Search

Matrix metalloproteinases are a growing family of neutral pH optima, zinc atom-dependent endopeptidases that collectively degrade all components of the extracellular matrix. This family of related proteases is further defined by their inhibition of protease activity by a class of low-molecular-weight endogenous inhibitors known as tissue inhibitors of metalloproteinases or TIMPs. Reverse zymography is an electrophoretic technique used to identify TIMP inhibitory activity within acrylamide gels. Previous methods have generally used biochemically complex sources of proteolytic activity (such as cell culture conditioned media) copolymerized with a proteinase substrate in the gel to identify the zones of inhibited proteolysis. We describe a novel system for reverse zymography using purified recombinant human gelatinase A or gelatinase B in place of conditioned media. These reverse zymograms using recombinant gelatinase have sensitivities for TIMPs that are favorable in comparison to immunoblotting techniques but have the benefit of visualizing multiple inhibitors simultaneously. We have developed and characterized these methods for the evaluation of inhibitors and have shown them to be highly sensitive, convenient, and reproducible. Both systems detect TIMPs 1, 2, and 3 simultaneously, but with differential sensitivities for TIMPs 1 and 2. Using gelatinase A the system can detect as little as 1 pg of rTIMP-2, but the limit of detection for rTIMP-1 is 40 pg. Gelatinase B shows less differential activity in that the limits of detection are 60 and 40 pg for TIMP-2 and TIMP-1, respectively. We demonstrate how these varied sensitivities of the gelatinases for the TIMPs can contribute to potential pitfalls in systems using uncharacterized reagents (i.e., conditioned media).

We describe the use of gelatinase standards from human capillary whole blood for calibrating gelatin zymograms. Capillary blood was obtained by fingerstick puncture and prepared in nonreducing Laemmli SDS-PAGE sample buffer without heating. Gelatin zymography revealed that whole blood (0.25- 1 microliter) contained substantial fibroblast-derived (72-kDa) and neutrophil-derived (92-, 130-, and 225-kDa) gelatinases which could be used for calibration purposes. Calibration was linear under different electrophoretic conditions using 6.5-10% polyacrylamide gels containing 1-2 mg/ml gelatin (correlation range, r = 0.990-0.998; perfect correlation, r = 1.000). The use of the standards for characterizing gelatinases from arthritic joint fluid is demonstrated. The standards are easily obtained and well characterized, and should facilitate interlaboratory comparison and standardization.

Two hemorrhagic metalloproteinases (LHF-I and LHF-II) were previously isolated from Lachesis muta muta (bushmaster snake) venom. The proteolytic activities of these hemorrhagic factors and of the crude venom were investigated using as substrate the oxidized B-chain of bovine insulin. LHF-II cleaves the Ala14-Leu15 bond of insulin B-chain very rapidly and the Phe24-Phe25, His10-Leu11 and His5-Leu6 more slowly, whereas LHF-I hydrolyzed only the Ala14-Leu15 bond. Both hemorrhagic factors cleaved the Leu-Leu bond in the fluorogenic peptide Abz-Pro-Leu-Gly-Leu-Leu-Gly-Arg-EDDnp. When the insulin B-chain was incubated with crude venom previously treated with 2.5 mM PMSF, the Ala14-Leu15 bond was also rapidly cleaved. In addition, the hemorrhagic activity and the digestion of casein remained unaltered. Both hemorrhagic and proteolytic activities were inhibited when the crude venom was treated with EDTA, confirming that only metalloproteinases are responsible for these activities. The hydrolysis of insulin B-chain and the fluorogenic heptapeptide by these proteinases was found to be in inverse relationship to their hemorrhagic activities.

The invasion of blood vessel walls is a critical step in cancer metastasis, in which endothelial cells and their vascular basement membranes act as barriers to tumor cell passage. Here we report that Walker 256 carcinosarcoma (W256) cells degrade subendothelial matrices by a process involving both the generation of hydrogen peroxide and the secretion of a matrix metalloproteinase. As an assay of basement membrane degradation, [3H]proline-labeled subendothelial matrices were exposed to W256 cells in the presence or absence of the chemotactic peptide N-formyl-L-methionyl-L-leucyl-L-phenylalanine (fMLP). The release of [3H]proline, in the presence of 5 x 10(6) W256 cells, was increased from 49 +/- 2.5 to 64 +/- 2.2% by the addition of 10(-6) M fMLP. In the presence of fMLP-activated W256 cells, [3H]proline release was completely inhibited by the addition of 2000 units/ml catalase or by the metalloproteinase inhibitors 1,10-phenanthroline and EDTA at concentrations > or = 10 micrograms/ml. alpha 1-Antitrypsin or alpha 2-macroglobulin were without effect. Cell-free supernatants obtained from activated W256 cells were also able to promote basement membrane degradation. Electrophoresis of the cell-free supernatants from fMLP or PMA-activated W256 cells in gelatin-containing sodium dodecyl sulfate-polyacrylamide gels revealed a major band of gelatinolytic activity at 94 kDa. The 94-kDa band represented the activity of a latent gelatinase since incubation with 1 mM 4-aminophenylmercuric acetate (APMA; a known activator of latent metalloproteinases) resulted in the loss of gelatinolytic activity at 94 kDa and the appearance of five new bands of lower molecular weight (M(r) 86, 79, 74, 70, and 66 kDa). Two of these lower molecular weight bands (M(r) 86 and 66 kDa) were also detected in the absence of APMA, following 10-fold concentration of the cell-free supernatants. When the cell-free supernatants of phorbol myristate acetate-activated W256 cells (concentrated 10-fold) were incubated with increasing concentrations of hydrogen peroxide (35 to 70 mM), the band at 66 kDa demonstrated enhanced gelatinolytic activity. We suggest that W256 cells can secrete a latent metalloproteinase of molecular weight 94 kDa which, when activated by hydrogen peroxide, can degrade subendothelial matrices.

