 almost specifically expressed in macrophages, where it is obligatory for the invasion of these cells through basement membranes. Since macrophages and osteoclasts are closely related cell types both originating from the haematopoietic stem cell and differentiating late in its development, a 35 similar role of MMP-12 in osteoclast invasion and migration ~s likely (see Fxample 3-4 and Figures 4 to 6) .

CA 02261~67 1999-01-27 The present invention provides the use of an agent in the manufacture of a medicament for the treatment of bone metabolic disease, characterised in that the agent acts by inhibition of the production or action of a membrane 5 associated protease or the matrix metalloprotease MMP-12 involved in the resorptive activity of osteoclasts. More preferably, the invention provides the use of an agent in the manufacture of a medicament for the treatment of bone metabolic disease by inhibition of the production or action 10 of a metalloproteinase involved in the resorptive activity of osteoclasts. Particularly, inhibition of the production or action of an MT-MMP but also of other membrane-associated metalloproteinases such as a meltrin or an ADAM as well as a secreted MMP such as MMP-12.
The treatment may be for prevention or for cure of such diseases.
Preferably, the metalloproteinase is involved in the recruitment, proliferation, differentiation, or migration of osteoclast precursor cells or in the migration, fusion, 20 attachment, polarisation, activity in removal of mineralised osseous substance, or death of osteoclasts.
Though MT-MMP and MMP-12 produced by osteoclasts and osteoclast precursors is a major target for the inhibitory agent of the invention, the invention also includes 25 regulation of bone metabolism by inhibition of non-osteoclastic proteinases which influences the life cycle of osteoclasts. Other bone cells such as osteoblasts and chondrocytes are able to produce both latent and active forms of MMPs, cathepsins and plasminogen activator as well 30 as natural inhibitors of some of these enzymes. These enzymes might be important for the initial degradation of the bone surface exposing the underlying mineralised matrix to subsequent osteoclastic action (Delaissé & Vaes 19925) and they might be involved in the degradation of collagen fibres 35 either released from the bone by the action of osteoclasts or still remaining in the resorption pit after the osteoclast has left (Foged et al, 19966). Furthermore, latent pro-forms of osteoblastic enzymes stored in bone CA 02261~67 1999-01-27 W098/04287 P~ 97/04110 might be activated during osteoclast resorption. Finally, proteolytlc enzymes of non-osteoclastic origin might have a chemotactic role in regulating the migration and maturation of osteoclasts.
The agent may be selectively inhibitory of MTl-MMP or MT-MMPs broadly, of MMP-12 or MMPs broadly, or of membrane-associated metalloproteinases or metalloproteinases broadly.
The agent may be an antibody selectively immunoreactive with an MT-MMP. Such an agent may alternatively be an antisense oligo-nucleotide or oligo-nucleotide analogue directed against a gene involved in the production of an MT-MMP or an agent regulating MT-MMP activity. It may be an MT-MMP substrate mimic inhibitor. It may be a broad spectrum matrix metalloproteinase (MMP) inhibitor or a broad 15 spectrum membrane-associated metalloproteinase inhibitor.
It may also be a peptide, peptide analogue or other peptide mimicking agent obtained by screening an appropriate library for compounds reactive with an MT-MMP, an MMP or a membrane-associated metalloproteinase.
A preferred inhibitor provided by the invention is the peptide S-K-Y-P-J-A-L-F-F-K (SEQ ID ~o.l) (J being the single letter code of hydroxyproline) and inhibitory variants thereof such as the peptide analogue S-K-Y(NO2~-P-J-A-L-F-F-K(Abz) (SEQ ID No.2).
In an alternative aspect, the lnvention includes the use of an agent in the manufacture of a medicament for the treatment of bone metabolic disease by inhibition of the recruitment, proliferation, differentiation, or migration of osteoclast precursor cells or in the migration, fusion, 30 attachment, polarisation, or death of osteoclasts.
Preferably, said agent produces said inhibition by inhibiting the production or action of a proteinase.
The invention includes an anti-bone resporption agent comprising a proteinase inhibitor active against a 35 proteinase involved in bone resorption operatively linked to a ligand having binding specificity targeting the inhibitor to said proteinase or to the environment of the proteinase.

The invention includes a new protease termed rabbit osteoclast MT1-MMP having the amino acid sequence given in ~igure 1 and Figure 2, as well as an isolated nucleic acid coding for such a protein, e.g. one having the sequence set 5 out in Figure 1. Proteins having high e.g. more than 75% eg more than 90% or 96% homology to the said rabbit osteoclast MT1-MMP are included also, as is human osteoclast MT1-MMP
and isolated nucleic acid sequences encoding it.
The invention also includes a new protease termed 10 rabbit osteoclast MMP-12 having the amino acid sequence given in Figure 4 and Figure 5, as well as an isolated nucleic acid coding for such a protein, e.g. one having the sequence set out in Figure 4. Human osteoclast MMP-12 and isolated nucleic acid sequences encoding it as well as other 15 proteins and nucleic acid sequences with a high homology (e.g. at least 50%, preferably at least 70, 80 or 90%) to rabbit osteoclast MMP-12 are also included in the invention.
. .
Inhibition of proteolytic activity can be obtained in several ways and by several classes of agents. The 20 inhibition could be direct, i.e. by an agent acting directly either on the proteinase in its active form(s) inhibiting its proteolytic activity or substrate recognition or on the latent form of the proteinase inhibiting its conversion into active proteinase. The most relevant directly acting 25 inhibitors of proteinases include:

1. natural inhibitors which form specific complexes with an active proteinase and in some cases even with its latent pro-enzyme (e.g. tissue inhibitors of metalloproteinases, TIMPs);
2. antibodies or antibody fragments which e.g. neutralise the active site or block the substrate recognition site;
3. synthetic pseudo-substrates which specifically interact at the catalytic site (e.g. synthetic peptides linked to a chelating group) or the natural substrate recognition site; and CA 02261~67 1999-01-27 WO 98/04287 PCT~EPg7/04110 4. so-called entrapping reagents which are cleavable substrates which when cleaved undergo a conformational change which leads to entrapment of the proteinase (e.g. a-macroglobulins).

The inhibition, however, could also be indirect i.e. by an agent regulating either the expression and/or production of the proteinase (e.g. a natural transcription factor or its naturally regulating systemic or local factor, or a l~ synthetic antisense probe specifically binding to and blocking the mRNA encoding the proteinase) or by an agent influencing the level or activity of a natural regulator of the proteinase (e.g. an inhibitor of an enzyme responsible for catalytic activation of the target proteinase).
The development of many types of proteinase inhibitor is assisted by having the proteinase itself available. The productior. of proteinases may be performed either directly in cultures of isolated osteoclasts or indirectly by transfection of an expression plasmid containing proteinase 20 encodirg c~NA into a recipient cell line. For proteinase productior. in osteoclasts, the majority of e.g. MMP-9 is produced n its latent proform (pro-MMP-9) and therefore needs a sLosequent activation process if the active form is required. The amount of proteinase obtained from production 25 in osteoclast is severely restricted by:

a) the r.on-proliferative nature of osteoclasts in culture and b) the _echnical difficulties in isolation of native osteoclasts in high numbers and purity.

For llustration, the production, purification and activatior of osteoclastic pro-MMP-9 is described in Example 3-1. In ~ontrast, both latent and active protelnase can be 35 produced ~irectly by recombinant techniques depending on whether ~he expression plasmid-transfected into the recipient cell is designed to contain the complete cDNA or a cDNA devo-d of the region encoding the propeptide moiety of CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 the latent enzyme. Since active proteinases are generally less stable than their corresponding latent pro-enzymes and particularly under cell culture conditions might be degraded, production of latent proteinases is often 5 preferable. For illustration, the identification and cloning of cDNA encoding several osteoclastic MMPs or parts thereof, including MMP-9, MMP-12 and MT1-MMP is described in Examples 1, 2 and 3-4.
Apart from natural regulators of metalloproteinase and lo particularly MMP production and activity, agents inhibiting metalloproteinases (including MMPs and especially MT-MMPs and MMP-12) involved in one or more phases of the osteoclast life cycle can include:

15 1. a substance which interacts at a specific site of the metalloproteinase or MMP thereby reducing its proteolytic activity to recognise a natural substrate, e.g. anti-MMP antibodies and fragments thereof as well as synthetic, peptide-mimicking proteinase inhibitors;
20 2. substances which influence the transcription or translation of metalloproteinase or MMP;
3. substances stimulating the level or activity of a natural inhibitor of metalloproteinase or MMP; and 4. substances reducing the level or activity of a natural activator of metalloproteinase or MMP, e.g. a substance analogous to the description in 1. and 2. but regulating a proteolytic enzyme responsible for activation of latent MMP.

Examples 5 and 6 below describe the development of inhibitory agents; the production and use of anti-proteinase antibodies (Example 5); the production, identification and characterisation of synthetic, peptide-mimicking proteinase inhibitors (Example 6 a-e); and the design and use of 35 antiser.se probes to proteinase mRNA (Example 6f).
Anti-proteinase antibodies are central tools for the development of proteinase inhibitors and under appropriate conditions can be used as inhibitors themselves (see Example CA 02261~67 1999-01-27 5e and Figure 9). Thus, the applications for anti-proteinase antibodies and parts thereof are several and inparticular anti-MMP antibodies and antibody fragments will be useful:

l. In the production of recombinant MMP by use in immuno-blotting or a similar immunodetection method for identification of clones expressing recombinant proteinases.
lo 2. In affinity chromatographical purification of native or recombinant MMPs by immobilisation on activated resins produced for affinity columns such as e.g. divinyl sulfone agarose.
3. In immunoassays such as ELISA or RIA for quantitative determination of the specific MMP concentration in samples for diagnostic analysis e.g. tissue extracts, sera or urine samples, and in samples for research analysis e.g. cell culture medium.
4. In immunocytochemical identificatlon of MMP-expression on the protein level by incubation with bone cells or tissue sections. As shown in Example 5, this can also lead to a demonstration of a particular cellular localisation of a MMP and thereby aid in the clarification of its biological role.
25 5. In the characterisation of MMP-activity by use as specific inhibitory agents. Antibodies have shown the highest specificity as MMP inhibitors in test tubes (i.e. selectivity for a particular MMP and not others) and therefore will be important tools for characterisation of individual proteinases (Birkedal-Hansen et al, l9932). Especially, antibodies raised by immunisation with peptides mimicking a region comprising the catalytic site of a particular MMP could be expected to interfere with the proteolytic activity '- of this member but not other members of the MMP-family and thereby be of importance for the demonstratiOn of the specific role of particular proteinase in bone metabolism.

CA 02261~67 1999-01-27 W0 98/04287 12 PCT~EP97/04110 6. In the manufacture of a medicament for the treatment of bone metabolic disease by use as direct MMP inhibitors or as constltuents of hybrid MMP inhibitors. Two general principles for using anti-MMP antibodies or fragments thereof for treatment of bone metabolic disease are relevant: as direct inhibitors of proteinase activity or as site-directing agents merely assuring that another inhibitory agent is transported to the right target cell or tissue, e.g. by hybridisation on the protein or gene level of the antibody or a fragment thereof to a peptide-mimicking synthetic inhibitor. In both cases the use of antibodies in treatment of a bone metabolic disease requires its administration to animal or man in a proper pharmaceutical composition to avoid degradation and to ensure a beneficial effect.

Synthetic peptide and peptide-mimicking inhibitors of proteinases are promising agents for use for treatment of 20 bone metabolic disease by inhibition of the action of proteinases involved in the recruitment, proliferation, differentiation, or migration or osteoclast precursor cells or in the migration, fusion, attachment, polarisation, removal of mineralised osseous substance, or death of 25 osteoclasts. Several methods for production of peptide and peptide mimicking inhibitory agents are available, two of which are described in Example 6 (a-e).
One is based on a recently developed beaded polyethylene glycol cross-linked polyamide (PEGA) resin 30 designed for peptide synthesis and with an open structure permitting biologically active proteins into the interior ~Meldal et al11, 1994: Meldal & Svendsen, 199512). The PEGA
bead peptide library was developed for the complete characterisation of the specificity of proteinases in 35 general and can be used for identification of first synthetic peptide substrates of osteoclast proteinases and subsequently inhibitors after a well-functioning substrate has been identified. In the first step of this procedure , W098/04287 l3 PCT~P97/04110 millions of randomly synthesised fluorogenic peptides are screened for their ability to become hydrolysed during incubation with an osteoclast proteinase. The major purpose of this step is to identify a syntheti peptide substrate 5 suitable for use in the second step of the procedure, i.e.
- the identification of inhibitors of the same proteinase.
However, the identification of substrates might lead directly to inhibitory agents, since substrates with high affinity for the proteinase but little ability to become lo hydrolysed (i.e. pseudo-substrates) can act as reversible inhibitors. In Example 6b, we report the finding of a peptide-mimicking molecule ~C~-l) identified by incubation of MMP-9 with a PEGA bead substrate library, which has a low Km (3.4 ~M) but also a low kCat/Km (<500 M-1s~1) suggesting its 15 potential use as an inhibitor of osteoclastic MMP-9. Even better nhibitory characteristics of pseudo-substrates can be expected after modification of the originally identified substrates, e.g. either by linking peptide-mimic~ing sub-strates to chelating groups such as hydroxamates, thiols, 20 phosphonamidates, phosphinates and phosphoramidates (reviewed by Birkedal-Hansen et al, l9932) or by designlng pseudo-substrates which easily forms acyl-proteinase complexes but which hydrolyse slowly due to interaction with the blndlng site on the enzyme for the leaving group ~Baggio 25 et al 19961), In the more regular cases where the ldentification of an appropriate synthetic substrate (i.e. showing a low Km and a high Kcat/Km by incubation with the proteinase) is obtained either by the first step of the PEGA bead procedure or by 30 simply being already commonly available, synthetic peptide inhibitors can be identified among millions of randomly desi~ned peptides in a PEGA bead synthetic peptide inhibitor library (Meldal and Svendsen, l9951Z; Meldal et al, 199721).
The screening is based on the rare ability of some peptides 35 to inhibit the hydrolysis of the established synthetic peptide-mimicking substrate. Inhibitors of MMPs, MT-MMPs and membrane-associated metalloproteinases can be found by this method also.

