MXPA97004015A - Metaloproteinase inhibitors of matrix, based on malon acid - Google Patents

Abstract

The present invention relates to a compound, characterized in that it is represented by the general formulas I, IIóIII, which binds and inhibits the matrix metalloproteinases (MMP), in the formulas: X1 is oxygen or sulfur, R1 is OH, SH, CH2OH, CH2SH or NHOH, R2 is a residue of 2 to 10 atoms in the main structure, which binds to amino acid 161 of HNC, the residue is saturated or unsaturated, is straight or branched and preferably contains homocyclic or heterocyclic structures, X2 is oxygen or sulfur and binds as a hydrogen bond acceptor in amino acid 160 of HNC, and is a residue which binds to the S1'from HNC and consists of at least 4 atoms in the main structure of Z1Z2Z3-Z4 -R3 (formula IV), R3 is n-propyl, isopropyl, isobutyl or a residue with at least 4 atoms in the main structure, which is no greater than a tricyclic ring system, and R4 is hydrogen, alkyl or aryl, or a salt thereof, with the condition that e the compound is not HONH-DL-CO-CH- (CH2C6H5) -CO-L-Ala-Gly-NHC6H4N

Description

METALOPROTEINASE INHIBITORS OF MATRIX. BASED ON MALONIC ACID DESCRIPTION OF THE INVENTION The invention relates to new matrix metalloproteinase inhibitors which are based on the structure of (pseudo) malonic acid. The invention additionally comprises methods for the production of the inhibitors and their use, especially in the field of therapeutics. Matrix metalloproteinases (MMP, Matrixins) comprise a family of Ca-containing Zn endopeptidases, which show proteolytic activities towards most, if not all, constituents of the extracellular matrix such as interstitial membrane collagen and basamental, fibronectin and laminin. They play a basic role in the remodeling of normal tissue and are particularly involved in other processes such as ovulation, embryonic growth and differentiation1-2,3,4. At least 11 different and highly homologous species of MMP have been characterized, including collagenase of interstitial fibroblasts (MMP-1, HFC), neutrophil collagenase (MMP-8, HNC), REF: 24738 two gelatinases, stromelysins (such as HSL-1) and HPUMP (for a recent review, see Birkedal-Hansen et al.2). These proteinases share many structural and functional characteristics but differ to some extent in their substrate specificity. Only HNC and HFC are able to break down native triple helix collagens, types I, II and III in a single bond with the production of fragments 3/4 and 1/4 of the native chain length. This lowers the melting point of collagen and makes it accessible for further attack by other matrix degrading enzymes. All MMPs are secreted as multiple domain proenzymes proteolytically activatable with an activation peptide of -80 residues which in most cases is followed by the catalytic domain of -165 residues, terminated by an opexin-like domain of -210 waste. The catalytic domain contains the conserved zinc binding sequence HEXXHXXGXXH, characteristic of the "metzincin" superfamily 5 and shows complete activity towards most of the small peptide substrates6,7,8,20. MMPs are important for the development and remodeling of normal tissue and have been implicated in various disease processes such as the growth of tumors and, metastases, rheumatoid arthritis and osteoarthritis, periodentitis, corneal ulceration, atherosclerosis and emphysema (for reference, see reviews1,2,3,4). Therefore, inhibitors for MMP can be used to treat these diseases. To date, three endogenous protein inhibitors (TIMPI1, II, III) have been described, which block the proteolytic activity of MMPs in a more or less specific manner10,11,12. Virtually all the specific synthetic collagenase inhibitors designed so far are reversible peptidyl inhibitors which interact with the active site of their target enzyme. They contain a chelating group capable of interacting with the catalytic zinc (without removing it), such as hydroxamate, thiol, carboxylate or phosphinic group, coupled with a peptide portion used to bind to the enzyme recognition site of the enzyme13,14,15,1 *, in this way, the inhibitors aim and are specific for the desired enzyme, with zinc. The invention defines a new class of MMP inhibitors, which bind to MMP in a completely different way compared to the synthetic inhibitors mentioned above. The new inhibitors are compounds which are expressed by the general formulas I, II or III and salts thereof, for the inhibition of matrix metalloproteinases (MMP), in which Xx is oxygen or sulfur. Rj. is OH, SH, CH2OH, CH2SH or NHOH. R2 is a residue of 2 to 10 atoms in the main structure, which binds to amino acid 161 of HNC, the residue is saturated or unsaturated, is linear or branched and preferably contains homocyclic or heterocyclic structures, X2 is oxygen or sulfur and binds as a hydrogen bond acceptor at amino acid 160 of HNC, and is a residue which binds to the SI bag of HNC and consisting of at least 4 main structure atoms Z1-Z2-Z3-Z4-R3, R3 is n-propyl, isopropyl, isobutyl or a residue with at least 4 atoms in the main structure , which is not larger than the tricyclic ring system, and R 4 is hydrogen, alkyl or aryl, preferably isopropyl, n-butyl, benzyl. The term "inhibition" according to the invention means a substantial reduction of collagenase activity in vitro and in vivo. The collagenase activity can be determined in vitro, for example, in an enzymatic assay according to F. Grams (1993) 51. "Substantial inhibition" means an inhibition of at least about 50%, preferably less than the concentration mmolar inhibitor (based on collagenase activity without an inhibitor) Generally, an inhibition of more than about 80% to 90% is found In a preferred embodiment of the invention, the Y residue consists of a peptide or peptidomimetic group. structure Z ^ ZJ-ZJ-Z ^ RJ consists of 4 atoms in the main structure that form a dihedral angle of approximately 0o (sp2 or sp3 hybridization), in which the distance between Z and Z4 is between 2.5 and 3.0 Á (see examples of formula IV) Zx and Z4 can be joined to form a cyclic structure Preferred radicals for cyclic substructures are mimetic peptide ring structures, such as phenylene, pyridinyl, pyrazole nyl-, pyrimidinyl, pyridazinyl, piperazinyl, indolinyl and morpholinyl. The preferred R3 residues are isopropyl, amino acid, piperidinyl, pyridinyl and furyl. The compounds according to formulas I, II or III consist of three parts which have different structures and different properties: The chelating group C i ^, phosphinoyl or phosphono), the primary binding site which is referred to herein in the following as the group of the tail (Y) and the secondary binding site (C2R2). The chelating group of the new inhibitors interacts with the catalytic zinc (which is located in the lower part of the active cleavage site in the MMP is penta-coordinated with three histidines and by Rt and by X of the inhibitor) in a bidentate manner. The group of the inhibitor queue according to the invention adopts a curved conformation and is inserted into the SI 'pocket (subsitium) and does not bind to the subsites S2' and S3 '. In contrast to this, the tail group of most known inhibitors bind extensively along the active site cleavage (for the definition of active sites see 39). The reason for the difference in binding between the new inhibitors and the inhibitors of the state of the art is based on several new and essential structural characteristics of the inhibitors according to the invention. 1. The basic structure of acid (pseudo) malonic.

The structure of malonic acid defines three binding positions of the inhibitor. The chelating group is bound, via R, and X as a bidentate to the active zinc site in the MMPs. Preferably, the bidentate structure can be a hydroxamate, thiol, carboxylate, phosphinoyl or phosphono group. The second binding site is defined by the interaction between R2 and amino acid 161 of HNC. The term "amino acid linkage 161" means binding to the surface area around amino acid 161, so, preferably, attachment to amino acid 161. is included. The linkage is derived, for example, from the van der Waals interaction. or hydrophilic. For an optimized bond it is preferred that R2 be an alkyl, alkenyl, or alkoxy residue with 2 to 10 atoms in the main structure (C, N, 0, S) or a cyclo (hetero) alkyl or aromatic residue with 5 to 10 atoms in the main structure (C, N, 0, S). An additional connection of the basic structure of malonic acid to the MMPs is carried out by means of oxygen or sulfur X2 to the group of the tail. This oxygen or sulfur is hydrogen bonded to the proton amide of leucine 160 of HNC. 2. The group of the tail The second carbonyl group (= C3X2) of the basic structure of malonic acid is bound to the primary linking group of the inhibitors according to the invention (tail group Y, see Examples of formula IV). This structure comprises 4 atoms in the main structure that form a dihedral angle of approximately 0 o (sp2 or sp3 hybridization). In a preferred embodiment of the invention, the rotation is part of a cycle with 5 or 6 atoms. Therefore, the distance between Zl and Z4 is preferred to be between 2.5 and 3.0 Á. At the Z4 position, there is an additional residue R3 which is n-propyl, isopropyl, isobutyl or a residue with at least four atoms in the main structure which, however, is not greater than a tricyclic ring system. The inhibitor according to the invention is bound, via Y, to the SI 'bag. The bag consists of: 1) human neutral collagenase (HNC): amino acids L 193, V 194, H 197, E 188 i L 214, Y 216, P 217, and 219, A 220, R 22 (numbering according to Reinemer et al. (1994) 17). 2) HFC: Amino acids L 181, A 182, R 214, V 215, H 218, SE 219, Y 237, P 238, S 239, and 240 (numbering according to Lovejoy et al. (1994) 49). 3) Stromelysin: Amino acids L 197, V 198, H 201, E 202, L 218, Y 220, L 220, L 222, Y 223, H 224, S 225, A 226 (numbering according to Gooley et al. 1994) 50).

Therefore, an essential characteristic of the compounds is that, in contrast to the collagenase inhibitors according to the state of the art, the structure of the group of the tail is highly relevant for the inhibitory activity of the compounds according to the invention. The groups in the queue of the inhibitors according to the state of the art do not bind to the essential binding sites in the MMP. In the case of the compounds according to the invention, however, the binding of the group of the tail Y to the SI 'bag of the MMPs constitutes the essential part of the binding between the inhibitor and the collagenase and, consequently, of the inhibitory activity. Therefore, it is essential that the tail group Y have a structure which fits well in the SI bag. These requirements are met by synthetic compounds which contain a zinc chelating group which is separated with a carbon of the substituent R2, which constitutes an auxiliary binding site. It interacts with the surface of the protein through van der Waals and / or hydrophilic interactions. The tail group Y must be designed for insertion into a protein pocket of well defined geometry and surface properties. In the upper part, the environment is mainly hydrophobic, where hydrophobic interactions can be exploited, while the lower part also contains several hydrophilic sites, which allows the connections with hydrogen. Correspondingly, the hydrogen bond acceptors and the donors are constructed in this portion of the inhibitory molecules. Due to the short distance between the zinc binding region and X2, in connection with the rotation structure mentioned above Y, an "L-based" structure of the inhibitor is obtained when it joins the MMPs. Most collagenase inhibitors according to the state of the art are based on a succinyl basic structure and show a larger distance between the zinc and C3 binding region. Therefore, R2 joins the SI bag, and there is no opportunity for the queue group to join this bag. Therefore, collagenase inhibitors according to the state of the art, in contrast to the invention, bind in a manner similar to the substrate and therefore exhibit an extended main structure in the MMP-bound state. Collagenase inhibitors of this type are described, for example, in U.S. Patent No. 4,595,700 (in which the separator is represented by the chiral center b), U.S. Patent No. 4,599,361, (the separator is represented by boc) , EP-A 0231081 (the separator is represented by (CH2) n of, formula I), EP-A 0 236 872 (the separator is represented by CHR3 of the formula I), EP-A 0 276 436 ( the separator is represented by CH2 of the formula I), WO 90/05719 (the separator is represented by the C atom which is connected with and with CONHOH), EP-A 0 489 579, EP-A 0 489 579 and WO 93/14096 (the separator is represented by CR2), EP-A 0 497 192 (the separator is represented by the C atom which connects to and RJ, WO / 92/16517 (the separator is represented by the C atom which connects C02H and CO), EP-A 0 520 573 (the separator is represented by NH, which connects CHC02H and CHR, WO 92/10464 ( wherein the separator is represented by one of the C atoms which connects ROCO and CCO) and WO 93/09097 (the separator is represented by the C atom which connects CONHOH and CR2). Additional collagenase inhibitors are disclosed in EP-A 0 320 118 and WO 92/21360. This structure differs especially for another essential characteristic. The molecule contains, instead of O, an NH group (located between CR2 and CR3). This NH group, in contrast to CiO, is an electron donor, and therefore, completely changes the properties of the molecule. From this, it is clear that this molecule can not bind to collagenase in a manner similar to that of the inhibitors according to the invention.