The purpose of the present research was to determine if metalloproteinase levels were elevated in human chronic wound fluid. Samples of blood and wound fluid from acute (mastectomy) and chronic (leg ulcer) wounds were collected, and metalloproteinase profiles of the samples were determined by gelatin zymography. Compared to serum, acute wound fluid (mastectomy fluid) contained markedly increased levels (five-to tenfold) of several metalloproteinases including 72-kDa and 94-kDa gelatinases (MMP-2 and MMP-9). In chronic wound fluid, not only were these enzyme levels increased another five- to tenfold over mastectomy fluid, but also both activated enzyme and proenzyme species appeared to be present. Our results suggest that non-healing ulcers develop an environment containing high levels of activated metalloproteinases, which may result in chronic tissue turnover and failed wound closure.

The sequence specificities of human 72-kDa fibroblast gelatinase (type IV collagenase), human 92-kDa neutrophil gelatinase (type IV collagenase), and putative metalloproteinase (PUMP or matrilysin) have been examined by measuring the rate of hydrolysis of over 50 synthetic oligopeptides covering the P4 through P4' subsites of the substrate. The peptides investigated in this paper were those employed in our previous study which systematically examined the sequence specificity of human fibroblast and neutrophil collagenases [Netzel-Arnett et al. (1991) J. Biol. Chem. 266, 6747]. The initial rate of hydrolysis of the P1-P1' bond of each peptide has been measured under first-order conditions ([S0] << KM), and kcat/KM values have been calculated from the initial rates. The specificities of these five metalloproteinases are similar, but distinct, with the largest differences occurring at subsites P1, P1', and P3'. The specificities of the two gelatinases are the most similar to each other. They tolerate only small amino acids such as Gly and Ala in subsite P1. In contrast, larger residues such as Met, Pro, Gln, and Glu are also accommodated well by PUMP. All five enzymes prefer hydrophobic, aliphatic residues in subsite P1'. PUMP exhibits a stronger preference for Leu in this subsite than is shown by the other enzymes. The P3' subsite specificities of the gelatinases and collagenases are very similar but different from those of PUMP which particularly prefers Met in this position. The specificity data from this study allow the design of optimized substrates and selective inhibitors for these metalloproteinases.

Mesothelial cells play a critical role in the remodeling process that follows serosal injury. Although mesothelial cells are known to synthesize a variety of extracellular matrix components including types I, III, and IV collagens, their potential to participate in matrix degradation has not been explored. We now report that human pleural and peritoneal mesothelial cells express interstitial collagenase, 72- and 92-kD gelatinases (type IV collagenases), and the counterregulatory tissue inhibitor of metalloproteinases (TIMP). Our initial characterization of the mesothelial cell metalloenzymes and TIMP has revealed: (a) they are likely identical to corresponding molecules secreted by other human cells; (b) they are secreted rather than stored in an intracellular pool; (c) a primary site of regulation occurs at a pretranslational level; (d) phorbol myristate acetate, via activation of protein kinase C, upregulates expression of collagenase, 92-kD gelatinase, and TIMP, but has no effect on expression of 72-kD gelatinase; and (e) lipopolysaccharide fails to upregulate the biosynthesis of either metalloproteinases or TIMP. Of particular interest is the observation that the state of cellular differentiation has a striking influence on the expression of metalloenzymes and TIMP, such that epitheloid cells display a more matrix-degradative phenotype (increased 92-kD gelatinase and decreased TIMP) than their fibroblastoid counterparts. We speculate that mesothelial cells directly participate in the extracellular matrix turnover that follows serosal injury via elaboration of metalloproteinases and TIMP. Additionally, the reactive cuboidal mesothelium which is characteristic of the early response to serosal injury may manifest a matrix-degenerative phenotype favoring normal repair rather than fibrosis.

Matrix metalloproteinases (MMPs) are a family of enzymes which contain zinc at their active site and can degrade most of the matrix macromolecules found in connective tissues. These MMPs are secreted by connective tissue cells and infiltrating leucocytes in response to inflammatory mediators. There is now widespread recognition that MMPs are the major class of proteinases responsible for the excessive degradation of cartilage that leads to joint dysfunction in rheumatoid arthritis. The properties of the MMPs are reviewed and a therapeutic role for synthetic, zinc-binding pseudopeptide MMP inhibitors in the treatment of arthritis is proposed.

The metalloproteinases are a family of enzymes that can degrade all the components of the extracellular matrix. These potent enzymes are often found in proenzyme forms and require activation before the substrate can be digested. To prevent unlimited connective tissue destruction a number of inhibitors exist to limit their activity. In a previous study it was found that metalloproteinases in proenzyme form and metalloproteinase inhibitors were often present in rheumatoid synovial fluids. Two of these inhibitors were identified in rheumatoid synovial fluid as alpha 2 macroglobulin and tissue inhibitor of metalloproteinase (TIMP), the specific metalloproteinase inhibitor. A third inhibitory peak was unidentified. In the study reported here it was shown that this third inhibitor can be purified using gelatin-Sepharose chromatography and consists of TIMP-2 bound to progelatinase (relative molecular weight 72,000) in a similar way to that found in concentrated connective tissue culture medium. The importance of these proteinase inhibitors in synovial fluid is discussed.

Entactin is the basement membrane protein which bridges laminin and type IV collagen. Entactin is known to be degraded by serine proteinases, but its susceptibility to matrix metalloproteinases has not been determined. We have studied the capacity of three matrix metalloproteinases (interstitial collagenase, 92-kDa gelatinase, and matrilysin) to degrade entactin. While all three metalloenzymes cleaved entactin, matrilysin was approximately 100-fold as effective as collagenase and 600-fold as effective as 92-kDa gelatinase. The Km of matrilysin for entactin was 8.9 x 10(-7) M. A Vmax of 21 molecules of entactin degraded/molecule of matrilysin/min at 37 degrees C was observed. An Arrhenius plot relating matrilysin's catalytic activity to temperature was linear from 15 to 37 degrees C and indicated an activation energy of 10,060 calories/mol. Matrilysin produced multiple, but distinct, cleavages in entactin resulting in peptide fragments ranging from 115 to 29 kDa. The precise sites of cleavage of six fragments were determined by Edman degradation. Cleavage sites consistently occurred amino-terminal to leucine or isoleucine. These data indicate that entactin is a substrate for matrix metalloproteinases. The effectiveness of matrilysin is noteworthy, however, particularly in relation to the minimal ability of other much more well described matrix metalloproteinases to attack this substrate. Our results suggest a potentially important role for matrilysin in disruption of basement membranes by tumor or inflammatory cells.