CA 0226l~67 1999-01-27 W O 98/04287 14 PCT~EP97/04110 A novel modification of the original PEGA bead inhlbitor technology was developed in order to optimise the synthesis of MMP inhibitors. It has previously been shown (Galardy et al, 199213) that substitu~ing the cleavable 5 peptide bond ~-CO-NH-) in a peptide substrate of fibroblast collagenase by a phosphorus-containing bond e.g., a phosphinate (-PO2-CH2-), phosphonamidate (-PO2-NH-) or phosphonate (-PO2-O-) bond can cause inhibition of the proteolytic activity. For the first time, this knowledge 10 has been used in combination with the PEGA bead technology by ex~ending the group of building blocks used for synthesis of putative inhibitory peptide analogues on the PEGA-beads from just natural amino acids (including hydroxyproline) and thei- corresponding D-forms to also including pseudo 15 dipeptides such as NH2-PlP/CPl'-COOH, NH2-PlP/NP1'-COOH or NH.-PlP/C?l'-COOH, where the two normal amino acids (P1 and P1') instead of being linked through the peptide bond are linked thro..gh the phosphinate, phosphonamidate or phosphonate bond (P/c~ P/N or P/~). This allows the synthesis of random PEGA-20 bead inhibitor libraries with a s~ructure such as: X1-X2-PlP/~?1'-X3-~4-"linker"-PEGA, where X1 to X4 are natural amino acids and PlP/CPl is a phosphinate pseudo dipeptide (as described in Example 6c and Figures 12-15).
By employment of the PEGA-bead substrate library 25 technology, it has been possible to identify peptide seqLences which are of use in the deslgn of novel highly specific MMP-substrates (see Example 6 a and b). These subs rates facilitate the design and use of PEGA-bead inhibitor libraries both through the use of one of these 30 selective substrates in the library and through the use of the substrate sequence data for the design of the structure o~ _he randomised inhibitors in the library (Meldal and Svendsen, 199512; Meldal et al, 199721). Particularly in the design of PEGA-bead inhibitor libraries based on inhibitors 3 witr a phosphorous containing bond, the substrate data were usec for determination of the two amino acid R-groups around the phosphinate, phosphonamidate or phosphonate of the pseLdo dipeptide (see Example 6c). Furthermore, the design r W098/04287 P~ 57/04110 of selective inhibitors based on the characteristics of the novel MMP-substrate specificities will be facilitated (see data for CL-l, CL-21, CL-25 and CL-29 in Example 6b).
Finally, the specific substrates could become important 5 tools for selective detection and quantification of MMPs in tissue samples in diagnosis and research.
The other method for identification of peptide and peptide mimicking inhibitory agents is based on the use of posltional combinatorial peptide inhibitor libraries. A few lo members of these libraries of randomly synthesised peptides having in a single amino acid position an abnormal amino acid, such as a D-amino acid instead of an L-amino acid, in some case will act in an inhibitory way to a particular enzyme, probably due to a pseudo-substrate effect. If an 15 inhibitory signal is obtained by incubation of a positional combinatorial peptide inhibitor library with a proteinase or a biological model system includlng essential proteinase activity, the peptide(s) in the library responsible for this inhibition must be subsequently identified by systematic 20 segmentation of the library as described in Example 6 (d-e) for incubation of positional combinatorial peptide inhibitor libraries with murine foetal metatarsal cultures. Some preferred inhibitory libraries and peptide structures provided by the invention are the libraries X-X-w-X-X, X-X-25 1-X-X and X-X-w-Y-X and the peptides C-L-w-Y-L, C-L-w-Y-M, C-Y-w-Y-L, V-Y-w-Y-M and L-F-w-Y-L, where X are natural amino acids including hydroxyproline, and w and l are D-tryptophan and D-leucine, respectively (see Example 6e).
Comparing the two methods, the major advantage and 30 disadvantage of the PEGA bead library are the immediate identification of inhibitors and the need for incubation with a preferably purified proteinase preparation in a test tube, respectively. The major advantage and disadvantage of the positional combinatorial peptide inhibitor library is 35 the possibility to screen directly for an lnhibitory effect in a biological test system and the need for several cumbersome segmentations of the initial library to identify the agent originally causing the inhibition, respectively.

CA 02261~67 1999-01-27 WO9B/04287 PCT~EP97/04110 ~inally, one feature of the positional combinatorial peptide inhlbitor library can be seen as both favourable and non-favourable, since the functional background for an inhibltory response induced in the biological system by this s type of library is uncertain i.e. the inhibitory peptides might not be proteinase inhibitors but have other regulatory functions.
A review by Eggleston and Mutter of methods for producing inhibitors mimicking inhibiting peptides appears lC in "Chemistry in Britain" May 1996, pages 39-4118. The techniques reviewed may be applied to peptides identified by the methods discussed above.
The benefits of using antisense probes to proteinases can be divided into two major aspects, an early aspect and a later aspect. The antisense probes are important tools for evaluation of the role of the corresponding proteinase in a biological process, because they can be used at an early stage of a study when anything else than the oligonucleotide sequence cf this proteinase is unknown, and this even with 20 usually high specificities i.e. with only a minor risk of cross-reaction to other proteinases if the design of the antisense probe and the experimental conditions are appropriate. Antisense probes were used successfully for inhibition of MMP synthesis by fibroblasts (Lin et al, 25 19959), and interfered with the proton pump activity of osteoclasts when assessed in both cell and tissue cultures (Laitala and Vaananen, 19948). Another major aspect of using antisense probes is their possible application in the treatment of diseases caused by over-expression of 30 particular genes. For specific reduction of proteinase levels, gene therapeutic use of antisense probes to MMPs may be expected to be effective.
The identification of an antibody-derived or synthetic peptide-mimicking inhibitor of an osteoclast proteinase may ~5 be followed by appropriate modification of this compound to assure its use as a medicament for the treatment of bone metabolic disease. Several characteristics are necessary, partlcularly sufficient uptake and stability in the living CA 02261~67 1999-01-27 organism to assure a beneflcial effect, sufficient tissue or cell specific action to assure maximal effects at the target site of the organism relative to effects at non-target sites lncluding acceptable levels of side effects, and a pharma-5 cologically acceptable dose- and time-response to the treatment.
Administration of proteins, peptides and peptide-like substances to animals and humans requires protective routes of administration and/or protective formulation of the o peptide in order to avoid degradation of the compound.
Though protective encapsulation for oral administration of peptides and peptide-like agents is a technology currently undergoing significant improvement, stabilisation of the agent itself prior to administration is advantageous. For 15 peptide-mimicking MMP-inhibitors this has been possible by chemical modification of an initially identified compound apparently without important changes in its inhibitory capacity (Brown & Giovazzi, 19954 and P. D. Brown personal communications June 1996).
Targeting of a proteinase inhibitor to e.g. osteoclasts and osteoclast precursors, can be obtained by two general means. One, is if the inhibitor due to its intrinsic specificity selectively reacts with the proteinase present on these cells either because the proteinase at this target 25 cell is particularly available to the inhibitor (due to e.g.
the localisation of the cell, the localisation of the proteinase in the cell or simply by a local high concentration of the proteinase) or because the proteinase when produced by these cells is different from the 30 corresponding proteinase as it is expressed in other cells and tissues (due to e.g. post-translational modifications).
The other way to obtain a specificity is by making hybrid molecules or conjugates combining one part of the agent having proteinase-inhibitory characteristics with another 35 part having antibody or ligand specificity for the particular cells or tissue. These hybrids can be made by recombinant expression of fusion-proteins after cloning of a hybrid cDNA. E.g. a piece of cDNA encoding the osteoclast-CA 02261~67 1999-01-27 W O98/04287 PCT~E~97/04110 18 specific ligand calcitonin (or a receptor-binding part thereof) can be ligated to another piece of cDNA encoding a pept_de lnhibitor for an osteoclast proteinase. Hybrids can alsG be conjugates of two compounds e.g. by chemically 5 linking an amino-bisphosphonate, which has high affinity for hydroxyapatite in bone, or an antibody specific for a component exposed in the osteoclast membrane, such as the calcitonin receptor with a peptide or peptide-mimicking proteinase inhibitor.
lo The invention will be further described and illustrated with reference to the examples which follow and the appended drawings in which:

Figure 1 shows the nucleotide (SEQ ID No.3) and deduced 15 amlno acid sequence ~SEQ ID No.4) of the MT1-MMP or MT1-MMP
analogue identified in rabbit osteoclasts;

Figure 2 shows a comparison between the amino acid sequence of the novel MT-MMP identified in rabbit osteoclasts 20 (Rabbit) (SEQ ID No.4) and the previously reported amino acid sequences of Human (SEQ ID No.5), Rat (SEQ ID No.6) and Mouse MT1-MMP (SEQ ID No. 7). Positions with an amino acid identical in all 4 proteins are indicated (*);

2~ Figure 3 shows schematically the structure of three MT1-MMP
cDN.. constructs and the corresponding control construct used in _xample 3-2;

Figure 4 shows the nucleotide (SEQ ID No.8) and deduced 30 amino acid sequence (SEQ ID No.9) of the MMP-12 or MMP-12 ana ogue identified in rabbit osteoclasts;

Figure 5 shows a comparison between the amino acid sequence of the novel MMP-12 identified in rabbit osteoclasts 3 (Rabbit) (SEQ ID No.9) and the previously reported amino acl~ sequences of Human (SEQ ID No.10), Rat (SEQ ID No. 11) ar.~ Mouse MMP-12 (SEQ ID No.12). Positions with an amino acid identical in all 4 proteins are indicated (*);

CA 02261~67 1999-01-27 W098/04287 rCTAEP97/04110 Eigure 6 shows schematically the structure of a MMP-12 cDNA
construct and the corresponding control construct used in Example 3-4;
-Figure 7 shows the effect of various proteinase inhibitorson the migration of purified osteoclasts through collagen coated membranes. The values are relative to the number of migrations observed in the absence of proteinase inhibitor.
lG
Figure 8 shows the effect of an MMP-inhibitor on pit formation by purified osteoclast seeded on dentine slices which were either not coated or coated with collagen. The values are relative to pit formation in the absence of 1 collagen coating and MMP-inhibitor;

Figure 9 shows the dose dependent inhibitory effect on MMP-9 proteolytic activity of sera from mice immunised with the conjugated femta-peptide RSGAPVDQMFPGVPL (SEQ ID No.13) 20 (peptide B, mimicking a region of the rabbit MMP-9 hemopexin domain) alone or together with purified intact rabbit osteoclast pro-MMP-9. No inhibitory effect was observed for sera from non-immunised mice and for mice immunised with another non-related femta-peptide (peptide A). The values 25 are relative to the average relative fluorescence generated during 30 minutes of incubation of the synthetic quenched fluorogenic substrate Mca-PLGL-Dpa-AR-NH~ (Bachem) (SEQ ID
No. 14) with a pre-incubated mixture of purified activated MMP-9 and the appropriate dilutions of 9 different control 30 sera (non-immunised or immunised with non-relevant femta-peptides);

Figure 10 shows the relationship between the initial velocity of enzymatic hydrolysis and the substrate 3 concentration determined by continuous fluorometric assay of MMP-9 or subtilisin with either (A) MR2:
Abz-G-P-L-G-L-Ln~r-A-R-Y(NO2)NH2) (SEQ ID No.15) or (B) CLl:
Abz-S-K-Y-P-J-A-L-F-Y(NO2)-D (SEQ ID No.16). Assays were CA 02261~67 1999-01-27 W 098/04287 PCT~EP97104110 performed at 37~C, pH 7.5 and fluorescence read at ~ex = 320 nm and ~em = 425 nm. Peptide origin and kinetic parameters are reported in Table 1;

5 Figure 11 shows inhibition of hydrolysis of CL1 by the MMP-inhibitor RP59794, but not the cysteine proteinase inhibitor E-64. MMP-9 (80 pmol) or subtilisin (3.4 pmol) were pre-incubated with either RP59794 or E-64 in a total volume of 40 ~l for 5 min at 37~C. Subsequently, 1 ml of 2.8 ~o ~M CL-1 was added and the incubation continued for 2 to 70 hrs. Inhibitor is listed in final concentrations;

Figure 12 shows the synthesis of the phosphinate analogue to hydroxyproline for use as a building block in the subsequent 5 generation of a hydroxyproline-methionine phosphinate pseudo dipeptide (see also Figure 13). The phosphinic acid analogue to trans-hydroxyproline is synthesised from potassium D- or L-erythronate. After bromination at the 2 and 4 position the acid is transformed into the methyl ester by methanol 20 quenching. The 2-position is reduced and the ester converted into the alcohol by sodium borohydride reduction.
The primary alcohol is oxidized by sodium hypochlorite to the aldehyde and condensed with tritylamine. The imine formed is reacted with bis-trimethylsilyloxyphosphine to 2~ yield the phosphinate. Upon acid hydrolysis and intra-molecular substitution of the bromine the free hydroxy-proline is obtained;

Figure 13 shows the synthesis of the hydroxyproline-30 methionine phosphinate pseudo dipeptide for use in preparation of the P~GA bead phosphinate inhibitor library IIa (see Example 6c). The phosphinic acid analogue of hydroxyproline (see Figure 12) is derivatised with benzyloxycarbonyl chloride. 2-methylene-4-methyl mercapto-3c butanoic acid ethyl ester was synthesised fromdiethylmalonate sodiation and reaction with methyl mercaptoethyl chloride followed by selective basic ester hydrolysis, acid decarboxylation and reaction with CA 02261~67 1999-01-27 WO 98/04287 P~ 97/04110 formaldehyde in the presence of piperidine. These reactions can be performed on a large scale. Reaction with the phosphinic acid analogue of hydroxyproline gives the dipeptide isosteric phosphinate. The phosphinate is 5 protected by reaction with adamantylbromide followed by ester hydrolysis with sodium hydroxide. The Cbz group is cleaved hydrogenolytically and the free amine protected by reaction with FmocCl and sodium carbonate;

o Figure 14 shows the synthesis of the glycine-leucine phosphinate pseudo dipeptide for use in preparation of the PEGA bead phosphinate inhibitor library IIb (see Example 6c). The phosphinic acid analogue of glycine is synthesised from tritylamine and formaldehyde to give the 5 imine which is reacted with bis-trimethylsilyloxyphosphine obtained from ammoniumphosphinate and hexamethyl disilazane. The product is deprotected by acid hydrolysis and is derivatised with benzyloxycarbonyl chloride. 2-Methylene-4-methyl pentanoic acid ethyl ester was 20 synthesised from diethylmalonate sodiation and reaction with isobutylbromide followed by selective basic ester hydrolysis, acid decarboxylation and reaction with formaldehyde in the presence of plperidine. Reaction with the phosphinic acid analogue of glycine gives the dipeptide ~5 isosteric phosphinate. The phosphinate is protected by reaction with adamantylbromide followed by ester hydrolysis with sodium hydroxide. The Cbz group is cleaved hydrogenolytically and the free amine protected by reaction with FmocCl and sodium carbonate;

Figure 15 shows the development and structure of the PEGA
bead phosphinate inhibitor library (IIa) based on the hydroxyproline-methionine phosphinate pseudo dipeptide. The invariable quenched fluorescent substrate (here: Ac-Y(NO2)PLJMKGK(Abz)G-"Linker"-) (SEQ ID No.17) and the randomly variable phosphinate inhibitor (here: X1X2J2/CMX3X4-"Linker"-) are independently associated to the PEGA bead.
Alternatively an FmocLys(Aloc) residue can be used to .