In addition, the collagenase inhibitors of WO 92/09563 show a completely different structure and, therefore, they must be linked to collagenases in a completely different way from that of the prior art inhibitors. The binding properties mentioned before the inhibitor can be determined by X-ray crystallographic techniques. Such methods are described, for example, by W. Bode et al., EMBO J. 13 (1994) 1263-1269, which is incorporated in the present as a reference for these techniques and for the crystalline structure of the catalytic domain of HNC.

Main characteristics of the catalytic domain of MMP MMPs, for example HNC, show a spherical shape, with a narrow active site cleavage separating a larger "upper" N-terminal domain from a smaller "lower" C-terminal domain. The upper main domain consists of a highly threaded, folded, 3-strand 3-sheet, central (with the ordered strands ß 2, 1, 3, 5, 4 and the strand 5 represents only the antiparallel strand), flanked by a double S-shaped curl and two additional curls that form a bridge on its convex side, and by two long a-helices that include an active site helix on its concave side. The important substrate binding and inhibitory regions are the strand of the "edge" Leu (160) to Phe (164) of the β-sheet placed "on top" of the active site helix and forming the edge "to the north "of the active site slit, and the preceding" outgoing "segment Gly (155) to Leu (160), hereinafter referred to as the" outgoing segment ", which is part of the S-shaped double loop. The zinc ion "catalytic" (Zn (999)) is located in the lower part of the active site cleavage and is coordinated with imidazole atoms Ne2 of three histidine residues of the zinc binding consensus sequence His (197) -Glu (198 ) -XX-His (201) -XX-Gly (204) -XX-His (207), and by one or two inhibitory atoms. In addition, the catalytic domain harbors a second "structural" zinc ion (Zn (998)) and two calcium ions packaged against the top of the β-sheet. The lower, small domain consists of two concatenated broad curls and three turns of a-helix C terminals. The first of these wide curls that rotate to the right include a narrow, 1.4-turn narrowing, from Ala (213) to Tyr (216). This "Met turn" represents a topological element conserved in the "metzincins" 5 which provides a hydrophobic base for the three His residues, which bind to the catalytic zinc. The peptide chain then advances to the molecular surface in Pro (217) where the chain is linked and continues in an extended strand Pro (217) -Thr (224). The SI bag is immediately at the "right" of the catalytic zinc and is formed by a long surface crack (running perpendicular to the crack in active site) separated from the bulky water by the initial part of this strand Pro (217) -Tyr (219) which forms the wall exterior (referred to as the "wall-forming segment"). The entrance to this bag is formed by i) the outgoing segment Gly (158) -He (159) -Leu (160) and the initial part of the edge strand Leu (l60) -Ala (161) that forms the side " top "of the bag, ii) the side chain Tyr (219)" right "side," iii) the wall-forming segment that includes the side chain Asn (218) ("bottom" side) and iv) the catalytic zinc together with the carboxylate group Glu (198) ("left" side). The features provide a series of polar groups for fixing bound peptide substrates and inhibitors by hydrogen bonding (see below). The polar opening of bottleneck opens inside the much more hydrophobic interior of the bag surrounded mainly by i) the side chains of Leu (160) and Val (194), ii) the side chain Tyr (219), iii ) the flat faces of the amide groups that make up the wall-forming segment Pro (217) -Tyr (219), and iv) the flat side of the imidazole ring of His (197). The inner part of the bag is filled with four molecules of "internal" water joined with 4 transverse hydrogens, in addition to 3 to 4 molecules of solvent located at the entrance of the bag. The lower part of the bag is partially insulated by the chain on the long side of Arg (222) which is flanked by the side chains of Leu (193) and Leu (214) which extend "behind" / "under" towards the Turn Met. The terminal guanidyl group is weakly bound by hydrogen bonds to Pro (211) 0, Gly (212) 0 and / or Ala (213) 0. Several interposed polar groups provide fixation points for enclosed water molecules, one of which is in direct hydrogen bonding contact with a water molecule, located through an opening that is on the left between the side chain Arg (222) and the wall-forming solvent.

Union of inhibitors of the state of the art In order to demonstrate the binding difference of the inhibitors of the state of the art and the inhibitors according to the invention, two models of inhibitors were designed, which represent the basic structure of the inhibitors according to the state of the art.

These inhibitors are mentioned, in the following, as MBP-AG-NH2 and PLG-NHOH.

Interactions of the main chain of inhibitors of the state of the art The chains of inhibitors of PLG-NHOH and of MBP-AG-NH2 in complexes with HNC join in a more or less extended conformation. A substrate model can be constructed which comprises both inhibitory conformations by exchange of the zinc chelating groups by a normal peptide bond. The main chain from P3 to P3 'is stabilized by four hydrogen bonds to the active site edge strand and two hydrogen bonds to the wall-forming segment of the SI' pocket. According to this main chain conformation, the Cat-C? Pl 'of a peptide chain linked to MMP with a Pl' residue L-shaped will point "back, down" which allows any bulky side chain (particularly aromatic) to be immersed in the SI 'bag.

The subsite SI ' The main interactions between the substrates and the inhibitors occur between the residue Pl 'and the subsite SI'. In the MBP-AG-NH2 / collagenase complex, the side benzyl chain is adjusted to the SI 'subsite which has a depth of approximately 9 A and a width of 5 x 7 A between the surfaces of van der Waals. The HNC hydrolyzes substrates with the relative preference in Pl 'of Tyr > Leu ~ Met ~ Ile ~ Leu ~ Phe > Trp > Val ~ Glu > Ser ~ Gln ~ Arg31"33, which suggests that hydrophobic interactions are more important for binding and catalysis compared to polar interactions, however, distal polar side chain groups such as the hydroxyl portion of Tyr seem to have some beneficial effect at the junction, probably through the hydrogen bonding interactions with the circumscribed water molecules.The base of the "bag" seems to be adaptable due to the mobility of the side chain Arg (222) which can act as a fin and maintaining the distinct shape of the bag by stabilizing the wall-forming segment The SI bag has a narrow bottle neck, but it is much larger than that required for any of the naturally occurring amino acids. , the SI 'bag of collagenase-related fibroblast differs considerably at the bottom, due to simultaneous substitutions of Arg (222) of HNC by Ser, and from Leu (193) by Arg, much larger and polar, in the active site helix that forms part of the "upper" bag wall. Similar to Arg (222) in HNC, the Arg side chain (193) encompasses the lower part of the bag in HFC which considerably reduces the depth of the SI'18 bag. This explains the much lower tolerance of HFC by Trp in Pl 'compared to HNC31. It is notable that no other human MMP (except HSL-3) has Arg residues at positions 222 or 193, but they contain Arg residues at positions 226 and / or 228, which may have a similar function. In this context, it is also interesting that in all MMPs except HPUMP the side chain of Val (194) is part of the SI 'wall. In HPUMP, the Val residue is replaced by a Tyr. It can be mentioned that the corresponding SI 'bag in ter smell is also of a much smaller size since it is limited to the "back" part by the active site helix and is completely embedded in the protein matrix.

Other subsites The narrow adjustment of the PLG-NHOH proline in the hydrophobic slit-like subsite S3 explains the beneficial aspect of P3-Pro on the inhibitor binding34,35 and the specificity of rupture31"33 and agrees with the strict presentation of Pro in any collagen substrate (see Birkedal -Hansen, 19932) .Gl (II) in Pl, although present in all sites of collagen rupture, does not use at all interactions with the edges of the cleft; the large hydrophobic chains as well as the polar laterals probably improve the union. This is consistent with studies of rupture activity31"33 that show that Glu, in addition to Ala, is a more favorable residue Pl for HNC compared to Pro, Met, His, Tyr, Gly and Phe. The excellent property of Glu and effect damaging Arg in Pl may be due to the favorable and unfavorable interactions of hydrogen bonds with the protonated Ndl atom of His (162), respectively.The Glu hindrance effect of HFC rupture may be due to the blocking effect of the proximity of Asn (159) which replaces He (159) of HNC in HFC The interaction of the side chain Leu (12) allows a considerable reduction of the hydrophobic surfaces accessible to the solvent in both components, this side chain, however, is not voluminous enough to fill the narrow S2 subsite completely Although Leu is apparently the optimal side amino acid at P231.32, a stronger binding can be obtained by introducing artificial amino acids into this position. The side chain of Ala (12) in position P2 'of MBP-AG-NH2 certainly does not contribute well to the junction, due to the lack of interaction both to the lack of interaction with both edges of flanking HNC, formed mainly by Gly (158) as to the side chain of He (159) in the "north" and by the side chain of Asn (218) in the "south". In fact, most of the two most powerful MMP inhibitors published so far30,36 contain voluminous side chains, mainly hydrophobic in this position, which has led to help differentiate between different MMPs, probably due to the substitution of Gly ( 158) by His (HSL-2) or Asn (HSL-1, HPUMP) and Ile (159) by Asn (HFC), Ser (HSL-2) or Thr (HPUMP). In model peptide substrates, replacement of Ala in this position by Phe, Trp or Leu is correlated with an increase in the rate of hydrolysis by HNC and HFC, and the most notable effects are observed in the case of substitution of Trp; interestingly, these implements in the specificity constants result mainly from K "values decreased36,38, indicating closer interactions.