Chicken embryo fibroblasts (CEF) produce a 70kDa progelatinase, a member of the matrix metalloproteinase family, and secrete elevated levels of the enzyme upon transformation by Rous sarcoma virus (RSV). This enzyme can be purified by affinity chromatography complexed with a 21kDa tissue inhibitor of metalloproteinases (TIMP)-like molecule. Gel-filtration of the purified progelatinase suggests the presence of a mixed population of enzyme: a TIMP-complexed and a TIMP-free progelatinase. These two species were separated by Mono Q FPLC in the absence of denaturants. Quantitation of the purified progelatinase reveals that the transformed RSVCEF produce more TIMP-free enzyme than the normal CEF. Native PAGE analysis indicates that purified TIMP-free progelatinase is capable of binding to TIMP and generating a TIMP-complexed progelatinase. Treatment of the TIMP-free gelatinase with organomercurials results in a rapid conversion to the active 62kDa species and indicates that the TIMP-free progelatinase is more susceptible to activation than the TIMP-complexed progelatinase.

Collagenase production by rodent osteoblasts in response to calciotropic hormones has led to the hypothesis that bone cells play a major role in bone resorption by degrading the surface osteoid layer, thereby exposing the underlying mineralized matrix to osteoclastic action. Many studies suggest, however, that this model might not apply to bone resorption in the human. Human osteoblasts have been shown to produce gelatinase-A (72 kDa) and TIMP-1 (tissue inhibitor of metalloproteinases), but previous investigators have been unable to demonstrate the synthesis of collagenase by human osteoblasts either constitutively or in response to bone resorptive agents. In the present study the ability of human osteoblasts to produce the matrix metalloproteinases (MMPs) collagenase, gelatinase and stromelysin, and their specific inhibitors TIMPs-1 and 2, was examined using highly sensitive and specific antisera and by zymography. Semi-quantitative histomorphometric data showed that cells cultured on either glass or a type I collagen substratum constitutively synthesized gelatinase-A and TIMP-1. On type I collagen, however, a small proportion of unstimulated cells produce both collagenase (7%) and gelatinase-B (95 kDa; 3%). Treatment of cells with either parathyroid hormone (PTH), 1,25-dihydroxy-vitamin D3 (1,25(OH)2D3), or partially purified mononuclear cell conditioned medium (MCM), stimulated the synthesis of collagenase, gelatinase-B and stromelysin; MCM was 2- to 3-fold more potent than either PTH or 1,25(OH)2D3. Zymography using SDS/PAGE on conditioned media from cells cultured on type I collagen films revealed the presence of active gelatinase-A and that MCM stimulated progelatinase-B synthesis.(ABSTRACT TRUNCATED AT 250 WORDS)

We found previously that two fibrinolytic enzymes (jararafibrases I and II) purified from Bothrops jararaca venom displayed a haemorrhagic activity. To elucidate the mechanisms involved and the role of the enzymatic activity in haemorrhage, the enzymatic properties of the purified enzymes were examined. The substrate specificity of the enzymes was determined using type I collagen, type IV collagen, gelatin, laminin and fibronectin as substrates. The enzymes degraded type IV collagen, gelatin, laminin and fibronectin into smaller fragments, but degraded type I collagen only partially in a non-specific manner. The specific activities of jararafibrase I for type IV collagen and gelatin were 172 +/- 5 units/mg protein and 1315 +/- 177 units/mg protein, respectively. The specific activities of jararafibrase II for type IV collagen and gelatin were 9.2 +/- 0.6 units/mg protein and 143 +/- 15 units/mg protein, respectively. It was evident that the enzymes had rather broad substrate specificities and degraded basement membrane components including type IV collagen. The number of type IV collagen units of bacterial collagenase which gave the minimal haemorrhagic dose was 191.4, while the numbers of type IV collagenase units of jararafibrases I and II which gave the minimal haemorrhagic dose were 1.5 and 0.25, respectively. It is suggested that the broad substrate specificity of the enzymes is essential for inducing haemorrhage with a single enzyme.

Matrix metalloproteinases play a central role in the catabolism of extracellular matrix macromolecules. Here the authors report that giant cell tumor of bone (GCT) produces two matrix metalloproteinases (MMPs) in zymogen form, which have been identified as proMMP-2 (also known as "72-kDa-progelatinase/type IV procollagenase") and proMMP-3 (prostromelysin). Giant cell tumor is known to consist of two major cell populations, multinucleated giant cells and stromal cells. On several passages of the tumor cells in culture, only stromal cells proliferated. These stromal cells produced proMMP-2 but not proMMP-3. Addition of the conditioned medium of primary GCT culture or human macrophage-conditioned medium to the passaged stromal cells induced the production of proMMP-3. The production of proMMP-3 was also induced by interleukin 1 (IL-1), but not by tumor necrosis factor alpha (TNF alpha). ProMMP-1 (tissue procollagenase) was not detected even after treatment with these stimuli. Immunohistochemical studies have demonstrated that multinucleated giant cells in GCT both produce IL-1 and TNF alpha, suggesting that IL-1 secreted by multinucleated giant cells may be responsible for in vivo production of proMMP-3 by the stromal cells. The authors propose that GCT has a self-stimulatory system for the production of matrix-degrading proteinases and that the ability of the passaged stromal cells to synthesize and secrete proMMP-3 with appropriate stimuli may contribute the malignant behavior of GCT.

We have previously reported that recombinant human interleukin-1 (IL-1) stimulates matrix erosion in bovine nasal cartilage explants (R. J. Smith et al., Inflammation 13, 367-382, 1989). This action of IL-1 is believed to be caused by matrix-degrading neutral proteinases produced by activated chrondrocytes. Accordingly, we investigated the effects of recombinant human interleukin-1 alpha (IL-1 alpha), recombinant human interleukin-1 beta (IL-1 beta), and recombinant human tumor necrosis factor alpha (TNF alpha) on bovine nasal chondrocyte (BNC) responsiveness. IL-1 alpha and IL-1 beta stimulated a time (0-72 hr) and concentration-dependent (0.01-10 ng/ml) production of collagenase, gelatinase, caseinase, and prostaglandin E2 (PGE2) in BNC monolayer cultures. Neutral proteinase and PGE2 production by BNC was also induced by TNF alpha (0.2-200 ng/ml) in a time-dependent (0-72 hr) manner. Recombinant human interleukin-6 (IL-6) caused a concentration-dependent (6-200 ng/ml) potentiation of IL-1-stimulated neutral proteinase and PGE2 production by BNC. However, recombinant human platelet-derived growth factor homodimer BB suppressed BNC responsiveness to IL-1. A recombinant human IL-1 receptor antagonist protein inhibited BNC activation by IL-1 but not TNF alpha.