CA 02261~67 1999-01-27 WO 98/04287 22 PCT~EP97/04110 obtain orthogonal protection and incorporation of the two compounds and the order of synthesis of the library and the substrate may be reversed. This gives the possibility to use the same library with several substrates. The analogous 5 library (IIb) was prepared similarly by using an invariable substrate corresponding to MR1 (see Table 3) and a randomly variable phosphinate inhibitor X1X2GP/CLX3X4-''Linker''-;

Figure 16 shows inhibition of the ~sCa2+-release from foetal 0 murine metatarsals cultured for 4 days in the presence of posit1onal combinatorial pentapeptide inhibitor libraries.
The results for 5 selected libraries with the sequence X-X-D-X-X are shown. In these 5 cases D was either D-isoleucine, D-leucine, D-lysine, D-serine or D-tryptophan, and X were 15 randomly varying L-amino acids. In contrast to the libraries with D-lys and D-ser, the pentapeptide libraries with a D-ile, D-leu or D-trp at the third position induced a significant reduction of bone resorption. The MMP-inhibitor RP59794 was included as a positive control.
Example 1 Isolation of cDNA encoding fragments of osteoclastic proteinases.
2~
The use in PCR of degenerate nucleotide primer sets (designed from existing data describing the amino acid sequences of proteinases) for cloning of osteoclastic proteinases was exemplified by the studies described below 30 leading to the identification of MMP-9, MMP-12 and MT1-MMP
mRNA in rabbit osteoclasts:

a. Isolation and purification of osteoclasts Osteoclasts were isolated from 10-day-old rabbits (125-35 150 g) according to a method described previously (Tezuka etal, 1992l5) but with some modifications. Briefly, bone cells were released from marrow-depleted long bones and shoulder blades by mincing and mechanical agitation. A preparation CA 02261~67 1999-01-27 W098/04287 PCT~7/04110 of unfractionated bone cells rich in osteoclasts was isolated by centrifugation (30 x g, 5 min) and seeded into tlssue culture dishes. After a settling period of 90 minutes, non-adhering cells were removed, and cultivation 5 continued for 20 hrs at 37~C and 5 to 7.5% CO2 in a-MEM (pH
7.3) supplemented with 5% foetal calf serum. The cells were washed with PBS and then treated with 0.001% pronase E and 0.02% EDTA for approximately lO min. to release all non-osteoclastic cells. The purified osteoclasts were cultured lo for another 2 hrs before isolation of mRNA.

b. Amplification of MMP cDNA fragments by PCR, molecular cloning and ~omology analyses To identify possible MMP gene expression by rabbit 15 osteoclasts, cDNA reverse-transcribed from mRNA from the purified osteoclasts was subjected to PCR with degenerate primers designed from conserved reglons of MMP genes.
Briefly, the poly(A)+RNA from purifled osteoclasts was prepared using a mRNA purification kit (Pharmacia Biotech, 20 Uppsala, Sweden); single strand cDNA was synthesised from mRNA by use of a cDNA synthesis kit (Pharmacia); and aliquots of the synthesised cDNA were amplified by PCR with degenerate primers corresponding to the conserved amino-acid sequences in either the cysteine switch region (PRCGVPD (SEQ
25 ID No.18)) or the region resembling a cleavage site for furin (RRKRYA (SEQ ID No.l9)) in combination with the catalytic domain (GDXHFDXXE (SEQ ID No.20), where X is a variable amino acid) present in most members of the MMP-family. The PCR reactions were cycled 45 times through the 30 following steps: l min at 94~C, l at 55~C, l min at 74~C.
Three cDNA bands 330-340, 380-390 bp and 560-570 bp in length were identified by electrophoresis in a 1% agarose gel. The cDNAs were purified and cloned into a pCRII vector (Invitrogen, San Diego, CA) according to the instruction manual and subsequently characterised by nucleotide sequencing.

CA 02261~67 1999-01-27 2q The high expression of MMP-9 mRNA by rabbit osteoclasts is well-known and from previous characterisation of the MMP-9 gene the expected size of MMP-9 cDNA fragments amplified with degenerate primers used in this PCR would be 336 bp.
5 Our cloning and subsequent nucleotide sequencing confirmed that the isolated 330-340 bp cDNA originated from MMP-9.
The cloning of isolated 560-~70 bp cDNA, resulted in a clone, B4 with a length of 567 bp which by nucleotide sequencing was found to share more than 80% similarity with 0 a segment of the human metalloelastase (MMP-12) gene. The presence of mRNA encoding MMP-12 has previously been preliminarily identified in rabbit osteoclasts by partially sequencing randomly chosen cDNAs of an osteoclast cDNA
library (Sakai et al, 1995l2) (see also Example 3-4).
The cloning of isolated 380-390 bp cDNA, resulted in another clone, A3 with a length of 387 bp, which shared more than 90% similarity with the human MT1-MMP cDNA sequence previously reported in cancer cells (Sato et al, 1994l3).
Since neither MT-MMPs nor any other membrane-associated 20 proteinases have been previously identified in osteoclasts, the remaining part of this example as well as Example 2 describes studies of A3 and MT-MMP in osteoclasts.

c . Isola tion of MTl -MMP cDNA from an osteoclast cDNA
25 library A rabbit cDNA library (Tezuka et al, 199415) was screened by colony hybridisation, using the random-primed 32P-labelled PCR product of A3 as a probe. By screening lx105 clones, one positive clone was identified and made into 30 the plasmid form according to the instruction manual ~Stratagene, lambda ZAP vector). This positive clone contained a cDNA insert of 1,842 bp which was isolated and sequenced. An open reading frame consisting of 1716 bp initiated with an ATG codon at nucleotide position 127 was ~5 found. According to gene bank searches, an identical nucleotide sequence did not exist and the highest similarity was 91% to the human MT1-MMP gene. Figure 1 shows the nucleotide sequence of the cloned insert. The deduced T

CA 02261~67 1999-01-27 amino-acid sequence of the insert showed 96% similarity with human MT1-MMP (~igure 2). There were no additions or deletions of specific sequences when compared to MT1-MMP of other species. Based on further comparisons of amino acid s sequences of other MMPs, we concluded that the isolated novel cDNA encoded the rabbit homologue of MT1-MMP or of a closely related but previously unreported human osteoclast MT-MMP.

10 d. Nucleotide sequence analysis The nucleotide sequence analysis of the A3 PCR fragment and of the rabbit MT1-MMP cDNA clone from the cDNA library was determined from both strands by the dideoxy chain-termination method using the Qiagen-purified plasmid DNA
~s (Qiagen, ~SA), the Sequenase kit (U.S.B., USA), and either pBluescript SK primers (Stratagene, USA) or synthetic oligonucleotide primers.

Example 2 Identification of MT1-MMP in osteoclasts.

The novel identification of MT1-MMP in osteoclasts was further substantiated by the studies described in the 25 following examples:

a. Cells and organs for RNA preparation Brain, kidney, liver, lung, calvaria, spleen and alveolar macrophages were isolated from 10-day-old rabbit.
30 Bone stromal cells were obtained from a culture of unfractionated rabbit bone cells (Tezuka et al, 199215) in alpha-MEM containing 10~ ~BS until confluence, and then subcultured 4 times. In all cases total RNA was prepared as reported previously ~Tezuka e~ al, 1992l5).
b. Nort~ern blotting To investigate the mRNA expression of MT1-MMP in purified osteoclasts and to compare its level with that in CA 02261~67 1999-01-27 other tissues and cells, we performed Northern blotting.
Five micrograms of total RNA isolated from various organs and cells were blotted on nylon membranes after formaldehyde agarose gel electrophoresis, and hybridised with radioactive s probes. The A3 PCR fragment and a fragment of human MT1-MMP
cDNA (position 1647-2880, Sato et al, 199413) as well as (for quantitative normalisation) a synthetic oligonucleotide corresponding to 28 S ribosomal RNA were used as probes.
The cDNA probes were radiolabelled with a multiprime DNA
lo labelling system (Amersham International plc., Buckinghamshire, England) using ~alpha-32P]dCTP and the oligonucleotide probe was radiolabelled with a 5'-end labelling kit (Amersham) using [gamma-32P]ATP. Hybridisation was performed as described previously (Tezuka et al, 1992l5) ]5 and visualised by a Phosphorimager SF (Stratagene, La Jolla, CA). For both MT1-MMP probes, we found the same pattern of distribution as those reported previously for adult human tissues (Takino et al, 199514; Will and Hinzmann, 199517), and in addition a prominent expression of MT1-MMP in purified 20 osteoclasts. It was noteworthy that expression was not detectable in liver and brain and low expressions were found in bone stromal cells and alveolar macrophages.

c . In si tu hybridisa tion The expression of MT1-MMP in osteoclasts in vivo was examined by in situ hybridisation on sections of rabbit metatarsals. Consecutive paraffin sections of metacarpal bones of new-born rabbits were prepared as previously described (Blavier and Delaissé, 19953). A fragment of 30 rabbit MT1-MMP cDNA (position 1-318, corresponding to 126 nucleotides in the non-coding 5'-region and 192 in the region encoding the N-terminal part of MT1-MMP) was used for probe synthesis. Digoxygenin-labelled antisense or sense RNA probes were prepared by use of a DIG RNA labelling kit 3~ (Boehringer Mannheim) according to the instruction manual and compared to paraffin sections stained for tartrate-resistant acid phosphatase (Blavier and Delaissé, 19953).
Many tartrate-resistant acid phosphatase-positive multi-CA 02261~67 1999-01-27 nucleated cells were positive for MT1-MMP, whether they were attached to calcified cartilage or to bone.

d. Immunocytoc~emistry An important property of the MT1-MMP in previous investigated non-osteoclastic cells is its localisation in their plasma membrane. The expression of MT1-MMP at the protein level and its cellular localisation in osteoclasts was investigated by immunocytochemistry. Unfractionated : rabbit bone cells were seeded on glass coverslips. After 1.5 hr cultivation the non-adherent cells were discarded and the remaining cells were cultured for 1 to 18 hr, fixed and processed for immunocytochemistry. They were incubated for 90 min in the presence of 1-3 ~g/ml of the monoclonal MT1-: MMP antibody 113-5BT (Fuji Chemlcal Industries, Ltd.
Takoaka, Japan). This antibody was raised against a synthetic peptide corresponding to an amino acid sequence (CDGNFDTVAMLRGEM) (SEQ ID No.21) which differs by 1 amino acid from the corresponding rabbit sequence ~V in rabbit instead of M in human at position 10). Rhodamine-labelled donkey anti-mouse IgG (Jackson ImmunoResearch Laboratories, Inc. West Grove, PA) was used as secondary antibody at 200 times dilution. When incubating osteoclasts with an antibody against MT1-MMP we found fluorescence at specific points of its plasma membrane. Fluorescence did not appear when the MT1-MMP antibody was replaced by non-immune IgG.
All bright signals were in the focal plane where the cells were seen in contact with their substrate. In moving cells, mainly the extremities of the lamellopodia were illuminated.
In spread cells, the signals were arranged in a ring of small dots at the cell periphery. This pattern is reminiscent of podosomes. These are small extensions of the plasma membrane, that become abundant and organise in this particular way when the osteoclast is attaching. To ~ investigate whether MT1-MMP is associated to podosomes, we stained the cell simultaneously for actin by addition of 10 mg/ml fluorescein-labelled phalloidin (Sigma, Saint Louis, MO) during the incubation with the secondary antibody.

CA 02261~67 1999-01-27 W098104287 rcT~Pg7/04110 Actin staining which is widely used to identify podosomes revealed the same ring of bright dots as shown with the anti MT1-MMP antibody. Therefore MT1-MMP appears to be localised on the podosomes. MT1-MMP staining was however somewhat 5 more diffuse as compared to the sharp actin staining, probably because the sharp actin dots are due to bundles of actin filaments in the core of the podosome and oriented perpendicularly to the attachment surface, while MT1-MMP
might be on the surface of the podosome. As expected, 10 staining for actin illuminated also the extremities of the lamellopodia, as did the anti-MT1-MMP antibody. Similar localisations of MT-MMP were found when the osteoclast was cultured on bone slices. Thus these observations do not only demonstrate the presence on the protein level of MT1-1. MMP in the plasma membrane of the osteoclast, but providenew information on where exactly on the plasma membrane MT1-MMP is localised, i.e. at the level of lamellopodia and of podosomes.

20 Example 3 3-1 Production, purification and activation of osteoclast proteinases.

As noted in the summary of the invention, the production of osteoclast proteinases can be performed in cultures of osteoclasts or in cell lines transfected with cDNA encoding the osteoclast proteinase or a part thereof.
In all cases a purification of the product is needed and in 30 those cases where the production leads to a latent pro-form of the proteinase a subsequent activation is also needed for some purposes. Exemplifying this process, the production, purification and activation of osteoclastic pro-MMP-9 was performed according tO the following descriptions:
, CA 02261~67 1999-01-27 WO 98/04287 P~ 57/04110 a. Osteoclast production of pro-MMP-9 When cultured at 37~C and 5% CO , under serum-free conditions to avoid contamination with serum-derived protelnases and natural inhibitors of proteinases, rabbit 5 osteoclasts secreted 92 kDa pro-MMP-9 into the culture medium. According to studies by gelatinase-zymography, addition of 40 nM of phorbol 12-myristate 13-acetate (PMA) to the cell culture increased the yield of pro-MMP-9 at least 3-fold.

b. Purification of osteoclastic pro-MMP-9 The osteoclast conditioned medium was concentrated by 10 kDa cut-off filtration (Amicon) and subsequently diluted in 2.5 mM sodium phosphate containing 0.04% Triton X-100 15 before application to an affinity column comprislng hydroxyapatite (Bio-Rad, Hercules, CA). By this novel method for purification of MMes, pro-gelatinases including pro-MMP-9 and pro-MMP-2 were observed to bind efficiently to the hydroxyapatite column. However, pro-MMP-9 was eluted 20 from the column already by increasing the phosphate concentration to 5-10 mM, whereas higher concentrations (above 20 mM) of phosphate were needed to elute other pro-gelatinases and gelatinases from the column.

25 c. Activation of osteoclastlc pro-MMP-9 The purified latent pro-MMP-9 was activated either by a traditional method based on incubation with 1 mM (4-amino-phenyl~mercuric acetate (APMA) for 2-8 hrs at 37~C or by a method based on the activation of gelatinases as it is 30 observed during analytical zymography. In the latter method ~he purified pro-MMP-9 was run into a slab gel by preparative SDS-PAGE. The SDS was substituted by Triton X-100 during subsequent incubation of the gel for 16 hrs in a buffer containing 50 mM Tris-HC1, pH 7.5, 5 mM CaCl2, 1 ~M
3~ ZnCl~, and 1% Triton X-100. A part of the gel corresponding to an electrophoretic migration distance of compounds with an approximate ~olecu~ar weight of 92+5 kDa (but including the by now activated approximate 68 kDa form of MMP-9) was CA 02261~67 1999-01-27 excised. The active MMP-9 was electrophoretically eluted from the excised gel.

3-2 Expression and characterisation of MT1-MMP fuslon proteins The MT1-MMP cDNA fragment encoding amino acid residues Gln40-Glus31, Ecl (containing the propeptide, catalytic, hinge and hemopexin, but not the signal peptide, transmembrane and lo cytoplasmic domains of rabbit osteoclast MT1-MMP, see Fig.
3), was PCR amplified using a 5 primer with an extra SnaBI
site and a 3 primer with an extra NotI site. This fragment was inserted between the SmaI and NotI sites of the pGEX-6P-2 vector (Pharmacia). The MT1-MMP cDNA fragments encoding 15 amino acid residues Gln40-Asn322 Ec2 (containing the propeptide, catalytic, and hinge, but not the signal peptide, hemopexin, transmembrane and cytoplasmic domains of rabbit osteoclast MT1-MMP, see Fig. 3) and Gln40-Leu282, Ec3 (containing the propeptide and catalytic but not the signal 20 peptide, hinge, hemopexin, transmembrane and cytoplasmic domains of rabbit osteoclast MT1-MMP, see Fig. 3) were PCR
amplified using 5 primers with an extra BamHI site and 3 primers with an extra XhoI site. These fragments were inserted between the BamHI and XhoI sites of pGEX-6P-2 25 vector (Pharmacia). The three corresponding constructs were used to express glutathione S-transferase (GST) fusion proteins in E.coli BL21 (Pharmacia).
Four overnight cultures of E.coli BL21 transformed with the three PGEX-MT1-MMP expression vectors and the PGEX
30 vector alone (without any insert), were diluted 1:100 in 500 ml 2X YTA medium (Pharmacia). The cultures were grown at 37~C to an OD600= 1.0 before adding isopropyl-~-D-thiogalactopyranoside (IPTG) to a final concentration of 0.1 mM .o induce expression. After induction for 3.5 hours at 3 30~C, the cells were pelleted and resuspended in 25 ml of ice-cold lX PBS. All subsequent steps were carried out at 4~C or on ice. E.coli cells were lysed by sonication (5 bursts of 10 seconds/burst). Cellular debris was pelleted T

CA 02261~67 1999-01-27 W098/04287 PCT~P97/04110 by centrifugation at 3000 rpm after incubation with l~
Triton X-lO0 for 30 minutes.
The purifications were carried out by affinity cnromatography using Glutathione Sepharose 4B contained in 5 the GST Purifications Modules, according to the manufacturer's instructions (Pharmacia). The supernatants obtained after the centrifugation of the sonicated samples were absorbed on l ml of the 50~ slurry of Glutathione Sepharose 4B equilibrated with PBS by incubation at room temperature for 30 minutes. After washing several times with lX PBS, the fusion proteins were eluted with 900 ul of Glutathione Elution buffer (lO mM reduced glutathione in 50 mM Tris-HCl, pH 8.0). The eluates were stored at -20~C until use.
The three fusion proteins migrated in SDS-PAGE as proteins of approx. 85, 60 and 55 kDa corresponding to their cDNA-deduced-sizes of 87, 61 and 57 kDa, respectively. The fusion proteins were confirmed to be GST-MTl-MMP fusion proteins by Western-blotting using an anti-GST antibody ~o reacting with all three proteins and an antibody to the hemopexin domain of MTl-MMP reacting with the large but not the two smaller proteins. Einally, amlno acid sequencing of ~helr propeptide domains further demonstrated that these proteins were truncated forms of MTl-MMP.
3-3 Proteolytic activity of GST-MT1-MMP fusion proteins after activation by trypsin or plasmin In order to obtain truncated MTl-MMP in active form, 30 Ecl, Ec2 and Ec3 were incubated with trypsin or plasmin leading to removal of the GST-part and the propeptide domain of the fusion proteins.

a. Trypsin activation 3C, Eighty ul (20 ug approximately) of the eluted Ecl, ~c2, E-3 and the GST tag alone were incubated at 25~C with 5 u~/ml t~ypsin (Promega) for 15-60 min in a final volume of lO0 ul.
The reactions were stopped by the addition of 50 ug/ml SBTI.