Mode of binding of the inhibitors of the invention The inhibitors according to the invention join unexpectedly in a geometry not similar to the substrate. This structure is a main structure for the design of more potent collagenase inhibitors. Substitution of portions of the structure with peptidomimetic groups or non-peptide groups and replenishment of surfaces accessible to the solvent can lead to substantial improvements in the potency of the inhibitor. Several factors together seem to affect this strange joining geometry and not seen before. Importantly, a favorable zinc coordination of the flat hydroxamic acid group appears to be incompatible with the observed placement of the adjacent isobutyl group in the SI 'pocket. This binding geometry represents an energy compromise since an optimal hydroxamate-zinc interaction is preferred instead of a favorable embedding of the "PI-like side chain" in the SI 'pocket. An isobutyl group used as R2 shows only a moderate reduction of its surface accessible to the solvent before complex formation and clearly represents an appropriate point at which the modifications can lead to improved binding and selectivity properties. Conversely, the hydroxamic acid compounds (such as BB-9437) possess an additional (substituted) methylene linker between the hydroxamate group and the "PI-like" carbon and the insertion of the Pl '-like side chains in the SI bag. ' The manufacture of the inhibitors according to the invention can be carried out according to methods known in the state of the art. As initial materials, suitable malonic acid esters are used. For the replacement of the acid hydrogen at the R 2 position with larger alkyl or aryl groups, standard or conventional base-catalyzed alkylation reactions of 1,3-dicarbonyl-CH (or enolate alkylation) acidic compounds are used. Hydroxamates are synthesized by acylating hydroxylamine with malonic acid derivatives, for example, mixed anhydride, DCC (dicyclohexylcarbodiimide) or active esters. The oxygen of the carbonyl groups can be replaced by adjustment by the use of O? Exchange reagents. S, for example, potassium thiocyanate 40,41,42, thiourea43,44, 3-methylbenzothiazole-2-thione 4S and triphenylphosphine sulfide 4S or Lowry 47 reagent. In a preferred embodiment, the residue Y consists of peptide or peptidomimetic groups. In the case of Y peptide groups, these are coupled, for example, via methods of peptide coupling to the basic structure of malonic acid according to methods known in the art (Houben-Weyl) 48. In the case of peptidomimetic groups, the methods according to the state of the art are applied. The compounds of the present invention, which specifically inhibit MMPs, are pharmacologically useful in the treatment of rheumatoid arthritis and related diseases in which collagenolytic activity is a contributing factor, such as, for example, ulceration of the cornea, osteoporosis, periodontitis, Paget's disease, gingivitis, tumor invasion, dystrophic epidermiolysis, bullosa, systemic ulceration, epidermal ulceration, gastric ulceration and the like. These compounds are particularly useful in the treatment of rheumatoid arthritis (primary chronic polyarthritis, PCP), systemic lupus erythematosus (SLE), juvenile rheumatoid arthritis, Sjógren's syndrome (RA + sicca syndrome), polyarteritis nodosa and related vasculitis, for example, granulomatosis of Wegener, giant cell arthritis, Goodpasture syndrome, angiitis with hypersensitivity, polyomyositis and dermatomyositis, metastasis, progressive systemic sclerosis, M. Behcet syndrome, Reiter (arthritis + urethritis + conjunctivitis), mixed connective tissue disease (Sharp syndrome) ), ankylopoetic spondylitis (M. Bechtere). The compounds of the present invention can be administered by any suitable route, preferably in the form of a pharmaceutical composition adapted for such day and in an effective dose for the treatment that is desired. Therapeutically effective doses of the compounds of the present invention require avoiding or suppressing the progress of the medical condition or disorder and such doses can be readily determined by a person ordinarily skilled in the art. Accordingly, the invention provides a class of novel pharmaceutical compositions comprising one or more compounds of the present invention, in association with one or more non-toxic pharmaceutically acceptable carriers and / or dilutions and / or adjuvants (collectively referred to herein as "carrier materials") and, if desired, other active ingredients. The compounds and compositions, for example, can be administered intravascularly, intraperitoneally, subcutaneously, intramuscularly or topically. For all administrations, the pharmaceutical composition may be in the form of, for example, a tablet, capsule, suspension or liquid. The pharmaceutical composition is preferably made in the form of a dosage unit contained in a particular amount of the active ingredient. Examples of such dosage units are tablets or capsules. The appropriate daily dose for a mammal can vary widely based on the condition of the patient and other factors. However, doses of from about 0.1 to 300 mg / kg of body weight, particularly from about 1 to 30 mg / kg of body weight, may be appropriate. The active ingredient can also be administered by injection. The dose regimen for the treatment of a disease condition with the compounds and / or compositions of this invention are selected according to various factors including the type, age, weight, sex and medical conditions of the patient. The severity of the infection and the role of administration and the particular compound used and therefore can vary widely. For therapeutic purposes, the compounds of the invention are usually combined with one or more adjuvants appropriate to the indicated route of administration. If administered per os, the compounds may be administered with lactose, sucrose, starch powder, cellulose esters or alkanoic acids, cellulose alkyl ester, talc, stearic acid, magnesium stearate, magnesium oxide, sodium and calcium salts of phosphoric and sulfuric acids, gelatin, acacia, sodium alginate, polyvinylpyrrolidone and / or polyvinyl alcohol and therefore tablets can be made or encapsulated for convenient administration. Alternatively, the compounds can be dissolved in water, polyethylene glycol, propylene glycol, ethanol, corn oil, cottonseed oil, peanut oil, sesame oil, benzyl alcohol, sodium chloride and / or various buffers. Other adjuvants and modes of administration are sufficiently and widely known in the pharmaceutical art. The appropriate dosages in any given case, of course, depend on the nature and severity of the condition treated, the route of administration and the species of the mammal involved, including its size and individual idiosyncrasy. Carriers, dilutions and representative adjuvants include, for example, water, lactose, gelatin, starch, magnesium stearate, talc, vegetable oils, gums, polyalkylene glycols, petroleum jelly, etc. The pharmaceutical composition can be prepared in solid form, for example in granules, powders or suppositories, or in liquid form, for example solutions, suspensions or emulsions. The pharmaceutical compositions may be subjected to conventional pharmaceutical operations such as sterilization and / or may contain conventional pharmaceutical adjuvants such as preservatives or preservatives, stabilizers, wetting agents, emulsifiers, buffers, etc. For use in the treatment of rheumatoid arthritis, the compounds of this invention can be administered by any convenient route, preferably in the form of a pharmaceutical composition adapted for such day and in an effective dose for the proposed treatment. In the treatment of arthritis, the administration can be conveniently carried out orally or by intra-articular injection in the affected joint. As indicated, the dose administered and the treatment regimen will depend, for example, on the disease, the severity thereof in the patient being treated and his response to treatment, and therefore, may vary widely. The following examples and publications are provided to aid the understanding of the present invention, the true scope of which is set forth in the appended claims. It should be understood that modifications can be made in the processes that are established without departing from the spirit of the invention.

Abbreviations: HNC.MMP-8 = Human neutrophil collagenase, HFC, MMP-1 = Human fibroblast collagenase, HSL-1 = Human stromelysin l, HSL-2 = Human stromelysin 2, HSL-3 human stromelysin 3, HPUMP = human PUMP, H72G = 72 kD human gelatinase, H92G = 92 D human gelatinase, rms = mean square root, HONHiBM-AG-NH2 = H0NH-2-isobutylmalonyl-L-alanyl-glycinamide, MBP-AG-NH2 = 2-benzyl -3-mercaptopropanoyl-L-alanylglycinamide; PLG-NHOH = L-propyl-L-leucyl-glycine-hydroxamate.

EXAMPLE 1 Isolation and purification of the catalytic domain of HNC The catalytic domain of Met (80) -Gly (242) of human HNC is expressed in E. coli and is renatured by dialyzing the inclusion bodies which are solubilized in 6 M urea and 100 M jS-mercaptoethanol, against a buffer that contains 100 mM NaCl, 5 mM CaCl 2, 0.5 mM ZnCl 2, 20 mM Tris / HCl, pH 7.5, as previously described21. The renatured enzyme is subsequently purified to apparent homogeneity determined by SDS-PAGE by affinity chromatography on hydroxamate.

EXAMPLE 2 Synthesis of inhibitors The inhibitors HONH-iBM-AG-NH2 were synthesized (ER029) and MBP-AG-NH2 according to Cushman et al. (1977) 22. PLG-NHOH was synthesized according to Nishino et al. (1978) 13. The inhibitors of the invention can be synthesized as described in Examples 2.1 to 2.4. 2. 1 Isobutylmalonyl-L-alanine-furfurylamide hydroxamate (formula V) Generalities: Solvent systems used: 2E: ethyl acetate: n-butanol: acetic acid: water 5: 3: 1: 1; 6E: ethyl acetate: n-butanol: acetic acid: pyridine 55: 30: 3: 12: 10; 36: cyclohexane: CHC13: acetic acid 45: 45: 10 1) Ter-butyloxycarbonyl-L-alanine furfurylamide (1) To a solution of Boc-Ala-OH (10 g, 53 mmol) and N-methylmorpholine (5.8 ml, 53 mmol) in 250 ml of CH2C12, isobutyl chloroformate (6.3 ml, 53 mmol) was added dropwise at -10 ° C under vigorous stirring. After 7 minutes pre-cooled furfurylamine (7 ml, 74 mmol) is added and the reaction mixture is stirred at room temperature for 12 h. The solvent is evaporated and the residue is partitioned between ethyl acetate and water. The organic phase is washed with 5% KHS04 and 5% NaHC03 and brine, dried over sodium sulfate and evaporated to dryness. The residue is recrystallized from ethyl acetate / hexane. Yield: 12.1 g (85%), homogeneous in c.c.f. (solvent systems: 2E, 36); p.f. 107 ° C; [a] 20D = - 32. 7th; [a] 20546 nn = - 38. 6 ° (c = 1 in MeOH). Analysis calculated for C 13 H 20 N 2 O 4 (268.3): C 58.19, H 7.51, N 10.44; found: C 57.83, H 7.74, N 10.22 By using this method, the following amines can be coupled with tert-butyloxycarbonyl-L-alanine: for example, isopropylamine, butylamine, tert-butylamine, isopentylamine, hexylamine, heptylamine, octylamine , 2-octylamine, cyclohexylamine, aniline, 4-nitroaniline, 4-chloroaniline, benzylamine, 4-chlorobenzylamine, 4-fluorobenzylamine, 2-chlorobenzylamine, l-phenylethylamine, 2-phenylethylamine, 2-piperazin-1-yl-ethylamine , morpholine; 1-naphthylamine, fluorenyl-2-amine, dehydroabietylamine, N- (2-aminoethyl) -morpholine, (aminomethyl) pyridine, 3- (aminomethyl) pyridine, etc. 2) L-alanine-furfurylamide hydrochloride (2) Boc-Ala-Fur (2 g, - 7.5 mmol) is dissolved in 1.4 M HCl / ethyl acetate. The solution is maintained at room temperature for 1 h; then the solvent is evaporated and the residue is re-evaporated twice from toluene and finally dried in KOH pellets.

Yield: 1.5 g (100%); homogeneous based on c.c.f. (solvent systems: 2E, 36); [I heard] 20D = + 10.3 °; [α] 2054β nn = + 12.3 ° (C = 1 in MeOH); MS-BAR: 169-1 [M + H] +. "H-NMR (MeOD): the spectrum is consistent with the structure. Analysis calculated for CβH13N202Cl (204.66): C 46.95, H 6.40, N 13.69; Found: C 46.20, H 6.62, N 13.29. 3) Diethyl isobutylmalonate (3) Diethyl malonate (80 g, 0.5 mole) is dissolved in a freshly prepared solution of sodium (11.5 g) in ethanol (500 ml). Then isobutyl bromide (71.5) is added dropwise; 0.52 moles) under vigorous stirring. The reaction mixture is kept under reflux until the pH is almost neutral (5-6 h). The insoluble material is separated by filtration and the reaction mixture is evaporated. The residue is distributed between water and ether, and the organic phase is washed with water and dried with Na 2 SO 4. The solvent is evaporated and the residue is distilled under vacuum to provide the title compound as a liquid: Yield: 76 g (70%); homogeneous in c.c.f. (solvent system: 2E, 63); MS-EI: 217; Analysis calculated for (216.3): C 61.07 H 9.33; found: C 60.50 H 9.54. In addition to the commercially available diethyl benzylmalonate, diethyl ethoxymethylmalonate and diethyl phenylmalonate, following this process other diethyl malonic acid esters are prepared using, for example, 1-bromobutane, 2-bromobutane, 1-bromohexane, 2-bromohexane, 1- bromoheptane, 3- (bromomethyl) heptane, 1-bromononane, benzyl bromide, bro-cyclohexane, 3-bromo-1-propanol, 2-bromo-4'-methoxyacetophenone, 2-ethoxyethyl bromide, 2-bromoacetophenone, N-bromomethylphthalimide . 4) Potassium salt of ethyl isobutylmalonic acid (4) Diethyl isobutylmalonate (1.2 g; . 6 mmoles) in 5 ml of ice-cooled ethanol, containing 5.6 mmoles of KOH. After 2 h, the solution is evaporated to a small volume and the title compound is precipitated with hexane. Yield: 1.2 g (95%). -RMN (MeOD): the spectrum is consistent with the structure of the title compound. Analysis calculated for C9H1S04K (226.31): C 47.77 H 6.68; found: C 45.89 H 7.93.