Connective-tissue cells in culture produce a family of metalloproteinases which, once activated, can degrade all the components of the extracellular matrix. These potent enzymes are all inhibited by the tissue inhibitor of metalloproteinases (TIMP), and it was thought that this inhibitor was solely responsible for the inhibition of these enzymes within connective tissue. However, other inhibitors have recently been described, including large inhibitor of metalloproteinases (LIMP) present in the culture medium of human foetal lung fibroblasts. Here we show that a large proportion of the inhibitory activity of LIMP consists of 72,000-M(r)-progelatinase bound to TIMP-2, a recently discovered low-M(r) metalloproteinase inhibitor closely related to TIMP. The physiological implications of the secretion of a complex of 72,000-M(r) progelatinase and TIMP-2 are discussed, and the separation of the complex in 6 M-urea is described.

Progelatinase A was purified as a complex with TIMP-2 from the conditioned medium of a human glioblastoma cell line. The TIMP-2/progelatinase complex was resistant to the activation by p-aminophenylmercuric acetic acid (APMA), and showed less than 10% of the activity of the TIMP-2-free active enzyme. When the complex was incubated with stromelysin in the presence of APMA, the 64-kDa progelatinase was effectively converted to the 57-kDa mature enzyme, increasing its gelatinolytic activity about 8-fold. These results suggest that stromelysin is a natural activator of TIMP-2-bound progelatinase A.

Fifteen posterior uveal melanoma cell lines were analyzed qualitatively for gelatinolytic and caseinolytic proteinase activity after one to five in vitro passages. All 15 cell lines secreted a gelatinolytic metalloproteinase, with an apparent molecular weight of 72 kD, into protein-free culture media; nine of these secreted an additional gelatinolytic metalloproteinase with an apparent molecular weight of 92 kD. Neither species had the ability to degrade casein. This approach may provide insight into the mechanisms of tumor metastasis in uveal melanoma.

During the involution of the mammary gland there is destruction of the basement membrane as the secretory alveolar structures degenerate. Immunofluorescence staining of sections of rat mammary gland with antibodies to 72 KD gelatinase (MMP-2) and stromelysin (MMP-3) revealed increased production of these two proteinases during involution. This increased expression was mostly restricted to myoepithelial cells. Increased expression during involution was also demonstrated by immunoblotting techniques. Gelatin zymography indicated that the predominant metalloproteinase present in involuting rat mammary glands was a 66 KD gelatinase.

The maintenance and developmental remodeling of extracellular matrix is crucial to such processes as uterine implantation and the cell migratory events of morphogenesis. When mouse blastocysts are placed in culture they adhere to extracellular matrix, and trophoblast giant cells migrate out onto the matrix and degrade it. The secretion of functional proteinases by developing mouse embryos increases dramatically at the time of implantation. By zymography we identified the major secreted gelatin-degrading proteinase, also known as type IV collagenase, as one migrating at 92 x 10(3) Mr. Several casein-degrading proteinases were also secreted. The tissue inhibitor of metalloproteinases (TIMP) inhibited all of the embryo-derived proteinases detected by gelatin gel zymography, indicating that they are metalloproteinases, whereas TIMP did not inhibit all of the caseinases. Urokinase was also secreted. Addition of TIMP at 5-500 nM effectively inhibited the degradation of matrix by the trophoblast outgrowths. Blocking antibodies directed against 92 x 10(3) Mr gelatinase abolished matrix degradation by the trophoblast cells. These observations suggest that several metalloproteinases are regulated in early development and that 92 x 10(3) Mr gelatinase, in particular, has a rate-limiting function in degradation of the maternal extracellular matrix by trophoblast cells.

Gelatinase and proteoglycanase are metalloproteinases that govern extracellular matrix remodeling. In the present study, immature rats were primed with eCG (20 IU) and hCG (10 IU). Ovarian gelatinase and proteoglycanase activity were determined at the time of hCG administration (0 h) as well as 4, 8, and 12 h later. Gelatinase and proteoglycanase were extracted by homogenization in Triton and by heating (i.e., heat extraction). An aliquot of the heat extract was reduced and alkylated to destroy metalloproteinase inhibitors. Heat extracts not reduced and alkylated showed low levels of gelatinase and proteoglycanase activity that did not change at the different time points. However, with reduction and alkylation, gelatinolysis increased approximately 4-fold (p less than 0.05) at 4 h, 8 h, and 12 h after hCG priming. Proteoglycanase activity increased approximately 2-fold (p less than 0.05) between 0 and 8 h and declined at 12 h after hCG. The ovarian gelatinolytic activity was due to a metalloproteinase as demonstrated by the inhibition of enzyme activity by phenanthroline and EDTA (97.1 +/- 0.7% and 97.4 +/- 0.6% inhibition respectively). Proteoglycanase activity was not inhibited by phenanthroline (11.5 +/- 3.5%), suggesting that the enzyme activity was not specifically a metal-dependent enzyme. Gelatin gel zymography of the ovarian extracts demonstrated four predominant and distinct gelatin-degrading enzymes of 78, 72, 66, and 62 kDa, similar to the size of gelatinase. The present findings demonstrate a periovulatory increase in ovarian gelatinolytic and proteglycanase activity that may play a pivotal role in connective tissue remodeling associated with ovulation.