CA 02261~67 1999-01-27 W098/04287 PCT~97104110 b . Plasmin activa tion Twenty-five ul (7 ug approximately) of the eluted Ecl, Ec2, Ec3 and the GST tag alone, were incubated with 2.7 pmol 5 of human plasmin (Boehringer) at 25~C for 30 minutes in a final volume of 45 ul. The reactions were stopped by the .addition of 10 ,uM aprotinin.

c . En zyma t i c a ssay o The proteolytic activities were evaluated by fluorescence measurements (excitation wavelength: 320 nm, emission wavelength: 387 nm) of the hydrolysis of the quenched fluorescent peptide substrate Mca-PLGL-Dpa~AR-NH2 (Bachem) (SEQ ID No.14) after incubation at 37~C for 180 5 minutes in 150 mM NaCl, 10 mM CaCl2, 0.05 % (v~v) Brij-35 in 50 mM Tris-HCl, pH 7.5 (see Table 1).

d. Effect of inhibi tors of MMPs Samples treated either with trypsin or plasmin in the 20 conditions described above were preincubated for 30 minutes at 37~C in the absence or presence of the endogenous MMP
inhibitors TIMP-1 (16.7 ~g/ml) or TIMP-2 (16.7 ~g/ml) or the synthetic MMP-inhibitor BB-94 (0.8xlO-s M, British Biotech).
The hydrolysis of the fluorescent substrate was evaluated 25 afterwards as described above (See Table 1).

CA 0226l567 l999-0l-27 W 098/04287 PCT~P97/04110 Table 1. Hydrolysis in relative fluorescence units (RFUs) per 180 min of a synthetic substrate in the presence or absence of MMP-inhibitors by truncated forms of recomblnant osteoclast MT1-MMP
activated by trypsin or plasmin.

RFU/ Trypsin act~vated Plasmin actlvated 180 min - + TIMP-1 + ~IMP-2 + ~B9q - + B~94 Inhibitor Inhibltor Ecl 139.9 ND ND 6.7 27.5 2.9 Ec2 172.4 148 7.1 6.0 109 3.2 Ec3 9.6 ND ND 6.9 4.1 3.8 pGEX 8.6 ~D ND 6 . 8 3.8 3.5 ND: not done lC 3-4 The cloning, recombinant expression, activation and characterisation of rabbit osteoclast MMP-12.

Due to the expression and use in cell invasion of MMP-12 in macrophages as well as the common hematopoieitic stem î5 cell origin of osteoclasts and macrophages, we investigated whether MMP-12 was also expressed in osteoclasts. As indicated in Example lb and shown in the present example, this was indeed the case, and we therefore expect that MMP-12 plays a similar role in osteoclast invasion and migration 20 as it does in macrophages.
The isolation and sequencing of MMP-12 cDNA from the rabbit osteoclast cDNA library, and the subsequent steps of expression, characterisation and recombinant production of the MMP-12 fusion protein was done essentially as described 25 for MT1-MMP cDNA (see Examples 1, 3-2 and 3-3). Briefly, the osteoclast preparations were obtained from rabbit long bones and the reverse transcribed mRNA from these osteoclasts was amplified by PCR using degenerate prlmers based on regionS
conserved in the MMP family (see Example lb). Among several 30 PCR fragments of the predicted sizes, one (B4) presented CA 02261~67 1999-01-27 W098/04287 rCT~P97/04110 34 homology with a sequence of human MMP-12. When a random-primed 32P-labelled probe based on the PCR product of clone B4 was used to screen a cDNA library of rabbit osteoclasts several positive clones were identified. One of these contained a cDNA insert of 1,792 bp including an open reading frame encoding a polypeptide of 464 amino acids sharing 74 ~, 66 % and 65 % identity to human, rat and mouse MMP-12, respectively (see Figures 4 and 5). Based on this and further comparisons to other available protein 10 sequences, we concluded that the isolated novel cDNA encoded the rabbit homologue of MMP-12 or of a closely related but previously unreported human MMP. The nucleotide sequence analysis of the B4 PCR fragment and rabbit MMP-12 cDNA
clones from the cDNA library was done as described for MT1-1_ MMP (see Example ld). Using this cDNA as a probe fornorthern blotting, we compared the levels of expression of MMP-12 in various cells and tissues from rabbits, including calvaria, brain, placenta, lung, liver, spleen, kidney, bone stromal cells, alveolar macrophages, and purified 20 osteoclasts. Interestingly, the level of expression in purified osteoclasts was as high as in macrophages, while expression was a~most not detectable in the other cells and tissues. To investigate whether MMP-12 is also expressed in osteoclasts in vivo, we performed in situ hybridisations on 25 sections of metacarpals of new-born rabbits, and clearly identified MMP-12 in typical osteoclasts.
For expression and characterisation of a MMP-12 fusion protein, rabbit MMP-12 cDNA containing the open reading frame (bp 58-1437, see Figure 4) was amplified by PCR using 30 primers sense 5 -CGGGATCCCTGTGGGTCACTTCTTCT-3 (SEQ ID
No.22) and antisense 5 -CCGCTCGAGCTGGCACCATTACTAGC-3 (SEQ
ID No.23). The cDNA fragment was inserted into the BamHI
and XhoI sltes of the pGEX-6P-2 vector as described for MT1-MMP. The cDNA was shown by direct sequence analysis to lie 3- just 5' to the GST-encoding moiety of the vector and in proper reading frame with the plasmid translation initiation site (Figure 6).

CA 02261~67 1999-01-27 W098/04287 PCT~P97/04110 E.coli strain BL-21, transformed with pGEX-6P-2 alone (ccntrol vector) and pGEX-6P-2/MMP-12, were plated on Luria 5 Broth (LB) agar plates with 50 ug/ml ampicillin at 37~C
overnight. Single colonies were grown overnight in 50 ml of LB containing 50 ,ug/ml ampicillin in a shaking incubator at 30~C. Subsequently, the overnight cultures were diluted 1:100 in 400 ml of LB containing 50 ug/ml ampicillin and grown at lo 30~C to an OD600 =0.6-1Ø IPTG (Sigma) was added to a final concentration of 0.1 mM to induce production of fusion protein, and cells were maintained in culture for an additional 3 h.
Cell pellets were resuspended in 20 ml of a Tris-HC1 15 bufter (2 mM CaC12 in 25 mM Tris-HCl, pH 7.6) containing 2 mg/ml of lysozyme and then lysed by sonlcatlon for 1 min in ice (6 bursts of 8 sec/burst). After sonication, 1 ml of 20~ Triton X-100 was added and extractiGn continued for 30 minutes at 4~C. After centrifugation for 10 min at 20,000 x 20 g, the fusion protein according to SDS-PAGE was localised in the pellet (estimated molecular weight approx. 75 kDa co-responding well to the cDNA-deduced size of 83 kDa).
The pellet was solubilized in 20 ml of buffer ccn ainlng 8 M urea and then stirred for 1 h at 4~C. The sample was clarified by centrifugation at qO,000 x g for 30 minutes at 4~C. Subsequently, the urea was removed completely by stepwise dialysis of the supernatant against the Tris-HCl buffer. The supernatant was subjected to SDS-PAGE and proteins stained by Coomassle Brilliant Blue R250.
3~ Fusion protein expression was confirmed by Western blot using an antibody against the GST molety. The presence of recombinant rabbit MMP-12 protein was ensured by frag-mentation and subsequent amino acid sequence analysis. The eiastolytic activity of the truncated recombinant MMP-12 was can lrmed by elastin and gelatine zymography.

CA 02261~67 1999-01-27 W098/04287 PCTrEP97/04110 Example 4 5 Assessment of the role of osteoclast MMPs in osteoclast migration.

In bone tissue cultures, we previously showed that MMPs are very important for the recruitment of osteoclasts lO to future resorption sites (Blavier and Delaissé, 1995), but until now osteoclast purification techniques did not allow the demonstration of whether these MMPs were from osteoclasts or other cells. We therefore developed an experimental model in order to address the latter question.
Briefly, we seeded purified or non-purified osteoclasts on membranes (12 ~m pore size) coated with type I collagen, and followed their migration to the lower surface of the membranes after an overnight culture in the absence or presence of MMP inhibitors. We found that not only when 20 using non-purified osteoclast preparations, but also when using purified preparations, osteoclasts could extend cell processes into the pores of the membranes and spread over the lower surface of the membranes. This migration process was inhibited by MMP inhibitors of both the synthetic 25 pseudo-substrate type (RP59794 and BB94) and the natural type (TIMP-2) (Figure 7). This indicates that osteoclasts themselves can overcome a collagen barrier by migrating through it via an MMP dependent pathway, without the participation of other cells.
In order to evaluate how important MMPs are for this migration as compared to other proteinases, we also tested inhibitors of other classes of proteinases on this migration. Cysteine proteinase inhibitors that are potent inhibitors of the degradation of bone matrix in the ~5 subosteoclastic resorption zone, affected only slightly the migrations, whereas a serine proteinase inhibitor was without any effect (Figure 7). Thus MMPs play a unique role in osteoclast migration as compared to other proteinases.

CA 02261~67 1999-01-27 W O 98/04287 PCT~EP97/04110 In order to confirm the role of MMPs in an overall migration/resorption sequence, we seeded purified osteoclasts on dentine slices that were coated or not with type I collagen, cultured them overnight in the presence 5 and absence of MMP inhibitor and followed the formation of pits in the dentine slices. We found that the MMP inhibitor inhibited pit formation only in the collagen coated dentine slices (Figure 8). This indicates clearly that the role of MMPs is on the migration of the osteoclasts to their future lo resorption site, and not on resorption itself.

Example 5 Preparation, characterisation and application of antibodies 15 to MMPs Two approaches were used for the production of anti-MMP
antibodies. In one approach, intact or truncated, native or recombinant MMP was used as an immunogen (see a, below) and 20 in the other approach synthetic peptide mimicking a specific MMP-region was used as an immunogen after having been conjugated to a larger carrier protein ~see b-d, below):

a. Preparation and use of intact or truncated MMP
25 immunogens As an example of the first approach, pro-MMP-9 purified from osteoclast cultures as described in Example 3-1 was used for immunisation either in its latent form or after activation by APMA or by in-gel treatment with SDS/Triton X-30 100. The preparations of pro-MMP-9 and MMP-9 were injected intra-peritoneally every third week in female BALB/c-CF1 murine hybrids. A final booster immunisation of the protein without adjuvant was given 3 days prior to splenectomy. The spleen cells were fused with P3-X-63-Ag8.653 myeloma cells 35 in the presence of 50% polyethylene glycol 4000 and the resulting hybridoma cells propagated and cloned according to CA 02261~67 1999-01-27 W 098/04287 PCT~EP97/04110 standard procedures. Monoclonal antibody was purified from the condi'ioned medium of hybridoma cultures by using proteln A affinity chromatography.

S b. Preparation of ~MP-mimicking conjugated peptide immunogens Based on the amino acid sequence of osteoclastic MT1-MMP (Figures 1 and 2) and sequences available for other members of the MMP family, such as MMP-9 and MMP-12, lo femtameric sequences (i.e. polypeptide sequences of 15 amino acids) were selected due to:

1. their specificity for one member of the MMP family when compared to other members;
2. their putative properties as immunogens according to computer-based algorithms used for analyses of their hydrophilicity, their position and their expected secondary structure in the intact MM~; and 3. their conservation i.e. their possible sequence identity or similarity in corresponding regions of the same MMP in the human, rabbit and mouse species.

Corresponding to the selected femtameric sequences, femta-peptides were synthesised by using Fmoc-amino-acids-O-25 pentafluorophenyl-esters in the presence of catalytic amounts of 3,4-dihydro-4-oxo,1,2,3-benzotriazin-3-yl in a fully automated custom made peptide synthesiser.
The femta-peptides were coupled to a proteinaceous carrier molecular (thyroglobulin). Briefly, thyroglobulin 30 and glutaric anhydride (1:2 w/w) were incubated for 2 hrs at 20~C in 0.1 M sodium borate, pH 9.0 and subsequently desalted on a Nap 10/Sephadex G-25 column (Pharmacia) and dried by vacuum centrifugation. The carrier was resolubilized in 0.01 M sodium phosphate, pH 5.0 and incubated for 3 min at 35 20~C with equal volumes of 5 mg/ml freshly prepared l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (CDI). The CDI-activated thyroglobulin was incubated 4 hrs at 20~C in equal volumes and amounts (w/w) with the femta-peptide in 0.2 M

CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 sodium phosphate, pH 9Ø The thyroglobulin/CDI/femta-peptlde conjugates were dialysed and their protein content 5 determined.

c. Production of polyclonal antibodies by use of conjugated peptide immunogens The thyroglobulin/CDI/femta-peptide conjugates were lo mixed with Freunds incomplete adjutant and injected intramuscularly once per month in female New Zealand White rabbits. Blood was collected and the immunoglobulin fraction purified from the correspondlng serum by ammonium sulphate precipitation.
d. Production of monoclonal antibodies by use of conjugated peptide immunogens The thyroglobulin/CDI/femta-peptide conjugates were mixed with Freunds incomplete adjuvant and injected 20 intraperitoneally every third week ln female BALB/c-CF1 murine hybrids. A final booster immunisation of the conjugate without adjuvant was given 3 days prior to splenectomy. The spleen cells were fused with P3-X-63-Ag8.653 myeloma cells in the presence of 50% polyethylene 25 glycol 4000 and the resulting hybridoma cells propagated and -loned according to standard procedures. Monoclonal antibody was purified from the conditioned medium of hybridoma cultures by using protein A affinity chromatography.
e. Characterisation and application of specific anti-MMP
antibodies The antisera and monoclonal antibodies were selected and initially characterised by enzyme-linked immunosorbent 3~ assay (ELISA) based on 96-well polystyrene plates coated ~ith either purified intact or truncated MMes or homologous or heterologous conjugated femta-peptides. As indicated above, antisera and monoclonal antibodies showing MMP-CA 02261~67 1999-01-27 WO 98/04287 P~ 57lo4llo specificity according to the initial characterisation by ELISA have several applications. One example is their use in immunohistochemical identification of MMP-expression on the protein level by incubation of an anti-MMP antibody with 5 bone cells or tissues. As described in Example 2d, the binding of a monoclonal antibody raised by immunisation with a MT1-MMP mimicking peptide to the actin-rich membranous areas of an osteoclast shows that MMP-antibodies not only are tools of central importance to the identification of the 0 cells which produce a particular MMP, but also can demonstrate the cellular localisation of a MMP and thereby aid in the clarification of its biological role.
Sera from mice immunised with the thyroglobulin-conjugated femta-peptide RSGAPVDQMFPGVPL (SEQ ID No.13) 5 corresponding to a region in the hemopexin domain of rabbit MMP-9 and either boosted with the same conjugated peptide or with purified native osteoclast proMMP-9 showed inhibitory effects to activated MMP-9. The analysis was done by a fluorometric enzymatic assay based on pre-incubation of 20 diluted sera with MMP-9 for 30 min at 37~ before incubation with the synthetic peptide-like substrate Mca-PLGL-Dpa-AR-NH2 (Bachem) (SEQ ID No.14) for 30 minutes at 37~ in 150 mM NaCl, 10 mM CaCl2, 0.05 % (v/v) Brij-35 in 50 mM Tris-HCl, pH 7.5 (see Figure 9).