) Isobutylmalonyl-L-alanine-furfurylamide (5) To a cooled solution of compound 4 (0.53 g, -2.2 mmol) in 20 ml of CH2C12 is added oxalyl chloride (0.38 ml, 4.4 mmol) and after 2 h at room temperature the solvent evaporates. The residue is again evaporated from CH2C12 and finally dissolved in CH2C12 and added to a solution of 2 (0.45 g, 2.2 mmoles) in CH2C12 containing triethylamine (0.61 ml, 4.4 mmoles). The reaction is allowed to take place overnight at room temperature, then the solvent is evaporated and the residue is partitioned between ethyl acetate and water. The organic phase is washed with 5% KHS04 and 5% NaHC03 and brine. The ethyl acetate phase is dried over MgSO 4 and evaporated. The residue is dissolved in 5 ml of ethanol containing 2.2 mmoles of KOH. After 1 h the solvent is evaporated and the residue is distributed between ethyl acetate and 5% KHS04. The organic phase is washed neutral, dried over Na 2 SO 4 and evaporated. Yield: 0.575 g (84%); homogeneous determined by c.c.f. (solvent system: 2E, 36); EM-BAR: [M + H] * = 311.2; 'H-NMR (MeOD): consistent with the structure. Analysis calculated for C1SH22N20S (310.2): C 58.04; H 7.15; N 9.03; found: C 57.77; H 7.32; N, 8.89. 6. Isobutylmalonyl-L-alanine-furfurylamide hydroxamate (6) Compound 5 (0.400 g, 1.3 mmol) is reacted in tetrahydrofuran with N-hydroxysuccinimide (0.148 g, 1.3 mmol) and dicyclohexylcarbodiimide (0.268 g, 1.3 mmol) in an ice bath for 5 hours. The dicyclohexylurea is filtered off and hydroxylamine "HCl (0.181 g, 2.6 mmol) is added to the filtrate with triethylamine (0.36 ml, 2.6 mmol) in dioxane / water. The reaction is allowed to take place overnight at room temperature. The solvent is evaporated and the residue is distributed between water and ethyl acetate. The organic phase is washed with 5% KHS04, water and dried. The solution is concentrated and the residue is precipitated with petroleum ether. Yield: 0.302 g (72%), homogeneous determined by c.c.f. (solvent system: 2E; 36). EM-BAR: [M + H] + = 326.1. Analysis calculated for C1SH23N30S (325.2): C 55.36, H 7.13, N 12.92; found: C 55.87, H 7.32, N 12.67. 2. 2 2-Isobutyl-3-carbonyl-3'- (4-acetylaniline) -propionic acid (7) (Formula VI) To a cooled solution of 4 (1.0 g, - 5.3 mmol) in CH2C12, oxalyl chloride (0.72 ml, 10.6 mmol) is added and after 2 h at room temperature, the solvent is evaporated. The residue is dissolved in CH2C12 and evaporated to remove excess oxalyl chloride. The acid chloride is dissolved in CH2C12 and added dropwise with stirring to aluminum chloride containing CH2C12. Then a solution of acetanilide (0.72 g, 5.3 mmol) is added and the reaction mixture is maintained at 20 ° C by cooling. The reaction mixture is treated with ice and after acidification with dilute H2SO4, the CH2C12 phase is separated and washed with water, dried and concentrated to a small volume. The product is precipitated with petroleum ether. Yield: 0.92 g (57%); EM-BAR: [M + H] * = 306.2. The monoethyl ester (0.80 g, 2.6 mmol) is saponified in ethanol containing KOH (1 equivalent) and after 2 h the solvent is evaporated and the residue is distributed between ethyl acetate and KHS04. The organic layer is washed with water, dried over MgSO4 and concentrated to a small volume. The title compound is isolated by the addition of petroleum ether. Yield: 0.69 g (95%); EM-BAR: [M + H] + = 278.3.

Analysis calculated for C1SH1904N (277.1): C 64.95 H 6.91 N 5.05; found: C 63.67 H 7.02 N 4.99. 2. 3 Analysis calculated for N-benzyloxycarbonyl-a-phosphonoglycyl-L-alanine furfurylamide (8) (Formula VID The trimethyl ester of N-benzyloxycarbonyl-a-phosphonoglycine (1.46 g, 4.4 mmol) is completely deprotected with concentrated HCl, according to Balsiger et al. [(1959) J. Org. Chem. 24, 434] and the free amino function is again protected with low benzoylcarbonyl chloride with the Schotten-Baumann conditions. The hydrochloride is prepared with thionyl chloride according to Balsiger et al. (1959) J. Org. Chem. 24, 434, and dioxane (20 ml) is reacted with compound 2 (0.9 g, 4.4 mmol) in the presence of triethylamine (4 equivalents). After 4 hours at room temperature, the solvent is separated and the residue is distributed between ethyl acetate and KSH04. The organic phase is washed with water, dried over MgSO 4 and evaporated. The residue is triturated with ether / petroleum ether and filtered off. Yield 0.735 g (38%); EM-BAR: [M + H] = 439.1.

Analysis calculated for ClßH22N30aP (439.4): C 49.21 H 5.05 N 9.56; found: C 48.95 H 5.31 N 9.43. 2. 4 Synthon for phosphonic and phosphinic acid derivatives, according to Formula III The synthon (formula VIII) can be obtained by methods known in the literature reviewed by Houben-Weyl, Methoden der Organischen Chemie, Vol. 12/1 and E2. Their coupling to the groups Y for example, to the compound 2 is obtained by classical methods of peptide synthesis and the saponification of the methyl ester is carried out with KOH in ethanol.

EXAMPLE 3 Crystallization The crystallizations were carried out in a hanging drop steam diffusion at 22 ° C. The droplets were made by mixing 1.8 μl of a 10 mg / ml solution of HNC in 3 mM of Mes / NaOH, 100 mM NaCl, 5 mM CaCl2 and 0.02% NaN3 at pH 6.0, 2 μl of approximately MBP-AG-NH2 90 M or a solution of HONHiBM-AG-NH2 and 6 μl of PEG 6000 solution (10% m / v in 0.2 M Na / NaOH at pH 6.0). The droplets were concentrated against a reservoir buffer consisting of 0.8 M potassium phosphate buffer (with MBP-AG-NH2) and 1.0 M (with HONHiBM-AG-NH2), 0.02% NaN3 at pH 6.0. The crystals of size 0.66 x 0.10 x 0.03 mm (HNC with MBP-AG-NH2) and 0.90 x 0.12 x 0.02 mm (HNC with HONHiBM-AG-NH2) are obtained in the following 3 days and are collected in PEG 6000 at 20% (m / v), 0.5 M NaCl, 0.1 M CaCl2, 0.1 M Month / NaOH, 0.02% NaN3, pH 6.0, containing 10 mM of the corresponding inhibitor. The crystals belong to the group of orthorhombic space ^ 2 2 and show constant networks a = 33.24 / 33.13, b = 69.20 / 69.37, c 72.33 / 72.31 A, a = ß =? = 90 ° (HNC with MBP-AG-NH2 / HNC with HONHiBM-AG-NH2) and are very similar to the original crystals of Met (80) -Gl (242) collagenase containing PLG-NHOH9. The asymmetric unit contains a monomer.

EXAMPLE 4 Structure analysis X-ray data are collected at a MAR image area detector (MAR Research, Hamburg) mounted on a Rigaku rotating anode X-ray generator (λ = 1.5418 Á, operated at 5.4 kW). The X-ray intensities are evaluated with the MOSFLM23 program package, and all X-ray data is loaded with PROTEIN24. The data collection statistics for the two complexes are given in Table 1 and compared with the data previously obtained from the HNC complex with PLG-NHOH. A 2Fo-Fc electron density map is calculated using all the reflection data (Table 1) and the 2.0 Á model of the Met (80) -Gly (242) form of HNC9 for phase formation. The non-peptide portions of the inhibitors were constructed with the ENIGMA program (a molecular graph program provided by ICI Wilmington), and the complete inhibitor models were fitted to the electronic density map using the interactive graphics program FRODO25. The complexes were subjected to least squares refinement of reciprocal space with energy constraints as implemented in X-PLOR26 using strength field parameters derived by Engh and Huber27. These refined models were compared with their improved density, reconstructed and refined until convergence. A patch residue with junction and energies at angles close to zero and including the central zinc and three surrounding HisNe2 atoms together with both hydroxamate oxygens (in the case of the HONHiBM-AG-NH2 complex) was defined for zinc at the active site . The other three metals were treated as previously described in the PLG-NHOH9 structure. The water molecules previously observed in the PLG-NHOH structure were partially retained and additional water was introduced at stereochemically reasonable positions if the appropriate density is present in the maps calculated without these molecules, and contoured to is. In the last stage of refinement the individual temperature factors were refined without any restriction. The final R factor is 0.17 / 0.16. The final refining statistics of the two HNC complexes are shown in Table 2 and compared with previous data obtained with the PLG-NHOH complex.

EXAMPLE 5 Pro-Leu-Gly-NHOH binding (PLG-NHOH) The peptide chain of PLG-NHOH binds to the HNC border strand in a slightly twisted antiparallel manner that forms two interprincipal chain hydrogen bonds with Leu (12) and Ala (163). The N-terminal Pro (II) fits in the hydrophobic pocket formed by the terminal chains of His (162), Phe (164) and Ser (151) with the Pro ring approximately parallel to the benzene ring of Phe (164) and perpendicular to the ring. His imidazole group (162). The imino nitrogen points point to the volume water and the site of a P4 residue. The side chain of Leu (12) is housed in, but not filled in, a narrow slot delineated by His (210), Ala (206) and His (207). The hydroxamic acid group (RCONHOH) is in the cis configuration and is protonated due to the unambiguous relationship of NH and OH in the hydrogen bonds with the protein groups. The hydroxyl oxygen of hydroxamate binds to zinc and forms a favorable hydrogen bond (2.6 Á) with one of the oxygens (Oei) of the carboxylate group Glu (l98). The N-H forms a hydrogen bond (3.0 Á) with the carbonyl group of the residue of the Ala border strand (161) '. Catalytic forms of zinc with one octahedron layer with both hydroxamate oxygens. His (197) Ne2 and His (207) Ne2 form an almost tetragonal plane, with the other histidine His (201) Ne2 at oxygen distances and nitrogen-zinc are between 1.9 and 2.2 Á and the deviation of the average angle from the octahedron with ideal lid is only 10 ° (see Table 3). The zinc ion is not exactly in the plane defined by the four different hydrogen atoms of hydroxamic acid, but is 0.7 A behind it. This suggests a non-optimal orbital interaction with zinc probably caused by steric constraints on the peptide portion of the inhibitor.