Human melanoma cells secrete a 21 kDa protein which binds with 1:1 molar stoichiometry to the matrix metalloproteinase type IV collagenase proenzyme (70 kDa gelatinase) secreted by the same cells. We have purified this binding protein and determined its complete primary structure by directly sequencing overlapping peptide fragments which span the entire protein. We refer to this protein as CSC-21K based on the amino-terminal amino acids CSC and the apparent molecular weight of 21,000 daltons on gel electrophoresis. The amino acid sequence of CSC-21K demonstrates that this protein shares significant homology with human TIMP (tissue inhibitor of metalloproteinase), including conservation of the positions of the twelve cysteine residues and three of four tryptophan residues. The identification of CSC-21K now indicates that a family of TIMP-related proteins exists. Individual members of this family may possess selective affinities for different members of the matrix metalloproteinase family. Based on its sequence homology to TIMP and ability to inhibit type IV collagenolysis we propose the name TIMP-2 for this inhibitor. TIMP-2 produced by tumor cells can also be considered as an onco-suppressor gene product, because it could play an important role in regulating the metalloproteinases involved in tumor invasion and angiogenesis.

This study was performed to characterize the matrix metalloproteinases (MMPs) produced during the degradation of cotton-wrapped cartilage, implanted into the murine air pouch. One, two or three weeks following cartilage implantation, proteins were extracted from the granulation tissue and MMP activities were measured. Although collagenase-, gelatinase- and stromelysin-like activities were detected at each time point, gelatinase activity was by far the most prominent. These enzymes were inhibited by EDTA, but not by NEM or PMSF, indicating that these proteinases were metalloproteinases. Gelatin zymography revealed several lysis zones amongst which a major 92-kDa band shifted to 83- and 68-kDa species during the course of implantation. The emergence of these species coincided with enhanced gelatinolytic activity and collagen loss from the implanted cartilage.

Procollagenase M(r) 85,000 (SDS-PAGE) was purified from buffy coat to homogeneity and represents a stable single polypeptide chain forming the entire proenzyme. The procollagenase can be activated by various proteinases, e.g. trypsin, chymotrypsin, cathepsin G, kallikrein and stromelysin and by different mercurial compounds. Proteolytic conversion of the latent enzyme to the active form by chymotrypsin is accompanied by a molecular weight reduction to an apparent M(r) 64,000. This active enzyme lacks the first 79 N-terminal residues. Activation by trypsin leads to a latent intermediate of apparent M(r) 70,000, lacking 48 N-terminal residues. The active enzyme is therefore generated upon prolonged incubation with trypsin by further cleavage of 22 N-terminal residues. Another latent intermediate form with apparent M(r) 69,000 is generated from the proenzyme upon incubation with leukocyte elastase by N-terminal cleavage of 53 or 64 residues, respectively. However, latent collagenase cannot be activated by plasmin. Activation by different mercurial compounds finally results in the formation of active collagenase with apparent M(r) 64,000. In contrast to the proenzyme, active collagenase can autolyse to give active M(r) 57,000 and 45,000 intermediates and two M(r) 28,000 fragments. Purification of latent leukocyte gelatinase yields three final products with apparent M(r) 98,000, 125,000 and 220,000 (SDS-PAGE; non reduced). Upon reduction, only the M(r) 98,000 form can be detected. The latent gelatinase can be activated in a similar manner as collagenase. Proteolytic activation by trypsin leads after N-terminal cleavage to an active gelatinase with sequence homology to leukocyte collagenase.

Highly purified human polymorphonuclear (PMN) leucocyte matrix metalloproteinases, collagenase and gelatinase, cleaved human plasma C1-inhibitor at the carboxyl site of Ala439 (P6). This led to a concomitant loss of C1-inhibitor activity. An additional cleavage site, at the carboxyl site of Ser441 (P4), was observed during PMN leucocyte gelatinase-induced inactivation, and a minor fragment of the plasma C1-inhibitor was generated.

We report that monolayers of human fibroblasts stimulated with concanavalin A were able to activate 72 kDa progelatinase but not 95 kDa progelatinase. The activating capacity of fibroblasts appeared approx. 6 h after concanavalin A stimulation and was blocked by cycloheximide. The activation of 72 kDa progelatinase was readily inhibited by TIMP-2 but only poorly by TIMP-1. Plasma membranes isolated from the fibroblasts were capable of activating 72 kDa progelatinase. The cleavage products of the plasma membrane-mediated activation of 72 kDa progelatinase corresponded to those of organomercurial-induced self-cleavage. Only inhibitors of metalloproteinase self-cleavage inhibited the activating capacity of plasma membrane preparations, although the activating capacity was destroyed by trypsin and heat. As with the fibroblast monolayers, TIMP-2 was a potent inhibitor of the membrane-mediated activation whereas TIMP-1 was less so.

Serpins encompass a superfamily of proteinase inhibitors that regulate many of the serine proteinases involved in inflammation and hemostasis. In vitro, many serpins are catalytically inactivated by proteinases that they do not inhibit, leading to the concept of proteolytic down-regulation of serpin inhibitory capacity. The extent to which down-regulation of serpin activity occurs in vivo is debated, since little is known of the rates at which the process occurs. To address this debate, we have measured the rates of inactivation of three serpins, alpha 1-proteinase inhibitor (alpha 1PI), alpha 1-antichymotrypsin (alpha 1ACT), and antithrombin III (ATIII), by three human matrix metalloproteinases (MMPs-1, -2, and -3) thought to be involved in tissue destruction and repair. Our object was to establish a working kinetic model which can be used to predict whether serpin inactivation by these proteinases is likely to occur in vivo. We determined the rates of inactivation of these three serpins by each of the MMPs and compared these to rates of inhibition of the MMPs by an endogenous inhibitor, alpha 2-macroglobulin. An equation designed to predict the extent of substrate hydrolyzed by an enzyme in the presence of an enzyme inhibitor gave the following predictions of the inactivation in vivo: (i) ATIII is unlikely to be inactivated by the MMPs. (ii) MMP-2 (72-kDa gelatinase/type IV collagenase) is unlikely to inactivate any of the three serpins. (iii) MMP-1 (tissue collagenase) will inactivate alpha 1PI and alpha 1ACT only when its concentration saturates that of its controlling inhibitors. (iv) MMP-3 (stromelysin) may inactivate small amounts of alpha 1PI and more significant amounts of alpha 1ACT, even in the presence of its controlling inhibitors. Any physiologic or pathologic inactivation of these serpins by these MMPs that occurs in vivo will probably be due to MMP-3, and will likely only take place in tissues and inflammatory loci where the concentration of MMP inhibitors is depressed.