Example 6 Production of non-immunoglobulin inhibitors of osteoclast proteinases.
Production of non-immunoglobulin inhibitors of osteoclast proteinase aimed at two main type of agents, one being peptide or peptide-mimicking proteinase inhibitors another being antisense probes specifically binding to 3. osteoclast proteinase mRNA. The peptide and peptide mimlcking agents were produced by two methods: a technology based on PEGA bead peptide substrate and inhibitor libraries CA 02261~67 1999-01-27 W O 98/04287 PCT~P97/04110 (see a-c, below), the other being based on positional combinatorial peptide inhibitor libraries (see d-e, below).
The design and use of antisense probes is described in f (see below):

a. Identification of Ml~P substrates by PEGA bead libraries According to previous descriptions (Meldal et al, 199~1~), two PEGA bead peptide substrate libraries were lo generated consisting each of approx. 106 different beads.
Each bead contained many copies of a single sequence: NXl-X2-Y(NO2)-X3-X4-X5-X6-X7-X8-K(Abz~C-PEGA (PEGA bead substrate library A) or ~Xl-X2-Y(NO2)-X3-X4-X5-X6-K~Abz)C-PEGA (PEGA
bead substrate library B), where Xl to X8 are amino acids varying randomly from bead to bead, and Y(NO2) and K(Abz) is a quenching 3-nitrotyrosine and a fluorogenic lysine(2-aminobenzoic acid), respectively. The libraries were incubated at 37~C with purified and activated osteoclast proMMP-9 (approx. 0.1 ~M) and fluorogenic beads subsequently 20 isolated by a micropipette under fluorescence microscopy.
The isolated beads were analysed by an amino acid sequencer.
The incubation of the randomised PEGA-bead substrate libraries lead to identification of 15 clearly fluorescent beads, indicating a specific cleavage of their corresponding 25 peptide in contrast to the millions of other structures in the libraries. The amino acid sequences of the cleaved substrates showed some consistency (see Table 2). In particular a proline at the third position (P3) towards the N-terminal from the cleavage site was highly conserved.

Table 2 Amino acid sequences and cleavage site of quenched fluorogenic peptide substrates identified on PEGA bead libraries (A) and (B) Cleavage sit~

P7 P6 P5 P4 P3 P2 P1 Pl' P2' P3' P4' P5' Bead A2 S K Y' P J A L F F K' A3 S R Y' ? P J G L? T K' A5 W G Y' E A J G F T K' B1A R Y' P ~ K V K' B2N J Y' P J J Y K' B3Y I Y' P J M L K' B5R P Y' P Y ? K K' B6L K Y' P K ? L K' B7F A Y' J M R ? K' B8P A Y' M K K M K' B9P L Y' M S ? J K' B10P V Y' M R G ~ K' B11V R Y' L H G J K' lv b. Synthesis and characterisation of soluble peptides analogous to peptides identified by the PEGA bead substrate library technology To further evaluate the results observed for peptide substrates bound to PEGA beads, a series of soluble peptide 1, substrates was synthesised by multiple column peptide synthesis (Meldal et al, 19941l) The amino acid sequences of ~hese pu.a~ive soluble substrates were based on either single pep.ide substrate sequences or consensus secuences from the ?E~A bead studies The hydrolysis by MMP-3 and ~ o~her MMDs of the soluble peptides was analysed by a s arda~~ _uorometric assay (excitation 320 nm, emission CA 02261~67 1999-01-27 W 098/04287 PCT~EP97/04110 43 425 nm).
As an example, one of the fluorescent beads (A2 in able 2) isolated from PEGA bead peptide substrate library ,A) contained two similar peptides with the sequences S-K-5 Y(NO~)-P-J-A-L-F-F-K(Abz)-PEGA (SEQ ID No.2) and L-F-F-K(Abz)-PEGA (SEQ ID No.24) indicating hydrolysis by osteoclastic MMP-9 of the novel peptide-mimicking substrate S-K-Y(NO2)-P-J-A-L-F-F-K(Abz) (SEQ ID No.2) at the P1-P1' position: A-L. Based on this information several soluble 0 quenched fluorogenic peptides were synthesised (e.g., CL-l and CL-6, see Table 3 and Figures 10 and 11). By a similar strategy for the other amino acid sequences of substrates identified in the eEGA bead substrate libraries A and B, the first 30 soluble quenched fluorogenic peptide substrate i~ candidates for MMP-9 (named CL-1 to CL-30) were synthesised cy multiple column peptide synthesls. Their individual kinetic properties (kCat and Km) were determined by incubation at 37~C with MMP-9 and recombinant truncated MT1-MMP of osteoclast origin, and as controls recombinant truncated 20 MMP-1 and -3; the osteoclast cysteine proteinase, cathepsin K; and the broad-reacting proteinase, subtilisin. Several of the hitherto produced 30 synthetic substrates showed a high selectivity for one or more MMPs; no or very low reactivity with cathepsin K; and kCat/Km ratios up to 50-fold 2[ higher for MMP-9 than for subtilisin. This was particularly clear for the peptide substrates CL-21, CL-25 and CL-29 (see ~able 4). Further peptide substrate designing based on the sequence information obtained from both those of the 30 peptides which were cleaved specifically by MMP-9 and those 3c that were not, can be expected to lead to other even more selective synthetic MMP substrates.
For some of the 30 soluble putative peptide substrates, _he kinetic behaviour was different from what was expected according to the hydrolysis of the corresponding peptide 3 ;mmobilised on the PEGA bead. E.g., the putative substrate, _L-1, was inhibitory to MMP-9 as would have been expected ~or a pseudo-substrate, i.e. with a low Km(3.4 ~M) and a low k.dt/Km (250 M-ls~1) (see Table 3).

,, ~ .

W098/04287 PCTnEP97/04110 ~ ~ooooo ~ ~ * ~ ~ * ~
m Q ~ ~ ~ ~ ~
~ ~ .
V ~~U~
r~ h h Q
V
U~ ~
Q r ~ Q ~d ~ ~ ~ o O O
,~ ~ ~,~ O ", ~ v Q h V
~ V~ 3 V
a) a~

O Q
Q U :~ O
~ cn 4 V ~ _ ~ c ~
a). ,, ~ u a.) h _ ~
a, o Y o v ~ ~ r ~ ~ ~ ~
r r r ~ ~ E
.,, v s O _ ~ ~ ~ _ v ~n C r Z
S ~ (v ~ I ~
U c) J ~z O I,Q¢ ~ ~ Q
O I Z O

h ~ ~, I ~ I ~ ~ _ ~v ~ ~ ~ z~ I ~ ~ ~a ~

,v ~v c~ ) J
1~ 1 11 _ O U .--1 O ~ I ~ ~ II O r 3 (~ N NN ~ S (r Z Z Z Z Z
~ ~ ~ ~ ~ ~ u c O ~ a a ~ ~
c o c~ z ~. C, 1 ,~ 3 C ~.1 U7 U~
r,J ~ U
u~ o T

W O 98/04287 PCT~Er97/04110 ~ u~ ~
v ~ 0\oo\~ o\~ s ~, "~ a. ~ ~ ~ v v ~ ~ I ~ o c~ ~ ~ o -- n, ~ , u~
~: o o o v v 0~00~00~O
v,_, v In ~' ~ S --I
Q I ~ ~ ~ ~
u~ s ~ ~ o ~ ~ O o ~~, a.
0~o0~00 o u v v ~ #
m Q~ ~ ~ o o o u) a.~ v ~ v ~ - O ~o~o - ~
u) ~ ~ 1 3 Q ~, s o ~ ,- v ~r ~
u~ ~ ~ o , ~ OO

Q (J _l 0,0 0~0 O\o O ~ r ~n N~D ~ C
~ ~ -~ V E
~ ~ v -~ ~ C -I
~ ~ ~ ~ O O
u) O Q O ~ v ~ a~
~n O V u~ o ~ ~ C I SV
a ~
~n ~ O\o oP O\o ~ ~ O
o o o u ~ E
c ~ a ~ ~ ~ ~ u 3 o v v ~ ~ ~
O ~ ~ u. o o c v v ~ n ,1 # oa) JJ
C 7~ ~ ~ ~ U
u) 1 c ~ m m ~
a) ~ ~ a c~ N N ~ n~ C
v o m m m m 3 SV ~-C n u) ~.
~~ V
C ~ ~ I Q ~ ~ ~
c ~ a 1~ v ~
a ~ s _, _ v ~
~ N \ a~ c n V N N N ~L ~ '~
C ~ ~, ~
{5 tll ~a Q Q O O ~J
." O

CA 02261~67 1999-01-27 W 098/04287 PCT~P97/04110 c. Identification of MMP inhibitors by PEGA bead libraries According to previous descriptions (Meldal et al, 19941', Meldal & Svendsen, 1995l1, Meldal et al, 1997 ), a PE~A bead peptide inhibitor library (I) was generated consisting of 5 approx. 106 different beads, each containing many copies of a single well-defined substrate sequence as well as many copies of a randomly generated putative inhibitor sequence: NXl-X2-X3-D-X4-X5-X6-VC-PEGA, where X1 to X6 are L-amino acids varying randomly from bead to bead, and D is a ~-amino acid varying 0 randomly from bead to bead. The library was incubated at 37~C
with active MMP-9 and beads remaining quenched (i.e. dark compared to the majority of brightly fluorescent beads) were isolated by a micropipette under fluorescence microscopy. The isolated beads were analysed by an amino acid sequencer and 15 since the substrate sequence was not degraded by the Edman degradation due to prior acylation at the N-terminus, the sequences obtained corresponded to potential peptide-mimicking MMP-9 inhibitors.
A novel type of PEGA bead inhibitor library was developed 20 in order to identify peptide substrate mimicking MMP-inhibitors with a phosphinate instead of a peptide bond at the susceptible cleavage site (i.e. between the expected P1 and P1' sites of the corresponding substrate). Two PEGA bead phosphinate inhibitor libraries (IIa and IIb) were generated. Each library 25 consisted of approx. 106 different beads, and each PEGA bead contalned many copies of a single well-defined substrate sequence as well as many copies of a randomly generated putative inhibitor sequence: NX1-X2-JP/CM-X3-X4-"Linker"-EGA (in IIa) or NX1-X2-GP/CL-X3-X4-"Linker"-PEGA (in IIb), where X1 to 30 X4 are L-amino acids varying randomly from bead to bead, and JP/CM and GP/CL is the phosphinate pseudo dipeptide used in library IIa and IIb, respectively (see Figures 12-15). The deslgn of the first two phosphinate pseudo dipeptides was based on the identity of suitable P1 and P1' amino acids in newly 35 developed and existing MMP-9 substrates. Other combinations of pseudo amino aclds around the phosphinate bond will also be CA 02261~67 1999-01-27 investigated according to the findings of MMP selective peptide substrates by use of e.g. PEGA bead substrate libraries.

d. Positional combinatorial peptide inhibitor libraries As an alternative to using the PEGA bead peptide libraries for identification of potential MMP inhibitors, 20 different positional combinatorial peptide inhibitor libraries (Houghten et al, 19917) were produced using pentapeptides constructs X-X-D-X-X, where D is the D-form of one of the 20 common amino lo acids (except glycine) or hydroxyproline, and X is a randomly varying natural L-form of one of the 20 common amino acids or hydroxyproline. The peptide libraries were purified by high performance liquid chromatography in order to remove salts and other substances which were toxic to bone tissue cultures 15 before being tested for inhibitory effects on osteoclast migration and bone resorption in murine foetal metatarsal cultures. Each of the 20 libraries contained 30 ~mol penta-peptides composed of up to 214(199,481) different structures.

20 e. Murine foetal metatarsal cultures for studying osteoclast migration and resorption in vitro 45Ca2+ pre-labelled metatarsals isolated from 17 day old NRMl mouse foetuses were used as an organ culture model (Blavier & Delaissé, 19953). Briefly, foetal bones were 25 labelled by subcutaneous injection of 45Ca2+ into pregnant mice at day 16 of gestation. Foetal metatarsals isolated on the following day thereby comprised 45Ca-labelled calcified matrix developed in uteri between day 16 and 17. In the periosteum surroundlng the calcified matrix numerous osteoclast precursors 30 cells were present. Corresponding to the development of bone and bone marrow in metatarsals in vivo, subsequent cultivation of the isolated metatarsals in BGJb medium containing 30 nM
la,25 dihydroxy-vitamin D3 and 0.1% Albumax for 1 to 7 days resulted in differentiation, fusion and migration of the 35 osteoclast precursor cells leading to the presence of mature osteoclasts in the central calcified matrix where they resorbed bone and formed the primitive marrow cavity. The development CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 and bone resorbing activity of the osteoclasts was estimated by measurement of the release of ~sCa + into the culture medium at various time points and by microscopic inspection of the positioning in the cultured metatarsals of osteoclasts stained 5 for tartrate-resistant acid phosphatase. The general MMP
inhibitor, RP59794 which has been shown previously to inhibit the migration of osteoclasts and thereby reduce the release of 45Ca2 in the metatarsal culture model (Blavier & Delaissé, l9g53) was included as a positive control in all experiments.
lo The effect of the 20 X-X-D-X-X combinatorial libraries on bone resorption was evaluated by measuring the change (%) in accumulated 45Ca-release into the conditioned medium of the treated metatarsal culture relatively to the 45Ca-release of the corresponding non-treated metatarsal culture originating from -~ the other leg of the same foetus at Day 1, 2 and 4. Each library was tested in 4 independent metatarsal cultures in the same experiment and in some cases the experiment was repeated.
Each of the 20 libraries was used in a concentration of 3 mM total peptide corresponding to a concentration of approx. 15 20 nM for each of the 194,481 structures in a library. The majority of the 20 libraries did not significantly affect the bone resorption, whereas 1 of the 20 libraries (D=ile) showed significant reductions in the 45Ca-release at Day 4 (see Figure 16), and most importantly 2 of the 20 libraries (D=leu and 25 D=trp) showed significant inhibitions at both Day 2 and Day 4 (see Figure 16 and Table 5).

Table 5 Change (in %) of 45Ca-release due to the addition of a X-X-D-x-x combinatorial library to 4-day metatarsal cultures Library Day 0-1 Day 0-2 Day 0-4 x-x-trp-x-x o% (ns) -20% (0.02) -40% (0.05) X-X-leu-X-X o% (ns) -34% (0.0001) -48% (0.0005) The P-values express the level of significance of the changes between the treated and corresponding non-treated group (n=4 for each).