Two-thirds of the solvent accessible surfaces of the free inhibitor are removed by complex formation despite incomplete peptidyl chain adjustment. This is surprising at first examination and may be due to the gaps between the protein surface and the inhibitor, which are too small to allow penetration of the solvent probe. The little complementarity of the enzyme and the inhibitor may explain the relatively weak affinity of PLG-NHOH for collagenase and suggest that there are ways to improve the structure by medical chemists.

EXAMPLE 6 Union of HS-CH2-S / R-CH (Bzl) CO-L-Ala-Gly-NH2 ("MBP-AG-NH2") In the MBP-AG-NH2 complex with the collagenase active site the sulfur atom is the fourth ligand of the catalytic zinc and together with the three imidazole nitrogens of His (197), His (201) and His (207) form a almost exact tetrahedron with a deviation of only 6.2 ° (Table 3). The refined sulfur metal distance is 2.3 Á which is slightly larger than the average distance of 2.1 Á29 for distances Zn-S in proteins. The Ne2-Zinc distances (between 1.9 and 2.3 Á, Table 4) are only slightly changed, compared to the PLG-NHOH structure. The diol group is supposed to coordinate with zinc in its non-cationic form since coordination with catalytic zinc with positive charge would divert pK to lower values. The peptide chain of the inhibitor binds to the "right" of the cleavage of the enzyme in an extruded geometry ("FI1 = -179o", "* I1 = 109 °, FI2 = -89 °, * I2 = + 147 °, FI3 = -93 °) The chain of the inhibitor is almost antiparallel to the voluminous segment Gly (158) -He (159) -Leu (l60), is parallel to the crossing on the forming segment of the wall SI 'Pro (217) - Asn (218) -Tyr (219), is under two rungs with ladder with the first (Phe (Il) O ** «Leu (l60) N: 2.8 A, Gly (13) N» «» Gly (158) O: 3.0 Á) as well as with the last segment (Ala (I2) N »» * Pro (217) 0: 3.0 Á, Ala (I2) 0 «« * Tyr (219) N: 2.8Á). dominant between the inhibitor and the enzyme are hydrophobicly made by the phenyl side chain and the central hydrophobic portion of the SI 'pocket which is flanked by the His (197) imidazole and the Glu (198) carboxylate group (left) ), Val (194) (in the back), the phenolic side chain of Tyr (219) (to the right) and the segment or of the main chain (Pro (217) -Asn (218) -Tyr (219).

The refined electronic density unambiguously shows that the S stereoisomer, which corresponds to the amino acid analog L, binds preferentially from the diastereomeric mixture of inhibitor. The Ca-CJ junction is (according to XI = -152 °) an essentially trans geometry with the amino group trans with respect to C ?. This SI bag is wider than required to accommodate any lateral amino acid side chains that can in fact bind tricyclic compounds (see below). Therefore, the phenyl group of the inhibitor occupies only 1/2 or 1/3 of the internal volume of SI ', which leaves space for three ordered solvent molecules which are in sites similar to those observed in the free enzyme. These "internal" water molecules are in partial contact with the phenyl group and are interconnected by hydrogen bonds, with themselves or with hydrogen bond acceptors or donors provided by surrounding protein groups (Wing (220) N, Leu (214) 0, Leu (193) 0). The side chain of the Ala points (12) from the collagenase surface and the larger lateral chain are likely to be housed along the shallow surface of the groove running through the bulky segment between Gly (158) and Ile (159). This last residue (He (159)) which is not conserved in the MMPs is probably the result of the differences in specificity between the MMPs and may offer an attractive target for the design of selective inhibitors. The residue Gl (13) is located between both crosslinking segments Pro (217) -Tyr (219) and Gly (158) -Leu (160). A larger side chain could collide with the enzyme and would require a rearrangement of the inhibitor chain. Additional residues can be placed on the flat molecular surface in which they could find various fixation points for polar interactions. In summary, the thiol group and the "first residue" are involved in a large number of intimate contacts that result in a considerable reduction of the accessible solvent surface when the bonding occurs. The Wing (12) and Gly (13) run along the active site slot, with their side chain positions extending into the volume solvent. The peptide binding geometry and the confirmation of MBP-AG-NH2 are similar to other "primed site inhibitors" shown to bind to other collagenases18'19.

EXAMPLE 7 Union of HONHC (O) -R, S-CH (iButil) CO-L-Ala-Gly-NH2 (HONHiBM-AG-NH2) The inhibitor of HONHiBM-AG-NH2 binds unexpectedly in a different way than anticipated from its design and binding mode in thermolysin. The hydroxamate subgroup evidently interacts with catalytic zinc in a favorable, bidentate manner, with its two oxygen atoms and three ligand histidines that form a trigonal bipyramidal coordination sphere with the catalytic zinc, but in contrast, its isobutyl "side chain" remains outside the SI bag, probably due to the severe restrictions imposed by the adjacent flat hydroxamate group. Instead, the C-terminal Ala-Gly-amide tail adopts a bent conformation and inserts into the SI bag, probably in a non-optimized manner. Both the isobutyl side chain and the C-terminal peptide tail can be replaced by other groups that fit better. Although this inhibitor inhibits MMPs very little, the inhibitors according to the invention, which are based on this structure, are unexpectedly very powerful MMP inhibitors. HONHiBM-AG-NH2 is used in this example as a model substance for binding studies of the inhibitors according to the invention. The hydroxyl oxygen His (197) Ne2 and His (207) Ne2 of the central trigonal plane around the zinc and the carbonyl oxygen of His (201) Ne2 occupies both vertices. The average angular deviation of the ideal geometry is 13.0 ° (see Table 3). Both the N-0 and the carbonyl group of the hydroxamic acid portion form a common plane with the catalytic zinc. As in the PLG-NHOH complex, the nitrogen hydroxamate is close to Ala (161) 0 (2.9 Á) and is favorably placed to form a hydrogen bond and consequently this hydroxamic acid is also modeled in its protonated form. Due to the interaction of the hydroxamate group and zinc, the "side chain" R2 (for example isobutyl) is not capable of being inserted into the SI 'pocket and remains on the outer surface of the collagenase cleft, exposed to the solvent. The electronic density which justifies this side chain is deviated outwards, towards the periphery, which indicates some increased mobility and is loosely placed in the groove formed by the bulky segment and the adjacent edge strand. The S stereoisomer fits much better, with its "side chain" placed in a gauche conformation "Cß-C?." It is put to the next carbonyl group.

In contrast to conventional "primed site inhibitors" 18,19, the segment of the peptide L-Ala (I2) -Gly (13) -NH2 binds in a curved conformation instead of an extended geometry. The carbonyl following the CH group (iButyl) and the carbonyl group Ala (12) form a hydrogen bond respectively with Leu (160) N of the bulky segment and Tyr (219) N of the wall-forming segment. Despite the slight rotation of the group CH (iBut) (II) -Ala (12) amide outside the trans conformation, Ala (I2) NH along Ala (12) in MBP-AG-NH2 is able to form a hydrogen bond with Pro (217) O. However, Ala (12) has adopted a 310-helical conformation (with F = -78 °, * = -4o) and the next peptide group is oriented almost opposite to the amide bond. -13 of conventional "primed site inhibitors". As a consequence of this curved conformation, the amino group and the first R3 atom are located in the bottleneck of the SI 'bag in van der Waals type contact with Glu (198)? Ei, His (197) imidazole and the side chain of Val (194), but with the amino group lacking any hydrogen-binding acceptor. In contrast to the total binding of the inhibitors of the state of the art, the side chains of R2 (for example, iButyl (Il)) and Zx to Z3 (for example Ala (I2)), remain essentially exposed to water and the residue of tail C has almost completely been removed from contact with it.

EXAMPLE 8 Synthesis of additional inhibitors Abbreviations: OSu: N-hydroxysuccinimide ester ONp: p-nitrophenyl ester iBM: 2-isobutylmalonic acid Bn: benzyl Z: benzylcarboxy Boc: t-butylcarboxy homphe: homophenylalanine (-t -) BnONH-iBM-OEt (8.1) O-benzylhydroxylamine hydrochloride is suspended (4.79 g, 30 mmol) in 50 ml of THF and sodium methylate (1.62 g 30 mmol) is added under stirring. After 10 minutes, the solvent is completely evaporated to remove the methanol. The residue and Et-0-iBM-0"K + (6.78 g, 30 mmol) are suspended in 50 ml of THF.The suspension is cooled to 0 ° C and l-ethyl-N '- (3- hydrochloride is added. dimethylaminopropyl) -carbodiimide (EDCI) (6.34 g, 33 mmol) The reaction mixture is stirred for 12 h while it is warmed to room temperature The solvent is evaporated, the residue is dissolved in 150 ml of ethyl acetate. The organic phase is washed three times with 30 ml of 5% KHS04, three times with 30 ml of 5% NaHC03 and with 30 ml of water, dried over MgSO4 and evaporated The untreated oily product is purified by column chromatography on 100 g of silica gel (0.040-0.063 mm particle size), eluent, ethyl acetate: n-hexane / l: 2 to provide 7.45 g (84.6%) of a colorless oil Pure product determined by ccf Rf: 0.26, ethyl acetate: n-hexane / l: 2.? -RMN (dß-DMSO): The spectrum is consistent with the structure.

(+ -) BnONH-ÍBM-OH (8.2) Dissolve 8.1 (3.53 g, 12 mmol) in a mixture of 10 ml of THF and 10 ml of methanol. A solution of sodium hydroxide (1.44 g, 36 mmol) in 2 ml of water is added, with stirring, and the reaction mixture is heated at 50 ° C for 1 h. The reaction mixture is diluted with 50 ml of methanol. 10 g of Amberlyst 15 (strongly acidic cation exchanger, form H + 4.6 mmol / g) are added under cooling with ice, and the mixture is stirred for 15 minutes. The cation exchanger is filtered off, washed with methanol and the filtrate is evaporated to dryness. 3.20 g (100%) of product are obtained as fine, pure needles, determined by c.c.f. Rf: 0.61, acetonitrile: water / 4: 1. 'H-NMR (dß-DMS0): The spectrum is consistent with the structure.

Z-Ala-NHBn (8.3) Z-Ala-OSu (6.40 g, 20 mmol) is dissolved in 300 ml of ethyl acetate, benzylamine (2.75 ml, 25 mmol) is added and the mixture is stirred for 1 h. The solution is washed three times with 30 ml of 5% KHS04, three times with 30 ml of 5% NaHCO3 and 30 ml of water, dried over MgSO4 and evaporated to dryness. 5.59 g (90%) of the product are obtained as a colorless, pure powder determined by c.c.f. Rf: 0.17, ethyl acetate: n-hexane / l: 2, m.p. 140 ° C. The -RMN (ds-DMSO): The spectrum is consistent with the structure.