Human gingival fibroblasts in culture were shown to secrete a 72 kDa progelatinase, of which a proportion in the medium was found to be complexed with tissue inhibitor of metalloproteinases-2 (TIMP-2). A purification procedure was devised to purify free enzyme and inhibitor. We also describe the purification of both 95 kDa progelatinase bound to TIMP-1 and free 95 kDa progelatinase from the medium of U937 cells. A polyclonal antiserum to TIMP-2 was prepared and it was shown that TIMP-1 and TIMP-2 are antigenically distinct. The ability to form stable complexes and the relative inhibitory activities of TIMP-1 and TIMP-2 towards 95 kDa and 72 kDa gelatinases, collagenase, stromelysins 1 and 2 and punctuated metalloproteinase were determined; only minor differences were found. Complex-formation between TIMP-2 and 72 kDa progelatinase was demonstrated not to reduce the metalloproteinase-inhibitory activity of TIMP-2, a finding that led to the characterization of high-molecular-mass TIMP activity. Competition experiments between progelatinases and active gelatinases for TIMPs indicated that the affinity of TIMPs for progelatinases is weaker than that for active gelatinases. In a study of the effects of TIMP-1 and TIMP-2 on progelatinase self-cleavage we found that both TIMP-1 and TIMP-2 inhibit the conversion of 95 kDa and 72 kDa progelatinases and prostromelysin into lower-molecular-mass forms. TIMP capable of complexing with progelatinase was shown to be no more efficient an inhibitor of gelatinase self-cleavage than TIMP not able to complex with progelatinase.

The enzyme collagenase participates in remodeling the extracellular matrix of corneal stroma during normal wound healing and mediates the degradation of extracellular matrix that occurs in many corneal pathologic states. Because this enzyme is synthesized and secreted by corneal cells, therapy of degradative disorders might be geared toward control of enzyme expression. The effects of two cytokines, transforming growth factor-beta (TGF-beta) and interleukin-1 (IL-1), on the expression of collagenase by cultured corneal stromal cells are reported. In addition, the concomitant effects of these cytokines on expression of three additional members of the matrix metalloproteinase (MMP) family--stromelysin, 72-kilodalton (kD) gelatinase, and 92-kD gelatinase--were investigated. When stromal cells are situated in the normal corneal stroma, they produce only a single MMP, 72-kD gelatinase. This pattern of expression was reproduced by stromal cells freshly plated in primary culture. However after passage in culture, the cells also began to secrete collagenase and stromelysin. Treatment of primary cultures with recombinant human IL-1 also induced collagenase and stromelysin expression. In addition, 92-kD gelatinase expression was induced and 72-kD gelatinase expression was increased further by IL-1 treatment. Treatment of passaged cultures or IL-1-treated primary cultures with recombinant human TGF-beta reverted the pattern of enzyme expression toward that exhibited by primary, untreated cultures, ie, expression of collagenase and stromelysin was repressed while expression of 72-kD gelatinase was increased. These results suggest that TGF-beta and IL-1 may be important agents for controlling MMP expression in healthy and diseased corneas.

1. Atherosclerosis and aneurysm of the abdominal aorta are associated with thinning of the medial connective tissue. We have investigated the presence of the connective-tissue-degrading metalloproteinases in homogenates prepared from atherosclerotic, aneurysmal and control aortic media. 2. Gelatinase activity was much increased in homogenates from atherosclerotic and aneurysmal aorta [10.9 +/- 1.8 and 13.3 +/- 3.3 micrograms of gelatin hydrolysed h-1 (mg of protein)-1 respectively]. This gelatinase activity was highest at the luminal aspect of the aortic media, where the activity increased three- to five-fold after the destruction of alpha 2-macroglobulin. Zymograms demonstrated the principal gelatinase in atherosclerotic aorta to have a molecular mass of about 92 kDa, whereas in aneurysmal aorta there was a spectrum of gelatinase activity from 92 to 55 kDa. 3. Collagenase and stromelysin (proteoglycanase) could be detected by immunoblotting in homogenates of aneurysmal aorta, but rarely in atherosclerotic aorta and never in control aorta. Collagenase and stromelysin activities were low, but increased two- to three-fold after the destruction of tissue inhibitor of metalloproteinases. Collagenase and stromelysin activities were highest at the adventitial aspect of aneurysmal media. 4. The secretion of gelatinase by inflammatory cells at the intima of diseased aorta could have a pathological role in establishing atherosclerotic plaques and medial thinning. Secretion of collagenase, gelatinase and stromelysin from the adventitia could accelerate connective tissue degradation in the media of aneurysmal aorta.

Transformed human fibroblasts secrete two structurally and functionally related inhibitors of matrix metalloproteinases, tissue inhibitor of metalloproteinases (TIMP) 1 and 2. In assays measuring the relative inhibitory capability of TIMP-1 and TIMP-2 against autoactivated 72-kDa gelatinase, which consists of two major active peptides and several inactive fragments, TIMP-2 was more effective than TIMP-1. The isolated 42.5-kDa active fragment that formed as a result of the autoactivation of 72-kDa gelatinase showed the greatest preference for TIMP-2; at half-maximal inhibition, TIMP-2 was greater than 10-fold more effective than TIMP-1. TIMP-2 was also greater than 2-fold more effective than TIMP-1 at inhibiting 72-kDa gelatinase-TIMP-2 complexes activated with 4-aminophenylmercuric acetate, and greater than 7-fold more effective than TIMP-1 at inhibiting 92-kDa gelatinase activated with 4-aminophenylmercuric acetate. Furthermore, these active gelatinases preferentially bound 125I-TIMP-2 when incubated with equal amounts of radiolabeled TIMP-1 and TIMP-2. The ratios of 125I-TIMP-2/125I-TIMP-1 binding to 92-kDa gelatinase, autoactivated 72-kDa gelatinase, and 42.5-kDa fragment were 4.4, 10, and 33, respectively. On the other hand, interstitial collagenase was inhibited by TIMP-1 greater than 2-fold more effectively than TIMP-2 in assays measuring cleavage of loose collagen fibrils.