CA 02261~67 1999-01-27 WO 98/04287 PCT~EP97/04110 49 Further investigations of the X-X-trp-X-X library was done by performing a second screening of 28 libraries with a selected variation at one of the 4 X-positions. The following conformations were used U-X-trp-X-X, X-U-trp-X-X, X-X-trp-U-X
5 and X-X-trp-X-U, where U is a random mixture of L-amino acids belonging to a specific undergroup: U1: K and R (n=2); U2: H, Y, F and W (n=4); U3: E and Q (n=2); U4: T, D, S and N (n=4);
U5: C, V, L, I and M (n=5); U6: P and J (n=2); and U7: A, G
(n=2). Each of the 28 libraries was used in a concentration of o 1.6-4.0 mM total peptide, corresponding to approx. 85 nM for each of the 18,522 to 46,305 structures in a library. The majority of the 28 libraries did not significantly affect the bone resorption, whereas 5 of the 28 libraries showed significant and/or marginally significant reductions in the 15 45Ca-release at Day 1, 2 and/or 4 (see Table 6).

Table 6. Change ~in %~ of ~sCa-release due to the addition of a U-x-trp-x-x, x-u-trp-x-x, x-x-trp-U-x or x-x-trp-x-u combinatorial 20 library to 4-day metatarsal cultures Library Day 0-1 Day 0-2 Day 0-4 U5-X-w-X-X -20% (0.28) -21% (0.12) -15% (0.11) x-U2-w-x-x -15% (0.35) -23% (0.05) -18% (0.15) X-u5-w-x-X -43% (0.007) -28% (0.04) -11% (0.05) x-x-w-u2-x -23% (0.21) -20% (0.003) -16% (0.001) X-x-w-x-U5 -23% (0.15) -24% (0.0008) -15% (0.07) ~urther investigations of the U5-X-trp-X-X, X-U2/5-trp-X-25 X, X-X-trp-U2-X and X-X-trp-X-U5 libraries was done by performing a third screening of 23 libraries with a single variation at one of the 4 X-positions. The following conformations were used Z5-X-trp-X-X, X-Z2/5-trp-X-X, X-X-trp-Z2-X and X-X-trp-X-Z5, where Z2, Z5 or Z2/5 is a single L-amino 30 acid belonging to undergroup(s) U2, U5 or U2 and U5, respectively. With a few exceptions, each of the 23 libraries was used in concentration of 3.2 mM total peptide, corresponding to approx. 340 nM for each of the 9,261 CA 02261~67 1999-01-27 W098/04287 PCT~P97/04110 structures in a library. More than half of the 23 libraries did not significantly affect the bone resorption, whereas 11 of the 23 libraries showed significant and/or marginally significant reductions in the 45Ca-release at Day 1, 2 and/or 4 5 (see Table 7).

~able 7 Change ~in ~) of 45Ca-release due to the addition of a Z5-X-trp-x-x, x-z2/5-trp-x-x, x-X-trp-z2-x or X-X-trp-X-Z5 combinatorial library to 4-day metatarsal cultures ~o Llbrary (Conc) Day 0-1 Day 0-2 Day 0-4 C-X-trp-X-X (3.2 mM) -22% (0.09) -~8% (0.03) ~34~~ (0.006) V-X-trp-X-X (3.2 mM) - 6~ (0.33) ~30% (0.17) -23% (0.11) L-X-trp-X-X (3.2 mM) -23~o (0.07) -32% (0.01) -20% (0.06) X-W- rp-X-X (3.2 mM) -19~~ (0.22) -26% (0.003) -22% (0.08) X-Y-trp-X-X (3.2 mM) -26-o (0.0~) -27% (0.06) -18% (0.17) X-r-trp-X-X (3.2 mM) -20% (0.19) ~33% (0.06) -23% (0.09) X-C-trp-X-X (3.2 mM) ~39% (0.02) -18% (0.14) -10% (0.13) X-L-trp-X-X (3.2 mM) -19% (0.39) -26~o (0.07) -24~o (0.20) X-X-trp-Y-X (3.2 mM) -25% (0.05) -48~o (0.0003) -38~~ (0.0006) (0.8 mM) ~45% (0.07) -26% (0.14) ~ 8% (0.22) (0.8 mM) ~39% (0.003) ~30~o (0.02) -18% (0.0~) X-X-trp-X-T (3.2 mM) -12% (0.34) ~34% (0.21) -17~~ (0.18) X-X-trp-X-~. (3.2 mM) -26~~ (0.21) -26% (0.14) -18% (0.03) In an early attempt to identify single peptide inhibitory structures a fourth screening was performed on 20 peptides of 15 the structure C/V/L-Y/F/W/C/L-trp-Y-M/L considered to be likely candidates according to the results in the 3rd screening. Each of the 20 single structure peptides was used in a concentration of 13 luM. The majority of the 20 peptides did not significantly affect the bone resorption, whereas 5 of the 20 structures 20 showed signiflcant and/or marginally significant reductions in the ~5Ca-release at Day 1, 2 and/or 4 (see Table 8). Even better single peptide inhibitors will be obtained upon further investigations based on the data from the first 4 screenings.
Particularly further investigations of X-X-trp-Y-X

r CA 0226l~67 l999-0l-27 W098/04287 PCT~P97/04110 combinatorial libraries and a similar screening programme for X-X-leu-X-X seem promising.

5 Table 8. Change (in %) of ~5Ca-release due to the addition of a single peptide structure with the sequence C/V/L-Y/F/W/C/L-trp-Y-M/L to 4-day-metatarsal cultures Structure Day 0-1 Day 0-2 Day 0-4 C-L-w-Y-L -30% (0.02) -22% (0.03) -15% (0.005) C-L-w-Y-M -29~ (0.06) -26% (0.05) -13% (0.40) C-Y-w-Y-L -17% (0.008) -18% (0.009) -12% (0.11) V-Y-w-Y-M -17% (0.21) -21% (0.04) -15% (0.02) L-F-w-Y-L -34% (0.003) -37% (0.007) -26~o (0.04) f. Design and use of antisense probes to MMPs.
Antisense oligonucleotide probes agalnst various MMPs were produced in order to study their influence on bone metabolism 5 and osteoclast biology in bone cell and tissue cultures as well as in animal models. The antisense oligonucleotide probes were designed by choosing sequences which were specific to a particular MMP and showing as little as possible similarity to any predictably relevant mammalian genes. In all cases a sense 2Q probe and/or a so-called scrambled probe was used as negative controls for comparison to the antisense probe. In order to stabilise the probes, some were produced in a partially phosphorthiolated form to protect them against degradation by nucleases (phosphate bonds which are phosphorthioate bonds 25 instead of normal phosphordiesters are marked with a * in the diagram below). In order to make the delivery of the probes to the interior of osteoclasts some of the probes were included in liposomes before application to the cell or tissue cultures.
The strategy in this type of experiments is exemplified by 30 results from design, synthesis and testing of antisense probes to mouse and rabbit MMP-9.

, CA 02261~67 1999-01-27 W 098/04287 52 PCT~E~97/04110 Two sets of probes (17-mers) to murine MMP-9 are shown in the Table below:

Table 9: Selected probes for use in experiments with MMP-9 5 expression in murine cells and tissues:

5'-T*G*GTATGTGGTCTGT*G~T Scrambled (SEQ ID No.29) First set 5'-T*G*TGGTTCAGTTGTG*G*T Antisense (SEQ ID No.30) 5'-A*C*CACAACTGAACCA*C*A Sense (SEQ ID No.31) Second set 5'-G*GAC*T*CA*TGG*TGAG*G*A*C Antisense (SEQ ID No.32) 5'-C*GGA*T*ACAGG*TG*TC~G*G*A Sense (SEQ ID No.33) 15The probes were used in the murine metatarsal system described in Example 6e and in a murine pre-osteoclast culture system. The latter was based on unfractionated bone cells isolated from 12 day old mice and cultured for 7 days in the presence of 5% fettle calf serum in order to eradicate all 20 multinucleated osteoclast leaving only stromal cells and osteoclast precursors. Upon subsequent culture of approxi-mately 10 days in the presence of 2 ~g/ml PGE2, new mature osteoclasts were formed. The continuous differentiation of pre-osteoclasts to mature osteoclasts in this culture system 25 correlated well to production of pro-MMP-9 according to gelatinase zymographical studies of the corresponding conditioned medium. For both test systems, the probe was added to the culture medium in a concentration varying between 1 and 10 ~g/ml and the medium was renewed every day.
Seven antisense probes (14- to 18-mers) to rabbit MMP-9 were constructed as shown in the Table below:

3c CA 02261~67 1999-01-27 Table 10: Selected probes for use in experiments with MMP-9 expresslon in rabbit cells and tissues:

Probe 1 G*T*C*TGG*GGC*T*CA*TGG*T*G*A
(start codon) (SEQ ID No.34) Probe 2 G*G*CT*CA*TGG*TGA*G*G
(start codon) (SEQ ID No.35) ~robe 3 G*G*GC*T*CA*T&G*TG*AGG*G*G*A
(start codon) (SEQ ID NO.36) Probe 4 C*T*CA*TGG*TG*AGG*GGA*G*C*A
(start codon) (SEQ ID No.37) Probe 5 A*T*GG*TG*AGG*GGAG*CA*G*C*G
(start codon) (SEQ ID No.38) Probe 6 A*G*GT*GAG*TGG*CGT*CA*C*C*G
(stem loop) (SEQ ID No.39) Probe 7 G*C*TGT*CA*AAG*T*TGGA*A*G*T
(stem loop) ~SEQ ID No.40) Scrambled 1 G*G*CC*T*C*TAC*CG*CAACT*G*C
(SEQ ID No.41) Scrambled 2 G*G*C*C*T*C*TAGG*GGAAC*T*G*C
(SEQ ID No.42) Five of the antisense probes spanned the start codon of the mRNA and two targeted single stranded loops (identified by mRNA secondary structure prediction algorithms) within the translated region.
lo Testing of the effects of the antisense and scrambled probes to rabbit MMP-9 was performed in osteoclasts isolated from long bones of 8 to 10 days old rabbits. The osteoclasts were cultured on bovine bone slices in 5 % foetal calf serum, with renewal of media and oligonucleotides every day. The 15 results were evaluated by quantification of MMP-9 in gelatinase zymography and by studies of osteoclast morphology and numbers as well as quantification of the secretion of tartrate-resistant acid phosphatase into the conditioned medium of the osteoclast cultures by enzymatic assay.

CA 0226l~67 l999-0l-27 W 098/04287 PCT~EP97/04110 References l. Baggio R, Shi Y, Wu Y, Abeles R H. From poor substrates to good inhibitors: Design of inhibitors for serine and thiol proteases. Biochem 35:3351-3353. 1996.
2. Birkedal-Hansen H, Moore W G I, Bodden M K, Windsor L J, Birkedal-Hansen B, DeCarlo A, Engler J A. Matrix metalloproteinases: A review. Crit Rev Oral Biol Med 4:197-250. 1993.
3. Blavier L, Delaissé J-M. Matrix metalloproteinases are obligatory for the migration of preosteoclasts to the developing marrow cavity of primitive long bones. J Cell Sci 108:3649-3659. 1995.
4. Brown P D, Giavazzi R. Matrix metalloproteinase inhibition: a review of anti-tumour activity. Annals of Oncology 6:967-974. 1995.
5. Delaissé J-M, Vaes G. Mechanism of mineral solubilization and matrix degradation in osteoclastic bone resorption.
In: Biology and physiology of the osteoclast (eds. Rifkin B R, Gay C V), pp 289-314. Boca Raton, CRC Press. l9g2.
2s 6. ~oged N T, Delaissé J-M, Hou P, Lou H, Sato T, Winding B, Bonde M. Quantification of the collagenolytic activity of isolated osteoclasts by enzyme-linked immunosorbent assay. J. Bone Miner Res 11:226-237. 1996.
7. Houghten R A, Pinilla C, Blondelle S E, Apell J R, Dooley C T, Cuervo J H. Generation and use of synthetic peptide combinatorial libraries for basic research and drug discovery. Nature 354: 84-86. 1991.
3s 8. Laitala T, Vaananen H K. Inhibition of bone resorption in vitro by antisense RNA and DNA molecules targeted against carbonic anhydrase II or two subunits of vacuolar H+-ATPase. J Clin Invest 93:2311-2318. 1994.

9. Lin M, Hultquist K L, Oh D H, Bauer E A, Hoeffler W K.
Inhibition of collagenase type I expression by psoralen antisense oligonucleotides in dermal fibroblasts. FASEB
9:1371-1377. 1995.
4s 10. Meldal M, Svendsen I, Breddam K, Auzanneau F-I. Portion-mixing peptide libraries of quenched fluorogenic substrates for complete subsite mapping of endoprotease specificity. Proc Natl Acad Sci 91:3314-3318. 1994.
sC

11. ~leldal M, Svendsen I. Direct visualization of enzyme inhibitors using a portion mixing inhibitor library containing a quenched fluorogenic substrate. J Chem Soc eerkin Trans: 1591-1596. 1995.
5s 12. Sakai D, Tong H-S, Minkin C. Osteoclast molecular CA 02261~67 1999-01-27 W098/04287 55 PCT~P97/04110 phenotyping by random cDNA sequencing. Bone 17~ 119.
1995.

13. Sato H, Takino T, Okada Y, Cao J, Shinegawa A, Yamamoto E, Seiki M. A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370: 61-65.
1994.

14. Takino T, Sato H, Yamamoto E, Seiki M. Cloning of a human gene potentially encoding a novel matrix metalloproteinase having a C-terminal transmembrane domain. Gene 155:293-298. 1995.

15. Tezuka K, Sato T, Kamioka H, Nijweide P J, Tanaka K, Matsuo T, Ohta M, Kurihara N, Hakeda Y, Kumegawa M.
Identification of osteopontin in isolated rabbit osteoclasts. Biochem Biophys Res Commun 186:911-917.
1992.
20 16. Tezuka K, Nemoto K, Tezuka Y, Sato T, Ikeda Y Kobori M, Kawashima H, Eguchi H, Hakeda Y, Kumegawa M.
Identification of matrix metalloproteinase 9 in rabbit osteoclasts. J Biol Chem 269:15006-15009. 1994.
25 17. Will H, Hinzmann B. cDNA sequence and mRNA tissue distribution of a novel human matr x metalloproteinase with a potential transmembrane segment. Eur J Biochem 231:602-608. 1995.
30 18. Eggleston Z M, Mutter M. Shaping up to Proteins.
Chemistry in Britain:39-41. May 1996.
19. Galardy R E, Grobelny D, Kortylewicz Z P, Poncz L.
Inhibition of human skin fibroblast collagenase by phosphorous-containing peptides. Matrix Suppl 1: 259-262.
1992.
20. Shapiro S D, Griffin G L, Gilbert D J, Jenkins N A, Copeland N G , Welgus H G, Senior R M, Ley T J.
4Q Molecular cloning, chromosomal localization, and bacterial expression of a murine macrophage metalloelastase. J. Biol. Chem. 267:4664-4671. 1992.
21. Meldal M, Svendsen I, Juliano L, Juliano M A, Del Nery E, Scharfstein J. Inhibition of cruzipain visualized in a fluorescence quenched solid-phase inhibitor library.
D-Amino acid inhibitors for cruzipain, cathepsin B and cathepsin L. J Pept Sci: in press, 1997.