H-Ala-NHBn (8.4) 8.3 (0.63 g, 2.0 mmol) is dissolved in 20 ml of methanol, 100 mg of 10% Pd / C catalyst is added in a slow stream of H2 which is allowed to run through the solution for 20 minutes. The catalyst is removed by filtration and washed. The filtrate is evaporated and the residue is used without further purification for the next coupling fraction. 2 Diastereomers of BnONH-iBM-A. -NHBn (8.5) Dissolve 8.4 (+ -) BnONH-iBM-OH 8.2) (0.27 g, 1.0 mmol) and hydroxybenzotriazole (136 mg, 1.0 mmol) in 10 mL of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (0.20 g, 1.05 mmol) is added. The reaction mixture is stirred for 12 h while heating to room temperature. The solvent is evaporated, the residue is dissolved in 100 ml of ethyl acetate. The organic phase is washed three times with 15 ml of 5% KHS04, three times with 15 ml of 5% NaHCO3 and 15 ml of water, dried over MgSO4 and evaporated. The product precipitates with ether / ethyl acetate. 0.26 g (61%) of product are obtained as a colorless, pure powder determined by c.c.f. Rf: 0.65, chloroform: methanol / 9: l. The '• H-NMR (ds-DMSO): The spectrum is consistent with the structure, two diastereomers are observed. 2 diastereomers of HONH-iBM-Ala-NHBn (ER014) (8.6) 8.5 (110 mg, 0.26 mmol) is dissolved in 10 ml of methanol, 50 mg of 10% Pd / C catalyst is added and a slow stream of H2 is poured through the solution for 20 minutes. The catalyst is removed by filtration and washed. The filtrate is evaporated and the product is precipitated with ether. 80 mg of product (92%) are obtained as a pure colorless powder determined by c.c.f. Rf: 0.37, chloroform: methanol / 9: 1. '? -NRM (ds-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (40:60).

Z-Asn-NHBn (8.7) Z-Asn-ONp (7.75 g, 20 mmol) is dissolved in 100 ml of THF, benzylamine (2.25 ml, 20.5 mmol) is added and the mixture is stirred for 2 h. The precipitated product is washed with 50 ml of THF, 100 ml of diethyl ether, 300 ml of NaHC035%, 100 ml of water and 100 ml of THF. The product is dried in vacuo. 3.94 g of product (55%) are obtained as a pure colorless powder determined by c.c.f. R £: 0.57, chloroform: methanol / 9: 1. p.f. 205-208 ° C. 'H-NMR (d6-DMS0): The spectrum is consistent with the structure.

H-Asn-NHBn (8.8) Z-Asn-NHBn (0.36 g, 1.0 mmol) is deprotected as described for 8.4 2 diastereomers of BnONH-iBM-Asn (8.9) 8.8 (+ -) BnONH-iBM-OH 8.2) (0.27 g, 1.0 mmol) and hydroxybenzotriazole (135 mg, 1.0 mmol) are dissolved in 10 ml of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (0.20 g, 1.05 mmol) is added. The reaction mixture is stirred for 12 hours while heating to room temperature. The solvent is evaporated and the solid residue is washed in a glass frit with 150 ml of 5% KHS04, 150 ml of 5% NaHCO 3 and 150 ml of water. The product is milled with ether, 0.29 g of product (62%) are obtained as a pure colorless powder determined by c.c.f. Rf: 0.27, chloroform: methanol / 9: l. 'H-NMR (d6-DMSO): The spectrum is consistent with the structure, two diastereomers can be observed. 2 diastereomers of HONH-iBM-Asn-NHBn (ER017) (8.10) 8.9 (0.20 g, 0.43 mmol) is dissolved in 10 ml of methanol, 50 mg of 10% Pd / C catalyst is added and a slow stream of H2 is supplied through the solution for 20 minutes. The catalyst is removed by filtration and washed. The filtrate is evaporated and the product is precipitated with ether. 80 mg of the product (92%) are obtained as a pure colorless powder, determined by c.c.f. Rf: 0.61 acetonitrile: water / 4: 1. 'H-NMR (ds-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (34:66).

Z-Ser-NHBn (8.11) Z-Ser-OH (4.78 g, 20.0 mmoles) benzylamine (2.75 ml, 25 mmoles) and hydroxybenzotriazole are dissolved. (2.70 g, 20.0 mmol) in 50 ml of THF. The suspension is cooled to 0 ° C and (EDCI) hydrochloride (4.23 g, 22.0 mmol) is added, the reaction mixture is stirred for 12 h while heating to room temperature. The solvent is evaporated, the residue is dissolved in 150 ml of ethyl acetate. The organic phase is washed three times with 30 ml of 5% KHS04, three times with 5% NaHCO3 and 30 ml of water, dried over MgSO4 and evaporated to dryness. 5.10 g of the product (71%) are obtained as a pure colorless powder according to c.c.f. Rf: 0.44, chloroform: methanol / 9: 1 p.f. = 153 ° C.

H-Ser-NHBn (8.12) 8.11 is unprotected (0.66 g, 2.0 mmoles) as described for 8.4. 2 diastereomers of BnONH-iBM-Ser-NHBn (8.13) 8.12, 8.2 (0.53 g, 2.0 mmoles) and hydroxybenzotriazole (0.27 g, 2.0 mmoles) are dissolved in 10 ml of THF. The suspension is cooled to 0 ° C and l-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (0.40 g, -2.1 mmol) is added. The reaction mixture is stirred for 12 hours while heating to room temperature. The treatment procedure is carried out as described for 8.5. The product is precipitated with ether / ethyl acetate. 0.72 g (82%) of product are obtained as pure colorless powder according to c.c.f. Rf: 0.48, chloroform: methanol / 9: 1. 2 diastereomers of HONH-iBM-Ser-NHBn (ER028) (8.14) 8.13 (0.25 mg, 0.57 mmol) is deprotected as described for 6. 190 mg of product (95%) is obtained as a colorless, pure powder, based on c.c. Rf: 0.16, chloroform: methanol / 9: 1. 'H-NMR (d6-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (28:72).

Boc-Asn-NHBn (m-N02) (8.15) Dissolve 3-nitrobenzylamine hydrochloride (0.943 g, - 5 mmol) and triethylamine (0.84 ml, 6 mmol) in 50 ml of THF, add Boc-Asn-ONp (1.77 g, 5 mmol) and stir for 2 h . The solvent is evaporated and dissolved in 200 ml of ethyl acetate. The solution is washed three times with 30 ml of 5% KHS04, three times with 30 ml of 5% NaHCO3 and 30 ml of water, dried over MgSO4 and evaporated to dryness. 1.15 g of product (63%) are obtained as a pure colorless powder, based on c.c.f. Rf: 0.34, chloroform: methanol / 9: l. p.f. 190-191 ° C. ? -RMN (d6-DMS0): The spectrum is consistent with the structure.

H-Asn-NHBn (m-N02) * HCl (8.16) 8.15 (0.73 g, - 2.00 mmol) is suspended in 10 ml of a 4 M solution of hydrochloride in dioxane and stirred for 12 h at room temperature. The deprotected and precipitated product is collected on a filter and washed in diethyl ether.

BnONH-iBM-Asn-NHBn (m-N02) (8.17) Dissolve 8.16, 8.2 (0.54 g, 2.0 mmol) and hydroxybenzotriazole (0.30 g, 2.0 mmol) in 20 ml of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (0.4 g, 2.1 mmol) is added. With N-methylmorpholine (0.22 ml, 2 mmol) the solution is brought to pH 6-7. The reaction mixture is stirred for 12 h while heating to room temperature. The treatment procedure is carried out as described for 8.5. 0.48 g of product (40%) is obtained as a pure colorless powder based on c.c.f. Rf: 0.32, chloroform: methanol / 9: 1. 2 diastereomers of HONH-iBM-Asn-NHBn (m-NH2) (ER031) (8.18) Dissolve 8.17 (0.26 g, 0.50 mmol) and 0.5 ml of IN hydrochloric acid in 10 ml of methanol, add 200 mg of 10% Pd / C catalyst and administer a slow stream of H2 through the solution for 10 h . The catalyst is removed by filtration and washed. The filtrate is evaporated and the product is precipitated with ether. 200 mg of the product (92%) are obtained as a pure colorless powder based on c.c.f. Rf: 0.13, chloroform: methanol / 4: 1. ? -RMN (ds-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (44:56).

Z-Gly-NHBn (8.19) Z-Gly-OSu (6.13 g, 20.0 mmoles) and benzylamine (2.30 ml, 21.0 mmoles) are transformed as described for 8. 3. 5.34 g (90%) of colorless product, pure based on c.c.f. R £: 0.46, chloroform: methanol / 9: l. p.f. = 117 ° C.

Z-Ser-Gly-NHBn (8.20) 8.19 (1.49 g, 5.0 mmol) is deprotected as described for 8.4. The residue, Z-Ser-OH (1.20 g; . 0 mmole) and hydroxybenzotriazole (0.6 g, 5.0 mmole) are dissolved in 10 ml of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (1.0 g, 5.3 mmol) is added. The reaction mixture is stirred for 12 h while heating to room temperature. The treatment procedure is repeated as described for 8.5. The product is precipitated with ethyl acetate. 1.28 g of product (66%) are obtained as a pure colorless powder based on c.c.f. Rf: 0.38, chloroform: methanol / 9: l. p.f. = 170 ° C. 2 diastereomers of BnONH-iBM-Sßr-Gly (8.21) 8.20 (193 mg, 0.5 mmol) is deprotected as described for 8.4. The residue, 2 (133 mg, 0.5 mmol) and hydroxybenzotriazole (70 mg, 5.0 mmol) is dissolved in 10 ml of THF. The suspension is cooled to 0 ° C and (EDCI) hydrochloride (100 mg, 0.6 mmol) is added. The reaction mixture is stirred for 12 h while heating to room temperature. The treatment procedure is carried out as described for 8.5. 190 mg of product (76%) are obtained as a pure colorless powder based on c.c.f. Rf: 0.27, chloroform: methanol / 9: 1. 2 diastereomers of HONH-iBM-Ser-Gly-NHBn (ER059) (8.22) 8.21 (190 mg, 0.38 mmol) is deprotected as described for 8.6. The product is precipitated with diethyl ether. 120 mg of product (77%) are obtained as a pure colorless powder based on c.c.f. Rf: 0.57, acetonitrile: water / 4: 1. 'H-NMR (d6-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (42:58).

Z-Homophe-NHCH3Ph (p-Me) (8.23) Dissolve Z-Homophe-OH (157 mg, 0.5 mmol), 2- (p-tolyl) ethylamine (68 mg, 0.5 mmol) and hydroxybenzotriazole (70 mg, 0.5 mmol) in 3 mL of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl-carbodiimide hydrochloride (EDCI) (100 mg, 0.53 mmol) is added.The reaction mixture is stirred for 12 h while heating up The ambient temperature The treatment procedure is carried out as described for 8.5, 200 mg of product (93%) are obtained as a pure colorless powder based on ccf, Rf: 0.23, n-hexane: ethyl acetate / 2: l. 2 diastereomers of BnONH-iBM-Homophe-NHCH2Ph (p-Me) (8.24) Z-Homophe-NHCH2CH2Ph (p-Me) (200 mg, 0.46 mmol) is deprotected as described for 8.4. The residue, 8.2 (132 mg, 0.5 mmol) and hydroxybenzotriazole (70 mg, 5.0 mmol) are dissolved in 6 ml of THF. The suspension is cooled to 0 ° C and 1-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (100 mg, 0.6 mmol) is added. The reaction mixture is stirred for 12 h while heating to room temperature. The treatment procedure is carried out as described for 8.5. 200 mg of product (80%) are obtained as a colorless, pure powder based on c.c.f. Rf: 0.3823, n-hexane ethyl acetate / 1.2. 2 diastereomers of HONH-iBM-Homophe-NHCH2CH2Ph (p-Me) (ERO070) (8.25) 8.24 (190 mg, 0.35 mmol) is deprotected as described for 8.6. The product is precipitated with n-hexane. 140 mg of product (88%) are obtained as a colorless, pure powder based on c.c.f. R £: 0.38, chloroform: methanol / 9: 1. 'H-NMR (ds-DMSO): The spectrum is consistent with the structure, the two diastereomers have the ratio (23:77).