To study the activation of human 72-kDa gelatinase, and its relation to tissue inhibitor of metalloproteinases 2 (TIMP-2), we purified human 72-kDa progelatinase both as a complex with TIMP-2 and as a free proteinase. Activation of progelatinase-TIMP-2 complexes with 4-aminophenylmercuric acetate yielded gelatinolytically active enzyme migrating at 62 kDa. TIMP-2 remained bound to the active enzyme. Removal of TIMP-2 from progelatinase by reverse-phase high performance liquid chromatography in the presence of trifluoroacetic acid, followed by complete dialysis in neutral pH buffer, resulted in multiple fragments. These fragments were formed as a result of the cleavage of 72-kDa progelatinase at several locations. Cleavage at the amino terminus was restricted to the removal of the propeptide, except in the case of degradation leading to inactive fragments. Two active species autocatalytically evolved upon removal of TIMP-2 from progelatinase. The 62 kDa-activated gelatinase lacked the amino-terminal propeptide, which is known to be removed upon treatment with 4-aminophenylmercuric acetate. In addition, an active 42.5-kDa fragment lacking both the propeptide and a portion of the carboxyl terminus was formed. This low-molecular-weight active form of 72-kDa progelatinase retained its ability to bind and degrade gelatin. Self-activation and degradation of 72-kDa progelatinase can be prevented by agents that inhibit metalloproteinases, including 1,10-phenanthroline. Evidence presented here suggests that TIMP-2 binds to a stabilization site that is independent of the active site. This stabilization site does not bind TIMP-1 (TIMP). Occupation of this site by TIMP-2 prevents autocatalytic activation and degradation but does not prevent gelatinolysis by the enzyme-inhibitor complex.

The abilities of the matrix metalloproteinases 95 kDa and 72 kDa gelatinases (type IV collagenases), stromelysins-1 and -2 and punctuated metalloproteinase (PUMP) to degrade insoluble elastin, type IV collagen films and proteoglycan have been compared. The gelatinases and PUMP were markedly more active in the degradation of elastin than were the stromelysins. PUMP and the stromelysins were more potent proteoglycan-degrading enzymes. All of the enzymes studied degraded soluble native type IV collagen, but the gelatinases were more effective at higher temperatures. These quantitative data allow an analysis of the potential relative roles of these metalloproteinases in the breakdown of the key components of connective tissue matrices.

Human rheumatoid synovial cells in culture secrete both 72-kDa progelatinase and a complex consisting of 72-kDa progelatinase and a 24-kDa inhibitor of metalloproteinases, TIMP-2. In addition, the culture medium contains TIMP-1, the classical inhibitor of metalloproteinases, with a molecular mass of 30 kDa. TIMP-1 does not form a complex with free 72-kDa progelatinase. Free progelatinase and progelatinase complexed with TIMP-2 can be activated with the organomercury compound p-aminophenylmercury acetate. The activated complex shows less than 10% the enzyme activity of activated free gelatinase. The progelatinase-TIMP-2 complex could be shown to be an inhibitor for other metalloproteinases, such as gelatinase and collagenase secreted by human rheumatoid synovia fibroblasts, as well as for the corresponding enzymes from human neutrophils.

Several N-carboxyalkyl peptides were synthesized and tested as inhibitors of pig synovial collagenase, 72-kDa gelatinase and stromelysin (matrix metalloproteinases MMP-1, MMP-2, and MMP-3). The most potent of the series, CH3CH2CH2(R,S)CH(COOH)-NH-Leu-Phe-Ala-NH2, competitively inhibited cleavage of dinitrophenyl-Pro-Leu-Gly-Leu-Trp-Ala-D-Arg-NH2 at the Gly-Leu bond by MMP-1 and MMP-2 (KI = 30 and 40 microM, respectively). A similar inhibitory potency was found for MMP-1 with soluble Type I collagen and MMP-3 with substance P as substrate. The inhibitor was coupled to EAH-Sepharose 4B through a C-terminal amide. In the presence of 2 M NaCl at pH 7.2, this matrix bound MMP-1, MMP-2, and MMP-3 from concentrated culture medium of pig synovial membranes. The enzymes coeluted at pH 4.1 and subsequently were resolved by chromatography on DEAE-Sephacel and heparin-Sepharose. Purified MMP-1 catalyzed the o-phenanthroline-sensitive cleavage of collagen into TCA and TCB fragments as well as slower hydrolysis of the alpha 2 chain. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of MMP-1 indicated a predominant polypeptide of approximately 44 kDa and minor species of approximately 24 and 21 kDa. The 44-kDa species and one of the smaller polypeptides reacted with an antiserum to residues 195-207 of human fibroblast MMP-1, indicating that porcine MMP-1 contains a similar sequence and that the smaller components were probably derived from MMP-1. Neither MMP-2 nor MMP-3 reacted with this antiserum. Purified porcine MMP-2 degraded gelatin but not collagen and exhibited an apparent Mr of approximately 71 kDa. Additional smaller polypeptides were present, one of which may correspond to tissue inhibitor of metalloproteinases. MMP-3 showed doublets of approximately 47/46 and 26/25 kDa and cleaved substance P at the Gly6-Phe7 bond. This procedure provides a rapid means of obtaining all three MMPs from one source in approximately 15% yield each.

Matrix metalloproteinases are an important group of zinc enzymes responsible for degradation of the extracellular matrix components such as collagen and proteoglycans in normal embryogenesis and remodeling and in many disease processes such as arthritis, cancer, periodontitis, and osteoporosis. A matrixin family is defined, comprising at least seven members that range in size from Mr 28,000 to 92,000 and are related in gene sequence to collagenase. All family members are secreted as zymogens that lose peptides of about 10,000 daltons upon activation. Latency is due to a conserved cysteine that binds to zinc at the active center. Latency is overcome by physical (chaotropic agents), chemical (HOCl, mercurials), and enzymatic (trypsin, plasmin) treatments that separate the cysteine residue from the zinc. Expression of the metalloproteinases is switched on by a variety of agents acting through regulatory elements of the gene, particularly the AP-1 binding site. A family of protein inhibitors of Mr 28,500 or less binds strongly and stoichiometrically in noncovalent fashion to inhibit members of the family. The serum protein alpha 2-macroglobulin and relatives are also strongly inhibitory.