CA 02261~67 1999-01-27 W098/04287 PCT~P97/04110 SEQUENCE LISTING

(1) GENERAL INFORMATION:
(i) APPLICANT:
(A) NAME: Center for Clinical & Basic Research (B) STREET: Ballerup Byvej 222, (C) CITY: Ballerup (E) COUNTRY: Denmark (F) POSTAL CODE (ZIP): DK-2750 (ii) TITLE OF INVENTION: The Use of Proteinase Inhibitors for the Prevention or Reduction of Bone Resorption (iii) NUMBER OF SEQUENCES: 23 (iv) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B~ COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release #1.0, Version #1.30 (EPO) (vi) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: GB 9615976.9 (B) FILING DATE: 30-JUL-1996 (2) INFORMATION FOR SEQ ID NO: 1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGT~: 10 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:5 (D) OTHER INFORMATION:/product= "x is hydroxyproline"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:
Ser Lys Tyr Pro Xaa Ala Leu Phe Phe Lys CA 02261~67 1999-01-27 (2) INFORMATION FOR SEQ ID NO: 2-(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 10 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:3 (D) OTHER INFORMATION:/product= "X is Y(N02)"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:5 (D) OTHER INFORMATION:/product= "X is hydroxyproline"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:10 (D) OTHER INEORMATION:/product= "X is K(Abz)"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:
Ser Lys Xaa Pro Xaa Ala Leu Phe Phe Xaa (2) INFORMATION FOR SEQ ID NO: 3:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 2546 base pairs (B) TYPE: nucleic acid ~C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Oryctolagus cuniculus (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:
TAACGCAGAG TTACATATAC ATACCTGGGG GGGGGGGGGG GGTTCTACTA

CATATAAGAT TCTACTACTT ATTCATGTAG CGATCACTAA TGTAGATTTT

CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 AGATTTTTAT TTATTTCCTA TATACTTTAC TTATTATTTA TTTCTTTGAT

TTAGCACACG CAAACTTACA ACACAGAGTT CTATCCTATC CCTATTAGTT

TTACCTATTA GTTACCTTAT TAGTTACCTA TTAGTTTTAC CTTATTAGTT

GTTTTACCTT ATTAGTTTTA CCTATTAGTT TTAAACTACT AATGTAGCGA

AATGTTAGCC GCTAGGAATC CAAAGTCGGT GCCTCCGGAA GACAAAGGCG

AGATGGCGGC GCGACCCCTA GGCGAGGGCC CCGCCGCGGA ACCGCCCAGC

CCGACGGTCG CGGACCATGT CTCCCGCCCC ACGACCCTCC CGCAGGCTCC

CTGCTCACAC TCGGCACCGC ACTCGCCTCC CTCGGCTCGG CCAAAAGCAA

GCCCCGAAGC CTGGCTGCAG CAGTATGGCT ACCTGCCTCC AGGGAAGACC

CACACAGCGC TCTCCTCAGT CACTGTCAGC TGCCATTGCT AAGCCATGCA

GGTTTGCGAG TGACAGGCAA GGCCGATACA GACACCAAAT GAAGGCCATG

GCTGCGGTGT TCCAGACAAG TTTGGGGCTG AGAAATCAAG GCCAATGTCC

CTACGCCATC CAGGGCCTCA AATGGCAGAA CATAATGAGA TCACTTTCTG

TACACCCCCA AGGTGGGCGA ATATAAAATC TAAATGTTAG GCCACATTCG

CAAGGCATTC CGCGTGTGGG AGAGCGCCAC ACCGAAATCT ACTAAATGTA

CGCTTCCGCG AGGTGCACTA TGCCTACATC CGCGATGGCC GTGAGAAGCA

TCATGATCTT CTTTGCCGAG GGCTTCCATG GCGACAGCAC GCCCTTCAAG

TGGCTTCCTG GCCCACGCCT ACTTCCCGGG CCCCAACATT GGAAAACTCT

CA 0226l~67 l999-0l-27 W O 98/04287 PCT~EP97/04110 AGAATCTACT AAATGTTAGG GGGACACCCA CTTTGACTCC GCGGAGCCCT

GAATGAGGAC CTGAAAACGG GAATGACATC TTCCTGGTGG CTGTGCATGA

GCCCTGGGCA ACTGGAGCAC TCCAATGACC CCTCAGCCAT CATGGCACCG

GG~TGAAGAC ACAGAGAACT TCGTGCTGCC TGATGATGAC CGCCGGGGCA

TAATATGGGA GCCAGTCGGG GTCCCCCACA AAGATGCCTC CTCCACCCAG

CCCGGACTTT TATCCCCGAT AAGCCCAGGA ACCCCACCTA CGGGCCCAAC

CGGGAACTTT GACACTGTGG CCGTGCTCCG AGGAGAGATG TTTGTCTTCA

TGGTTCTGGA GGGTGAGGAA CAACCAAGTG ATGGACGGCT ACCCAAAATG

AGTTCTGGCG GGGCCTGCCT GCTTCCATCA ACACCGCCTA CAAGAGAGGA

ATTCGTCTTC TTCAAAGGAG ATAAGCACTG GGTGTTTAAG ACGAGGCTTC

GGCTACCCCA AGCACATCAA GGAGCTGGGC CGAAACTCTA CTAAATGTTA

ACCGACAAGA TCGATGCCGC TCTCTTCTGG ATGCCCAATG GAAAGAATCT

AGAACCTACT TCTTCCGGGG AAACAAGTAC TACCGATTCA ACGAGGAGCT

TGGACAGCGA GTACCCCAAG AACATCAAAG TGTGGGAAGG CATCCCCGAG

AGGGTCGTTC ATGGGCAGTG ATGAAGTCTT CACTTACTTC TACAAGGGGA

TGGAAATTCA ACAACCAGAA GCTGAAGGTG GAGCCCGGCT ACCCCAAAAG

GGGACTGGAT GGGCTGCCCG GCTGGGGGCC GTCCGGATGA GAAGGGACTG

GGAGGTGATC ATCATCGAGG TGGACGAGGA GGGCAGCAAG GAGCCGTGAG

CA 02261~67 1999-01-27 W098/04287 rCTAEP97/04110 TCTTCAGGCG CCACGGGACT CCGAAGCGAA ACTGCTCTAC TGCCAGCGTT

CAAGGTCTGA CCCCCACCGC TGGCCAACAC CCACTCCCAC CGCAAGGACT

GATTGTATCC AATAAAAAAT AAGCATCAGC AAAA~AAAAA AAPAAAAAAA

CTAAATGTTA GAACTACTAA TGTAGA

(2j INFORMATION FOR SEQ ID NO: 4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 582 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECUL~ TYPE: protein (iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Oryctolagus cuniculus (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:
Met Ser Pro Ala Pro Arg Pro Ser Arg Arg Leu Leu Leu Pro Leu Leu Thr Leu Gly Thr Ala Leu Ala Ser Leu Gly Ser Ala Lys Ser Asn Ser Phe Ser Pro Glu Ala Trp Leu Gln Gln Tyr Gly Tyr Leu Pro Pro Gly Asp Leu Arg Thr His Thr Gln Arg Ser Pro Gln Ser Leu Ser Aia Ala Ile Ala Ala Met Gln Arg Phe Tyr Gly Leu Arg Val Thr Gly Lys Ala I

CA 02261~67 1999-01-27 WO 98/04287 PCT~P97/04110 Asp Thr Asp Thr Met Lys Ala Met Arg Arg Pro Arg Cys Gly Val Pro Asp Lys Phe Gly Ala Glu Ile Lys Ala Asn Val Arg Arg Lys Arg Tyr Ala Ile Gln Gly Leu Lys Trp Gln His Asn Glu Ile Thr Phe Cys Ile llS 120 125 Gln Asn Tyr Thr Pro Lys Val Gly Glu Tyr Ala Thr Phe Glu Ala Ile ~30 135 140 Arg Lys Ala Phe Arg Val Trp Glu Ser Ala Thr Pro Leu Arg Phe Arg 145 . 150 155 ~60 Glu Val His Tyr Ala Tyr Ile Arg Asp Gly Arg Glu Lys Gln Ala Asp Ile Met Ile Phe Phe Ala Glu Gly Phe His Gly Asp Ser Thr Pro Phe Asp Gly Glu Gly Gly Phe Leu Ala His Ala Tyr Phe Pro Gly Pro Asn Ile Gly Gly Asp Thr His Phe Asp Ser Ala Glu Pro Trp Thr Val Arg Asn Glu Asp Leu Asn Gly Asn Asp Ile Phe Leu Val Ala Val Hi s Glu Leu Gly His Ala Leu Gly Leu Glu His Ser Asn Asp Pro Ser Ala Ile CA 02261~67 1999-01-27 W098/04287 PCT~EP97104110 Met Ala Pro Phe Tyr Gln Trp Met Asp Thr &lu Asn Phe Val Leu Pro Asp Asp Asp Arg Arg Gly Ile Gln Gln Leu Tyr Gly Ser Gln Ser Gly Ser Pro Thr Lys Met Pro Pro Pro Pro Arg Thr Thr Ser Arg Thr Phe Ile Pro Asp Lys Pro Arg Asn Pro Thr Tyr Gly Pro Asn Ile Cys Asp Gly Asn Phe Asp Thr Val Ala Val Leu Arg Gly Glu Met Phe Val Phe Lys Glu Arg Trp Phe Trp Arg Val Arg Asn Asn Gln Val Met Asp Gly Tyr Pro Met Pro Ile Gly Gln Phe Trp Arg Gly Leu Pro Ala Ser Ile Asn Thr Ala Tyr Glu Arg Lys Asp Gly Lys Phe Val Phe Phe Lys Gly Asp Lys His Trp Val Phe Asp Glu Ala Ser Leu Glu Pro Gly Tyr Pro Lys His Ile Lys Glu Leu Gly Arg Gly Leu Pro Thr Asp Lys Ile Asp Ala Ala Leu Phe Trp Met Pro Asn Gly Lys Thr Tyr Phe Phe Arg Gly CA 0226l~67 l999-0l-27 W098/04287 PCT~EP97/04110 Asn Lys Tyr Tyr Arg Phe Asn Glu Glu Leu Arg Ala Val Asp Ser Glu Tyr Pro Lys Asn Ile Lys Val Trp Glu Gly Ile Pro Glu Ser Pro Arg Gly Ser Phe Met Gly Ser Asp Glu Val Phe Thr Tyr Phe Tyr Lys Gly Asn Lys Tyr Trp Lys Phe Asn Asn Gln Lys Leu Lys Val Glu Pro Gly Tyr Pro Lys Ser Ala Leu Arg Asp Trp Met Gly Cys Pro Ala Gly Gly Arg Pro Asp Glu Gly Thr Glu Glu Glu Thr Glu Val Ile Ile Ile Glu Val Asp Glu Glu Gly Ser Gly Ala Val Ser Ala Ala Ala Val Val Leu Pro Val Leu Leu Leu Leu Leu Val Leu Ala Val Gly Leu Ala Val Phe Phe Phe Arg Arg His Gly Thr Pro Lys Arg Leu Leu Tyr Cys Gln Arg Ser Leu Leu Asp Lys Val CA 02261~67 1999-01-27 (2j INFORMATION EOR SEQ ID NO: 5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 582 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Homo sapiens (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:
Met Ser Pro Ala Pro Arg Pro Ser Arg Cys Leu Leu Leu Pro Leu Leu Thr Leu Gly Thr Ala Leu Ala Ser Leu Gly Ser Ala Gln Ser Ser Ser Phe Ser Pro Glu Ala Trp Leu Gln Gln Tyr Gly Tyr Leu Pro Pro Gly Asp Leu Arg Thr His Thr Gln Arg Ser Pro Gln Ser Leu Ser Ala Ala Ile Ala Ala Met Gln Lys Phe Tyr Gly Leu Gln Val Thr Gly Lys Ala Asp Ala Asp Thr Met Lys Ala Met Arg Arg Pro Arg Cys Gly Val ?ro Asp Lys Phe Gly Ala Glu Ile Lys Ala Asn Val Arg Arg Lys Arg Tyr CA 02261~67 1999-01-27 W098/04~7 PCT~P97/04110 Ala Ile Gln Gly Leu Lys Trp Gln His Asn Glu Ile Thr Phe Cys Ile Gln Asn Tyr Thr Pro Lys Val Gly Glu Tyr Ala Thr Tyr Glu Ala Ile Arg Lys Ala Phe Arg Val Trp Glu Ser Ala Thr Pro Leu Arg Phe Arg Glu Val Pro Tyr Ala Tyr Ile Arg Glu Gly His Glu Lys Gln Ala Asp Ile Met Ile Phe Phe Ala Glu Gly Phe His Gly Asp Ser Thr Pro Phe Asp Gly Glu Gly Gly Phe Leu Ala His Ala Tyr Phe Pro Gly Pro Asn Ile Gly Gly Asp Thr His Phe Asp Ser Ala Glu Pro Trp Thr Val Arg Asn Glu Asp Leu Asn Gly Asn Asp Ile Phe Leu Val Ala Val His Glu Leu Gly His Ala Leu Gly Leu Glu His Ser Ser Asp Pro Ser Ala Ile Met Ala Pro Phe Tyr Gln Trp Met Asp Thr Glu Asn Phe Val Leu Pro Asp Asp Asp Arg Arg Gly Ile Gln Gln Leu Tyr Gly Gly Glu Ser Gly . , .