Z-Phe-NHBn (8.26) Z-Phe-OSu (1.98 g, 5.0 mmoles) and benzylamine (0.60 ml, 5.5 mmoles) are transformed as described for 3. 1.6 g of pure colorless product (95%) are obtained based on c.c. Rf: 0.59, ethyl acetate: n-hexane / 2: 1.

H-Pro-NHBn (8.27) 8.26 (155 mg, 0.40 mmol) is deprotected as described for 8.4. 2 diastereomers of HONH-iBM-Phe-NHBn (ER074) (8.28) 8.27, 8.2 (100 mg, 0.37 mmol) and hydroxybenzotriazole (60 g, 0.40 mmol) are dissolved in 5 ml of THF. The solution is cooled to 0 ° C and l-ethyl-N '- (3-dimethylaminopropyl) -carbodiimide hydrochloride (EDCI) (0.20 g, 1.0 mmol) is added. The reaction mixture is stirred for 12 h while heating to room temperature. The treatment procedure is carried out as described for 8.5. Pure BnONH-iBM-Phe-NHBn based on c.c.f. (Rf: 0.27, ethyl acetate: n-hexane / l: 1) is deprotected as described for 8.6. 161 mg of product (98%) are obtained as a pure, colorless powder based on c.c.f. Rf: 0.41, chloroform: methanol / 9: 1.

EXAMPLE 9 Determination of the inhibitory effect (enzyme assay) In order to determine the inhibition of MMPs, for example HNC, the catalytic domain is incubated (for isolation and purification see Example 1) with inhibitors having various concentrations. Subsequently, the initial reaction rate is measured in the conversion of a standard or conventional substrate in a manner analogous to F. Grams et al. (1993) 5 '. The results are evaluated by plotting the record or the reaction rate against the concentration of the inhibitor. The inhibition constant (Ki) is obtained as the negative section of the abscissa by the graphical method according to M. Dixon (1953) 28. The synthetic collagenase substrate is a heptapeptide which is coupled, in the C-terminal portion with DNP (dinitrophenol). The DNP residue suspends, by steric hindrance, the fluorescence of the adjacent heptapeptide tryptophan. After the breakdown of a tripeptide, which includes the DNP group, the fluorescence of tryptophan is increased. The proteolytic cleavage of the substrate can be measured by the fluorescence value. a) First method The test is carried out at 25 ° C in freshly prepared 50 mM Tris buffer (pH 8.0) treated with dithiozone to remove traces of heavy metals. 4 mM CaCl 2 is added and the buffer is saturated with argon. Adamalysin II concentrated solutions are prepared by centrifugation of the protein from a suspension of ammonium sulfate and subsequent dissolution in the assay buffer. The concentrated collagenase solutions are diluted with the assay buffer. The enzyme concentrations are determined by uv measurements (e2ß0 = 2.8 • 104 M "1 • cm" 1, e2ßß: 2.2 • 104 M "1 cm" ') and the concentrated solutions are stored cold. This solution is diluted 1: 100 to obtain a final assay concentration of 16 mM. The fluorogenic substrate DNP-Pro-Leu-Gly-Leu-Trp-Ala-D-Arg-NH2 with a K ,, of 52 μM is used at a concentration of 21.4 μM; for the determination of K1 (a concentration of 12.8 μM is also used.) The fluorescence of the substrate is measured as an excitation and an emission wavelength of = = 320 and 420 nm, respectively, in a spectrofluorimeter (Perkin Elmer, Model 650 -40) equipped with a support for thermosetting cell The hydrolysis of the substrate is verified for 10 minutes immediately after adding the enzyme All the reactions were carried out at least in triplicate The Kt values of the inhibitors are calculated from the point of intersection of the straight lines obtained by the graphs of v0 / vi vs. [inhibitor concentration] while the CIS0 values are calculated from the graphs of v ,, / v0 [inhibitor concentration] by regression non-linear with simple general weighting. b) Second method Test buffer: 50 mM Tris / HCl pH 7.6 (Tris = Tris- (hydroxymethyl) -aminomethane) 100 mM NaCl 10 mM CaOH 2 MeOH 5% (if necessary) Enzyme: 8 nM catalytic domain (Met80-Gly242) human neutrophil collagenase Substrate: 10 microM DNP-Pro-Leu-Gly-Leu-Trp-Ala-D-Arg-NH2 Total test volume: 1 ml A solution of the enzyme and the inhibitor and the assay buffer (25 ° C) are prepared. The reaction begins directly by providing the substrate in the solution. The breakdown of the fluorogenic substrate is followed by fluorescence spectroscopy with an excitation and emission wavelength of 280 and 350 nm respectively. The CIS0 value is calculated as the inhibitor concentration, which is necessary to decrease the reaction rate by half compared to the reaction without the inhibitor. Table IV shows the IC50 values found.

TABLE I Data collection statistics MMP-8 with MBP-AG-NH2 HONHiBM-AG-NH2 PLG-NHOH Number of measurements 20226 34521 47920 Number of observations 20050 33433 45439 Number of unique reflections (I / s (I) > 0) 6429 9501 9740 Complete condition of the data [%] Infinal. - smaller resolution 92.8% (-2.40 A) 86.2% (-2.05?) 86.3% (2.03 Á) Last coverage (resol. [Al) 88.0% (2.46-2.40) 75.7% (2.10-2.05) 57.8% (2.07-2.03) RMert. "12.9% 12.1% 11.3% R ^ 7.9% 5.2% 5.0% Cell constants 33.24.69.20.72.33 33.13.69.37,72.31 33.09,69.37,72.48 (a, b, c, [Á]: a, 0. ? = 90 °) 1) RMßrgß = EhEi (II) (h, i) - < I (h) > I / EhSiI (h, i), where I (h, i) is the intensity value of the i th measurement of h and < I (h) > is the corresponding mean value of h for all measurements of i of h; the sum is on all the measurements 2) Rsyn = S (IIF- <IF> I) / SIF, where IF is the averaged value of the reflections related to the group of point e < IF > is the averaged value of a Bijvoet pair.

Table II Final refinement statistics MMP-8 with MBP-AG-NH, HONHiBM-AG-NH, PLG-NHOH Resolution interval [? | 8.00-2.40 8.00-2.17"8.00-2.03 Number of unique data in the resolution range 6409 7958 9600 Total number of protein atoms (excluding H) 1266 1266 1266 Atoms of solvent (excluding H) 95 89 111 Factor Rn 8.0-2.40 / -2.17 / -2.03Á 15.7% 19.1 % 18.2% 2. 51-2.40 / 2.20-2.17 / 2.05-2.03Á 20.4% 24.0% 25.1% Deviations of mean square root of target values (excluding metals and H) Unions [Á | 0.012 0.016 0.012 Angles [°] 1.7 1.9 1.8 1) R = (SIFo-FcI / SFo Table III Zinc Ligands and Ligands Distances and Zinc Active Site Angles MMP8 / MMP8 / MMP8 MBP-HONHiBM-PLG-NHOH AG-NH2 AG-NH, Link Lengths [Á]: Zn Ne (197) 1.9 1.9 2.0 Zn Ne (201) 2.3 2.3 2.2 Zn Ne (207) 1.9 2.1 1.9 Zn SH (Inh.) 2.3 SH (Inh.) 0ei (l98) 3.0 SH (Inh.) 0e2 (198) 3.8 Zn OH (Il) 2.2 2.2 Zn 0 (11) 2.2 2.2 OH (Il) 0ei (198) 3.2 2.6 OH (Il) 0e2 (198) 3.0 3.4 Angles [degrees]: Ne (207-Zn-Ne (197) 100.0 99.7 100.8 Ne (197 -Zn-N £ (201) 102.4 102.3 93.7 Ne (207 -Zn-Ne (201) 106.2 94.4 100.0 Ne (207 -Zn-SH (Inh. ) 126.1 Ne (197 -Zn-SH (Inh.) 110.1 Ne (201 -Zn-SH (Inh.) 109.4 Ne (197 -Zn-SH (Il) 108.1 84.7 Ne (197 -Zn-O (Il) 108.2 158.3 Ne (197-Zn-O (Il) 87.6 102.6 Ne (201-Zn-O (Il) 147.1 102.8 Ne (201 -Zn-OH (Il) 151.0 156.4 Ne (207 -Zn-O (Il) 92.6 90.2 0H ( I1) Zn-O (Il) 7.15 78.1 Table IV CIS0 values for the different inhibitors Code Substance CIso ER029 HONH-iBM-Ala-Gly-NH2 139 μM ER017 HONH-iBM-Asn-NHBn 63 μM ER059 HONH-iBM-Ser-Gly-NHBn 61 μM ER014 HONH-iBM-Ala-NHBn 58 μM ER028 HONH- iBM-Ser-NHBn 40 μM ER031 HONH-iBM-Asn-NHBn (m-NH2) 30 μM ER074 HONH-iBM-Phe-NHBn 29 μM ERO70 HONH-iBM-hPhe-NHhBn (p-Me) 1.6 μM (iBM = isobutylmalonic acid, hPhe = homophenylalanine, NHhBn (p-Me) = 2- (4-methyl) phenylethylamine).

References : 1. Oessner Jr., J. F., FASEB J. (1991) 5, 2145-2154. 2. Birkedal-Hansen, H., Moore, W.G.I., Bodden, M.K., windsor, L.J., Birkedal-Hansen, B., DeCarlo, A., Engler, J.A., Critical Rev. Oral Biol. Med. (1993) 4, 197-250. 3 Murphy, G. & Docherty, A.J.P., J. Res. Cell. Mol. Biol. (1992) 7, 120-125. 4 Matrisian, L. M., Trends Genet. (1990) 6, 121-125.

. Bode, W., Gomis-Rüth, F.-X., Stócker, W., FEBS Lett. (1993) 331, 134-140. 6. Sanchez-Lopez, R., Alexander, C.M., Behrendtsen, O., Breathnach, R., Werb, Z., J. Biol. Chem. (1993) 268, 7238-7247. 7. Murphy, G., Allan, J.A., Willenbrock, F., Cockett, M.I. O'Connell, J.P., Docherty, A.J.P., J. Biol. Chem. (1992) 267, 9612-9618. 8. Knuuper, V., Osthues, A., DeClerck, Y. A., Langley. K. E. Bláser, J., Tschesche H, Biochem. J. (1993) 291, 847-854 9. Bode, W., Reinemer, P., Huber, R., Kleine, T., Schnierer. S., Tschesche. H., EMBO J. (1994) 13, 1263-1269.

Docherty, A., J. P., Lyons, A., Smith, B.J., Wright, E.M. Stephens. P. E., Harris, T. J. R., (1985) Nature 318, 66-69. Boone, T. C, Johnson, M., J., DeClerck, Y., A., Langley, K.E., PNAS (1990) 87, 2800-2804. Yang, T.-T., Hawkes, S. P., PNAS (1992) 89, 10676-10680. Nishino, N., Powers, J.C., Biochem. (1978) 17, 2846. Powers. J. C, Harper, J. W, (Barrett, A.J., Salvesen, G., eds.) Elvesier. Amsterdam New York Oxford Johnson,. H., Roberts, N.A., Borkakoti, N., J. Enz. Inhib. (1987) 2, 1-22. Beeley, N.R.A., Ansell, P.R.J., Docherty, A.J.P., Curr. Opin. Ther. Patents (1994) 4, 7-16. Reinemer, P., Grams, F., Huber, R., Kleine, T., Schnierer. S., Pieper, M., Tschesche. H., Bode,., FEBS L. (1994) 338, 227-233. Stams, T., Spurlino, L., ahl, R.C., Ho, T.F, Qoronfleth, M.., Banks, T.M., Rubin, B., Nature Struct. Biol. (1994) 1, 119-123. Borkatoti N., Winkler, F.K., Williams, D.H., D'Arcy, A. .. Broadhurst, M.J., Brown, P.A., Johnson, W.H., Murray, E.J., Nature Struct. Biol. (1994) 1, 106-110.

. Dieckmann, O., Tschesche, H., Brazilian J. Med. Biol. Res. (1994) 27, in press. 21. Schnierer, S., Kleine, T., Gote, T. Hillemann. A., Knauper, V., Tschesche, H., Biochem. Biophys. Res. Commun. (1993) 191, 319-326. 22. Cushman, D., Cheung, H.S., Sabo, E.F., Ondetti, M.A., Biochemistry (1977) 16, 5484-5491. 23. Leslie, A. G. W. (1991) Daresbury Lab. Inf.Quart Prog. Crystallogr. 26 (available from the Librarian, SERC Laboraory, Daresbury, Warrington, WA4 4AD, UK). 24. Steigemann, W. (1991) in: From Chermistry to Biology (Moras, D., Podjarny, A. D. and Thierry, J. C. eds) Crystallographic Computing vol. 5, pp.115-125, Oxford University Press, Oxford, U.K. 25. Jones, T. A., I. Appl. Crystallogr. (1978) 15, 23-31. 26. Brünger, A. T., Karplus, M., Petsko, G.A., Acta Cryst. Sect. A (1989) 45, 50-61. 27. Engh, R. A., Huber, R., Acta Cryst. Sect. A (1991) 47, 392-400. 28. Dixon, M., Biochem. J. (1953) 55, 170-202 29. Chakrabarti, P. Biochemistry (1989) 28, 6081-6085.

. Beszent, B., Bird, J., Gaster, L.M., Harper, G.P., Hughes, I., Karran. E. H., Markwell. R.E., Miles-Williams, A, Smith, S.A., 1. Med. (Chem. 1993) 36, 4030-4039. 31. Netzel-Arnett, S., Fields, G.B., Birkedal-Hansen, H., van Wart, H.E., 1. Biol. Chem. (1991) 266, 6747-6755. 32. Netzel-Arnett, S., Sang, Q.-X., Moore, W.G.I., Navre, M., Birkedal-Hansen, H. van Wart, H.E., Biochemistry (1993) 32, 6427-6432. 33. Niedzwiecki, L., Teahan, J., Harrison, R, K., Stein, R.L., Biochemistry (1992) 31, 12618-12623. 34. Moore, W. I., Spilburg, C. A., Biochemistry (1986) 25, 5189-5195. 35. Odake, S., Okayama, T, Obata, M., Morikawa, T., Hattori, S .. Hori, H., Nagai, Y. Chem. Pharm. Bull. (1991) 39, 1489-1494. 36. Darlak, K., Miller, R.B., Stack, M.S., Spatola, A. F., Gray. R.D., 1. Biol Chem. (1990) 265, 5199-5205. 37. Davies, B., Brown, P.D., East, N., Crimmin, M.J., Balkwill, F.R., Cancer Res. (1993) 53, 2087-2091. 38. Holmes, M.A., Matthews, B.W., Biochemistry (1981) 20, 6912-6920. 39. Schechter, J., Berger, A., Biochem. Biophys. Res. Comm. (1967) 157-162. 40. Snyder, H.R., Stewart, I.M., Ziegler, J.B., J. Am. Chem. Soc. (1947) 69, 2672. 41. van Tamelen. E.E., J. Am. Chem. Soc. (1951) 73, 3444. 42. Price, C C, Kirk, P.F., J. Am. Chem. Soc. (1952) 75, 2396. 43. Ketcham, R., Shah, V.P., J. Org. Chem. (1963) 28, 229. 44. Franks, F., Chem. Abstr. (1964) 60, 5757b. 45. Caló, V., López, L., Márchese, L., Pesce, G., J. Chem. Soc. Chem. Commun. (1975) 621. 46. Chan, TH, Finkenbine, IR, J. Am. Chem. Soc. (1972) 94, 2880. 47. Lowry, OH, Rosebrough, NJ, Faro, AL, Randall, RJ, JBC ( 1951) 193, 265-275. 48. Houben-Weyl, Vol. 15 I / II. 49. Lovejoy, B., Cleasby, A., Hassel, A.M., Longley, K., Luther, M.A. , Weigl, D., McGeehan, G., McErroy, A.B., Drewry, D., Lambert, M.H., Jordan, S.R. Science 263 (1994) 375-377 50. Gooley, PR, O'Connell, JF, Marcy, AI, Cuca, GC, Salowe, SP, Bush, BL, Hermes, JD, Esser, CK, Hagmann, WK, Springer, JP, Johnson, BA, Nature Struc. Biol. I (1994) 111-118 51. F. Grams et al., FEBS 335 (1993) 76-80 Formulas IV: 20 Examples for IV: It is noted that in relation to this date, the best method known by the applicant to carry out the aforementioned invention, is the conventional one for the manufacture of the objects to which it relates. Having described the invention as above, property is claimed as contained in the following:

Claims (19)

1. A compound, characterized in that it is represented by the general formulas I, II or III, which binds and inhibits the matrix metalloproteinases (MMP), in the formulas: XL is oxygen or sulfur, R1 is OH, SH, CH2OH, CH2SH or NHOH, R2 is a residue of 2 to 10 atoms in the main structure, which binds to amino acid 161 of HNC, the residue is saturated or unsaturated, is linear or branched and preferably contains homocyclic or heterocyclic structures, X2 is oxygen or sulfur and binds as a hydrogen bond acceptor at amino acid 160 of HNC, Y is a residue which binds to the SI 'bag of HNC and consists of at least 4 atoms in the main structure of (formula IV), R3 is n-propyl, isopropyl, isobutyl or a residue with minus 4 atoms in the main structure, which is no greater than a tricyclic ring system, and R4 is hydrogen, alkyl or aryl, or a salt thereof, with the proviso that the compound is not HONH-DL-CO- CH- (CH2C6H5) -CO-L-Ala-Gly-NHC6H4N02.

2. The compound according to claim 1, characterized in that R 2 contains an alkyl, alkenyl or alkoxy residue with 2 to 10 atoms in the main structure (C, N, O, S) or a cyclo (hetero) alkyl or aromatic residue with 5 to 10 carbon atoms. to 10 atoms in the main structure (C, N, 0, S).

3. The compound according to claim 1 or 2, characterized in that the structure Z1-Z2, Z3-Z4 of the formula IV consists of 4 atoms in the main structure forming a dihedral angle of approximately 0o (sp2 or sp3 hybridization), that the distance between Z1 and Z4 is between 2.5 and 3.0 Á.

4. The compound according to claims 1 to 3, characterized in that Z1-Z2-Z3-Z4 consists of a peptidomimetic ring structure, for example phenylene, pyridinyl, pyrazinyl, pyrimidinyl, pyridazinyl, piperazinyl, indolinyl and morpholinyl.

5. The compound according to claims 1 to 4, characterized in that Y consists of a peptide group or mimetic peptide.

6. The compound according to claims 1 to 5, characterized in that Y is one of the following residues:

7. The compound according to claims 1 to 6, characterized in that R 4 is hydrogen, isopropyl, n-butyl or benzyl.

8. The use of a compound or a salt thereof, represented by the general formulas I, II or III, for the inhibition of matrix metalloproteinases (MMP), in the formulas: X1 is oxygen or sulfur, Rj is OH, SH, CH2OH, CH2SH or NHOH, R2 is a residue of 2 to 10 atoms in the main structure, which binds to amino acid 161 of HNC, the residue is saturated or unsaturated, it is linear or branched and preferably contains homocyclic or heterocyclic structures, X2 is oxygen or sulfur and binds as a hydrogen bond acceptor at amino acid 160 of HNC, and is a residue which binds to the SI 'bag of HNC and consists of at least 4 atoms in the main structure of Z ^ Z ^ Z ,, ^ (formula IV), R3 is n-propyl, isopropyl, isobutyl or a residue with at least 4 atoms in the main structure, which is no greater than a tricyclic ring system, and R4 is hydrogen, alkyl or aryl, or a salt thereof, with the proviso that the compound is not HONH-DL-CO-CH- (CH2C6H5) -CO- L-Ala-Gly-NHC6H4N02.

9. The use according to claim 8, characterized in that R2 contains an alkyl, alkenyl or alkoxy residue with 2 to 10 atoms in the main structure (C, N, O, S) or a cyclo (hetero) alkyl or aromatic residue with to 10 atoms in the main structure (C, N, 0, S).

10. The use according to claim 8 to 9, characterized in that the structure Z1-Z2-Z3-Z4 of formula IV consists of 4 main structure atoms that form a dihedral angle of approximately 0o (sp2 or sp3 hybridization), wherein the distance between Z1 and Z4 is between 2.5 and 3.0 A.

11. The use according to claims 8 to 10, characterized in that it consists of a peptidomimetic ring structure, for example, phenylene, pyridinyl, pyrazinyl, pyrimidinyl, pyridazinyl, piperazinyl, indolinyl and morpholinyl.

12. The use according to claims 8 to 11, characterized in that Y consists of a peptide or peptide mimetic group.

13. The use according to claims 8 to 12, characterized in that Y is one of the following residues:

The use according to claims 8 to 13, characterized in that R 4 is hydrogen, isopropyl, n-butyl or benzyl.

15. A therapeutic composition characterized in that it is based on a compound according to claims 1 to 7.

16. The therapeutic composition according to claim 15, characterized in that it is in association with one or more carriers and / or diluents and / or pharmaceutically acceptable non-toxic adjuvants.

17. The use according to claims 1 to 7, characterized in that it is for the manufacture of a therapeutic agent for the treatment of rheumatoid arthritis and related diseases in which the collagenolytic activity is a contributing factor.

18. The use according to claim 17, characterized in that the dose of the therapeutic agent is from 0.1 to 300 mg / kg of body weight.

19. The use according to claim 17 or 18, characterized in that the therapeutic agent is administered intravascularly, intraperitoneally, subcutaneously, intramuscularly or topically. BEg3MEN DB TO ONW IOTX The use of a compound represented by the general formulas (I), (II) or (III), for the inhibition of matrix metalloproteinases (MMP), in which X1 is oxygen or sulfur, R1 is OH, SH, is described. CH2OH, CH2SH or NHOH, R2 is a residue of 2 to 10 atoms of hydrocarbon backbone, which binds to amino acid 161 of HNC, the residue is saturated or unsaturated, is linear or branched and preferably contains homocyclic structures or heterocyclic, X2 is oxygen or sulfur and binds as a hydrogen acceptor at amino acid 160 of HNC, and is a residue which binds to the SI 'bag of HNC and consists of at least 4 atoms in the main structure of Z1Z2Z3 -Z4-R3, and R3 is n-propyl, isopropyl, isobutyl or a residue with at least 4 atoms in the main structure, which is no greater than a tricyclic ring system. These compounds bind to the MMPs in a manner different from the binding mode of the inhibitors of the prior art.