CA 02261~67 1999-01-27 W 098/04287 PCT~EP97/04110 Phe Pro Thr Lys Met Pro Pro Gln Pro Arg Thr Thr Ser Arg Pro Ser Val Pro Asp Lys Pro Lys Asn Pro Thr Tyr Gly Pro Asn Ile Cys Asp Gly Asn Phe Asp Thr Val Ala Met Leu Arg Gly Glu Met Phe Val Phe Lys Glu Arg Trp Phe Trp Arg Val Arg Asn Asn Gln Val Met Asp Gly Tyr Pro Met Pro Ile Gly Gln Phe Trp Arg Gly Leu Pro Ala Ser Ile Asn Thr Ala Tyr Glu Arg Lys Asp Gly Lys Phe Val Phe Phe Lys Gly Asp Lys His Trp Val Phe Asp Glu Ala Ser Leu Glu Pro Gly Tyr Pro Lys Hls Ile Lys Glu Leu Gly Arg Gly Leu Pro Thr Asp Lys Ile Asp Ala Ala Leu Phe Trp Met Pro Asn Gly Lys Thr Tyr Phe Phe Arg Gly Asn Lys Tyr Tyr Arg Phe Asn Glu Glu Leu Arg Ala Val Asp Ser Glu Tyr Pro Lys Asn Ile Lys Val Trp Glu Gly Ile Pro Glu Ser Pro Arg CA 0226l~67 l999-0l-27 W O 98/04287 PCT~EP97/04110 Gly Ser Phe Met Gly Ser Asp Glu Val Phe Thr Tyr Phe Tyr Lys Gly Asn Lys Tyr Trp Lys Phe Asn Asn Gln Lys Leu Lys Val Glu Pro Gly 49c Tyr Pro Lys Ser Ala Leu Arg Asp Trp Met Gly Cys Pro Ser Gly Gly Arg Pro Asp Glu Gly Thr Glu Glu Glu Thr Glu Val Ile Ile Ile Glu Val Asp Glu Glu Gly Gly Gly Ala Val Ser Ala Ala Ala Val Val Leu Pro Val Leu Leu Leu Leu Leu Val Leu Ala Val Gly Leu Ala Val Phe Phe Phe Arg Arg His Gly Thr Pro Arg Arg Leu Leu Tyr Cys Gln Arg Ser Leu Leu Asp Lys Val (2) INFORMATION FOR SEQ ID NO: 6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 582 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (ili) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Rattus rattus CA 02261~67 1999-01-27 W098/04287 PCT~ ~7/04110 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:
Met Ser Pro Ala Pro Arg Pro Ser Arg Ser Leu Leu Leu Pro Leu Leu Thr Leu Gly Thr Thr Leu Ala Ser Leu Gly Trp Ala Gln Ser Ser Asn Phe Ser Pro Glu Ala Trp Leu Gln Gln Tyr Gly Tyr Leu Pro Pro Giy Asp Leu Arg Thr His Thr Gln Arg Ser Pro Gln Ser Leu Ser Ala Ala Ile Ala Ala Ile Gln Arg Phe Tyr Gly Leu Gln Val Thr Gly Lys Ala Asp Ser Asp Thr Met Lys Ala Met Arg Arg Pro Arg Cys Gly Val Pro Asp Lys Phe Gly Thr Glu Ile Lys Ala Asn Val Arg Arg Lys Arg Tyr Ala Ile Gln Gly Leu Lys Trp Gln His Asn Glu Ile Thr Phe Cys Ile Gln Asn Tyr Thr Pro Lys Val Gly Glu Tyr Ala Thr Phe Glu Ala Ile Arg Lys Ala Phe Arg Val Trp Glu Ser Ala Thr Pro Leu Arg Phe Arg T __ CA 02261~67 1999-01-27 WO 98/04287 PCT~EP97/04110 Glu Val Pro Tyr Ala Tyr Ile Arg Glu Gly His Glu Lys Gln Ala Asp Ile Met Ile Leu Phe Ala Glu Gly Phe His Gly Asp Ser Thr Pro Phe Asp Gly Glu Gly Gly Phe Leu Ala His Ala Tyr Phe Pro Gly Pro Asn Ile Gly Gly Asp Thr His Phe Asp Ser Ala Glu Pro Trp Thr Val Gln Asn Glu Asp Leu Asn Gly Asn Asp Ile Phe Leu Val Ala Val His Glu Leu Gly His Ala Leu Gly Leu Glu His Ser Asn Asp Pro Ser Asp Ile Met Ala Pro Phe Tyr Gln Trp Met Asp Thr Glu Asn Phe Val Leu Pro Asp Asp Asp Arg Arg Gly Ile Gln Gln Leu Tyr Gly Ser Lys Ser Gly Ser Pro Thr Lys Met Pro Pro Gln Pro Arg Thr Thr Ser Arg Pro Ser Val Pro Asp Lys Pro Arg Asn Pro Thr Tyr Gly Pro Asn Ile Cys Asp Gly Asn Phe Asp Thr Val Ala Met Leu Arg Gly Glu Met Phe Val Phe CA 0226l~67 l999-0l-27 W098/04287 PCT~EP97/04110 Lys Glu Arg Trp Phe Trp Arg Val Arg Asn Asn Gln Val Met Asp Gly Tyr Pro Met Pro Ile Gly Gln Phe Trp Arg Gly Leu Pro Ala Ser Iie Asn Thr Ala Tyr Glu Arg Lys Asp Gly Lys Phe Val Phe Phe Lys Gly Asp Lys His Trp Val Phe Asp Glu Ala Ser Leu Glu Pro Gly Tyr Pro Lys His Ile Lys Glu Leu Gly Arg Gly Leu Pro Thr Asp Lys Ile Asp Ala Ala Leu Phe Trp Met Pro Asn Gly Lys Thr Tyr Phe Phe Arg Gly Asn Lys Tyr Tyr Arg Phe Asn Glu Glu Phe Arg Ala Val Asp Ser Glu Tyr Pro Lys Asn Ile Lys Val Trp Glu Gly Ile Pro Glu Ser Pro Arg Gly Ser Phe Met Gly Ser Asp Glu Val Phe Thr Tyr Phe Tyr Lys Gly Asn Lys Tyr Trp Lys Phe Asn Asn Gln Lys Leu Lys Val Glu Pro Gly Tyr Pro Lys Ser Ala Leu Arg Asp Trp Met Gly Cys Pro Ser Gly Gly CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 Arg Pro Asp Glu Gly Thr Glu Glu Glu Thr Glu Val Ile Ile Ile Glu Val Asp Glu Glu Gly Ser Gly Ala Val Ser Ala Ala Ala Val Val Leu Pro Val Leu Leu Leu Leu Leu Val Leu Ala Val Gly Leu Ala Val Phe Phe Phe Arg Arg His Gly Thr Pro Lys Arg Leu Leu Tyr Cys Gln Arg Ser Leu Leu Asp Lys Val (2) INFORMATION FOR SEQ ID NO: 7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 582 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Mus cookii (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:
Met Ser Pro Ala Pro Arg Pro Ser Arg Ser Leu Leu Leu Pro Leu Leu Thr Leu Gly Thr Ala Leu Ala Ser Leu Gly Trp Ala Gln Gly Ser Asn CA 02261~67 1999-01-27 W098/04287 PCTrEP97/04110 Phe Ser Pro Glu Ala Trp Leu Gln Gln Phe Gly Tyr Leu Pro Arg Gly Asp Leu Arg Thr His Thr Gln Arg Ser Pro Gln Thr Leu Ser Val Asp Ile Ala Ala Ile Gln Lys Phe Tyr Gly Leu Tyr Val Thr Gly Lys Ala Tyr Ser Glu Thr Met Lys Ala Met Arg Arg Pro Arg Cys Gly Val Pro Asp Lys Phe Gly Thr Glu Ile Lys Ala Asn Val Arg Arg Lys Arg Tyr Ala Ile Gln Gly Leu Lys Trp Gln His Asn Glu Ile Thr Phe Cys I le 51n Asn Tyr Thr Pro Lys Val Gly Glu Tyr Ala Thr Phe Glu Ala Ile Arg Lys Ala Phe Arg Val Trp Glu Ser Ala Thr Pro Leu Arg Phe Arg 16(~
5lu Val Pro Tyr Ala Tyr Ile Arg Glu Gly His Glu Lys Gln Ala Asp Ile Met Ile Leu Phe Pro Glu Gly Leu His Gly Asp Ser Thr Pro ?he Asp Gly Glu Gly Gly Phe Leu Ala His Ala Tyr Phe Pro Gly Pro .-:sn CA 02261~67 1999-01-27 WO 98/04287 PCT~EP97/04110 Ile Gly Gly Asp Thr His Phe Asp Ser Ala Glu Pro Trp Thr Val Gln Asn Glu Asp Leu Asn Gly Asn Asp Ile Phe Leu Val Ala Val His Glu Leu Gly His Ala Leu Gly Leu Glu H1s Ser Asn Asp Pro Ser Asp Ile Met Ser Pro Phe Tyr Gln Trp Met Asp Thr Glu Asn Phe Val Leu Pro Asp Asp Asp Arg Arg Gly Ile Gln Gln Leu Tyr Gly Ser Lys Ser Gly Ser Pro Thr Lys Met Pro Pro Gln Pro Arg Thr Thr Ser Arg Pro Ser Val Pro Asp Lys Pro Lys Asn Pro Ala Tyr Gly Pro Asn Ile Cys Asp Gly Asn Phe Asp Thr Val Ala Met Leu Arg Gly Glu Met Phe Val Phe Lys Glu Arg Trp Leu Trp Arg Val Arg Asn Asn Gln Val Met Asp Gly Tyr Pro Met Pro Ile Gly Gln Phe Trp Arg Gly Leu Pro Ala Ser Ile Asn Thr Ala Tyr Glu Arg Lys Asp Gly Thr Phe Val Phe Phe Lys Gly CA 02261~67 1999-01-27 Asp Lys His Trp Val Cys Val GlU Ala Ser Leu Glu Pro Gly Tyr Ala Asn His Ile Lys Glu Leu Val Arg Gly Leu Pro Ser Asp Lys I l e Asp Thr Ala Leu Phe Trp Met Pro Asn Gly Lys Thr Tyr Phe Phe Arg Gly Asn Lys Tyr Tyr Arg Phe Asn Glu Glu Phe Arg Ala Val Asp Ser Glu Tyr Pro Lys Asn Ile Lys Val Trp Glu Gly Ile Pro Glu Ser Pro Arg Gly Ser Phe Met Gly Ser Asp Glu Val Phe Thr Tyr Phe Tyr Lys Gly Asn Lys Tyr Trp Lys Phe Asn Asn Gln Lys Leu Lys Val Glu Pro Gly Tyr Pro Lys Ser Ala Leu Arg Asp Trp Met Gly Cys Pro Ser Gly Gly Arg Pro Asp Glu Gly Thr Glu Glu Glu Thr Glu Val Ile Ile I le Glu Val Asp Glu GlU Gly Ser Gly Ala Val Ser Ala Ala Ala Val Val Leu Pro Val Leu Leu Leu Leu Leu Val Leu Ala Val Gly Leu Ala Va l Phe CA 02261~67 1999-01-27 W098/04287 PCT~P97tO4110 Phe Phe Arg Arg His Gly Thr Pro Lys Arg Leu Leu Tyr Cys Gln Arg Ser Leu Leu Asp Lys Val (2) INFORMATION FOR SEQ ID NO: 8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:1 (D) OTHER INFORMATION:/product= "X is Abz-G"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:6 (D) OTHER INFORMATION:/product= "X is Lnor"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:9 (D) OTHER INFORMATION:/product= "X is Y(NO2)"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8:
Xaa Pro Leu Gly Leu Xaa Ala Arg Xaa (2) INFORMATION FOR SEQ ID NO: 9:
(i) SEQUENCE CHARACTERISTICS:
(A'' LENGTH: 10 amino acids (B' TYPE: amino acid (C) STRANDEDNESS: single (D' TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES

CA 02261~67 1999-01-27 (iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:1 (D) OTHER INFORMATION:tproduct= "X is Abz-S"
. (ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:5 (D) OTHER INFORMATION:/product= "X is hydroxyproline"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:g (D) OTHER INFORMATION:/product= "X is Y (N02)"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9:
Xaa Lys Tyr Pro Xaa Ala Leu Phe Xaa Asp (2) INFORMATION FOR SEQ ID NO: 10:
(l) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 10:
Cys Asp Gly Asn Phe Asp Thr Val Ala Met Leu Arg Gly Glu Met (2) INFORMATION FOR SEQ ID NO: 11:
; SEQUENCE CHARACTERISTICS:
(A) LENGTH: 6 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear r CA 02261~67 1999-01-27 (il) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:1 (D) OTHER INFORMATION:/product= "X is Mca-P"
(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:5 (D) OTHER INFORMATION:/product= "X is Dpa-A"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 11:
Xaa Leu Gly Leu Xaa Arg (2) INFORMATION FOR SEQ ID NO: 12:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA lgenomic) (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Oryctolagus cuniculus (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 12:
CGGGATCCCT GTGGGTCACT TCTTCT

(2) INFORMATION FOR SEQ ID NO: 13:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) CA 02261~67 1999-01-27 (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: YES
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Oryctolagus cuniculus (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 13:
CCGCTCGAGC TGGCACCATT ACTAGC

(2) INFORMATION FOR SEQ ID NO: 14:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 4 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:4 (D) OTHER INFORMATION:/product= "X is K (Abz)-PEGA"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 14:
Leu Phe Phe Xaa (2) INFORMATION FOR SEQ ID NO: 15:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 8 amino acids (B) TYPE: amino acid ~C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

T

CA 02261~67 1999-01-27 WO 98/04287 PCT~EP97/04110 (ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:1 (D) OTHER INFORMATION:/product= "X is Abz-G"
(ix) FEATURE:
(A) NAME/KEY: Cleavage-site tB) LOCATION:4..5 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 15:
Xaa Pro Leu Gly Leu Xaa Ala Arg (2) INFORMATION FOR SEQ ID NO: 16:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECVLE TYPE: peptide (iii) HYPOTHETICAL: YES
~iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modi~ied-site (B) LOCATION:4 (D) OTHER INFORMATION:/product= "X = J"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 16:
Tyr Pro Leu Xaa Met Lys Gly Lys Gly (2) INFORMATION FOR SEQ ID NO: 17:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

CA 02261~67 1999-01-27 W098/04287 PCT~EP97/04110 (ix) FEATURE: -(A) NAME/KEY: Modified-site (B) LOCATION:2..6 (D) OTHER INFORMATION:/product= "each X =J"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 17:
Asn Xaa Tyr Pro Xaa Xaa Tyr Lys Gly (2) INFORMATION FOR SEQ ID NO: 18:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iil) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(ix) FEATURE:
(A) NAME/KEY: Modified-site (B) LOCATION:3..8 (D) OTHER INFORMATION:/product= "each X = J"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 18:
Tyr Pro Xaa Xaa Met Lys Gly Xaa Gly (2) INFORMATION FOR SEQ ID NO: 19:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA ~genomic) (iil) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

WO 98/04287 PCT~EP97/04110 (x~,) SEQUENCE DESCRIPTION: SEQ ID NO: 19:
TGGTATGTGG TCTGTGT

(2) INFORMATION FOR SEQ ID NO: 20:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: YES

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 20:
TGTGGTTCAG TTGTGGT

(2) INFORMATION FOR SEQ ID NO: 21:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 21:
ACCACAACTG AACCACA

(2) INFORMATION FOR SEQ ID NO: 22:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear W098/04287 PCT~EP97/04110 (il) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: YES

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 22:
GGACTCATGG TGAGGAC

(2) INFORMATION FOR SEQ ID NO: 23:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 23:
CGGATACAGG TGTCGGA

Claims (12)

1. The use of an agent in the manufacture ofa medicament forthe treatment of bone metabolic disease, characterised in that the agent acts by inhibitionof the production or action of a membrane associated protease or the matrix metalloprotease MMP-12 involved in the resorptive activity of osteoclasts.

2. The use claimed in Claim 1, wherein the agent acts by inhibition of the production or action of a membrane-type matrix metallo-proteinase (MT-MMP) or the matrix metalloproteinase MMP-12 involved in the resorptive activity of osteoclasts.

3. The use claimed in Claim 2, wherein a protease is inhibited which is involved in the recruitment, proliferation, differentiation, or migration of osteoclast precursor cells or in the migration, fusion, attachment , polarisation, activity in removal of mineralised osseous substance, or death of osteoclasts.

4. The use claimed in any preceding claim, wherein the agent is an antibody selectively immunoreactive with a said protease.

5. The use claimed in any one of Claims 1 to 3, wherein the agent is an antisense oligonucleotide or oligonucleotide analogue directed against a gene involved in the production of a said protease.

6. The use claimed in any one of Claims 1 to 3, wherein the agent is a protease substrate mimic inhibitor.

7. The use claimed in any one of Claims 1 to 3, wherein the agent is a broad spectrum matrix metalloproteinase (MMP) inhibitor or a broad spectrum membrane-associated metallo-proteinase inhibitor.

8. The use claimed in any one of Claims 1 to 3, wherein the agent is a selective inhibitor of MT1-MMP, MMP-12 or a specific member of one of the families of membrane-associated metallo-proteinase, such as the meltrins or ADAMs.

9. The use claimed in any one of Claims 1 to 3, wherein the agent is a peptide or peptide analogue obtained by screening a peptide library for peptides reactive with a said protease.

10. The use of an agent in the manufacture of a medicament for the treatment of bone metabolic disease by inhibition of the recruitment, proliferation, differentiation, or migration of osteoclast precursor cells or in the migration, fusion, attachment, polarisation, or death of osteoclasts.

11. The use claimed in Claim 10, wherein said agent produces said inhibition by inhibiting the production or action of a proteinase.

12. An anti-bone resorption agent comprising a proteinase inhibitor active against a proteinase involved in bone resorption operatively linked to a ligand having binding specificity targeting the inhibitor to said proteinase or to the environment of the proteinase.