The concentration of the metalloproteinases type I collagenase and gelatinase was measured in isolated polymorphonuclear leukocytes (PMNLs) of renal transplant recipients treated either with cyclosporin A (CyA) and prednisolone (Pr) (n = 8) or azathioprine (Aza) and Pr (n = 8), and of healthy subjects (n = 12). PMNLs of CyA- and Aza-treated transplant patients displayed markedly higher gelatinase content (2427 +/- 489 and 3284 +/- 357 ng/10(7) cells) than PMNLs of controls (528 +/- 83 ng/10(7) cells). There was also a higher content of type I collagenase in PMNLs (3374 +/- 292 ng/10(7) cells) of Aza-treated patients and significantly elevated levels in PMNLs of patients receiving CyA (3625 +/- 229 ng/10(7) cells) compared with healthy subjects (2878 +/- 151 ng/10(7) cells). In contrast, neutrophil lactoferrin content was lower in transplant patients. Thus, immunosuppressive drugs may reduce the release of leukocyte proteinases, which are known for their deleterious role in proteolytic tissue and matrix breakdown. In vitro, the effects of different immunosuppressive drugs on the release of lactoferrin, collagenase and gelatinase were investigated on FMLPNTL-stimulated PMNLs isolated from healthy subjects. CyA but not Aza or Pr caused inhibition of gelatinase, collagenase and lactoferrin release.

Extracellular matrix (ECM) turnover and remodeling are initiated, at least in part, by the regulated secretion of members of a family of matrix metalloproteinases. Human and bovine trabecular mesh-work in culture secrete interstitial collagenase, both the 72- and the 92-kD forms of type IV collagenase (gelatinases) and stromelysin, and the tissue inhibitor of metalloproteinases (TIMP). These proteinases and TIMP were identified by immunoblotting western transfers from sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using several specific antiprotein and antipeptide polyclonal antibodies. Gelatinase and stromelysin enzymatic activities were also analyzed by substrate SDS-PAGE, in which proteinase substrates were polymerized into the gels before electrophoresis to allow subsequent activity assays. These matrix metalloproteinases and TIMP are secreted at low basal levels into trabecular culture medium; their secretion levels are increased several-fold by treatment of the cultures with the phorbol mitogen. 12-O-tetradecanoylphorbol-13-acetate (TPA). Characteristics of the trabecular matrix metalloproteinases and TIMP are similar to those secreted by numerous other tissues, including the retinal pigment epithelium. These proteinases may serve an important role in the maintenance and regulation of the trabecular extracellular matrix and, subsequently, of the aqueous humor outflow pathway in normal and glaucomatous eyes.

Cultures of transformed fibroblasts actively involved in extracellular matrix degradation have been examined for initial activation of serine and metallo protease cascade systems. Rous sarcoma virus transformed chick embryo fibroblasts (RSVCEF), in contrast to transformed mammalian cells, produce active, two chain urokinase-type plasminogen activator (tcu-PA). Active tcu-PA is found in serum-free, plasmin-free conditioned medium from RSVCEF cultures as determined by two independent methods, immunoprecipitation and differential DFP sensitivity. RSVCEF cultures synthesize and secrete inactive, single chain uPA (scu-PA) which is converted to tcu-PA in a time dependent manner by a catalytic mechanism that appears to involve a functioning uPA receptor on the surface of intact cells. The enzyme activity responsible for this conversion may represent the initiating catalytic event in the PA/plasminogen serine protease cascade system. A 70 kDa prometalloprotease capable of degrading denatured collagen following its activation also is significantly elevated in RSVCEF cultures over that of normal CEF. Trace amounts of the active 62 kDa form of the metalloprotease (gelatinase) is found in the transformed RSVCEF cultures indicating that these cultures produce a natural activator of the prometalloprotease. Plasmin and/or PA do not appear to be the activator of this enzyme as determined by indirect inhibition assays and direct assays employing purified enzymes. The possible central position of pro PA and the 70 kDa prometalloprotease in an interacting, complex protease cascade system involved in extracellular matrix degradation is discussed.

Extracellular matrix turnover is initiated, at least in part, by the regulated secretion of members of a family of matrix metalloproteinases. The authors show that interstitial collagenase, stromelysin, two gelatinases (the 72-kD and 92-kD type IV collagenases), and the tissue inhibitor of metalloproteinases (TIMP) are secreted into the culture medium of human retinal pigment epithelium (RPE). These enzymes and their inhibitor were identified by probing immunoblots of western transfers with specific polyclonal antibodies that were made against these proteins or against peptides containing unique sequences from these proteins. Stromelysin and the gelatinases are also active against substrates that are polymerized into polyacrylamide gel before electrophoresis and require metal ions (probably zinc and/or calcium) for activity. The phorbol mitogen, 12-tetradecanoylphorbol-13-acetate, differentially increases the levels of these metalloproteinases and TIMP found in retinal pigment epithelium culture medium with stromelysin and the 92-kD type IV collagenase responding most strongly and TIMP actually decreasing in certain cases. Additional changes in metalloproteinase profiles are observed after approximately 20 passage of several RPE lines in culture. Modulation of extracellular matrix turnover by changing RPE secretion of these matrix metalloproteinases and their TIMP, may play a central role in the normal function and in the pathology of the retina.

We have examined the pattern of expression of four different matrix metalloproteinases (MMPs), collagenase, stromelysin, 92 kD gelatinase, and 72 kD gelatinase, by primary and passaged cultures of rabbit corneal fibroblasts. Primary cultures of this cell type have previously been shown to reproduce the normal tissue regulation of collagenase expression. We demonstrate qualitative and quantitative changes in the pattern of MMP expression as the cells are passaged in culture. Only a single MMP, 72 kD gelatinase, is constitutively expressed by primary fibroblast cultures. Phorbol myristate acetate (PMA) treatment upregulates expression of 72 kD gelatinase and turns on the expression of collagenase and stromelysin, as well as 92 kD gelatinase. However, the degree to which MMP expression is induced is minimal. Cells subcultured but a single time constitutively produce not only 72 kD gelatinase, but also collagenase and stromelysin. In addition, PMA treatment upregulates expression of collagenase, stromelysin and 92 kD gelatinase to high levels. In contrast, the expression of 72 kD gelatinase is repressed by treatment of passaged cell cultures with PMA. Our data indicate that the cell does not simply turn the MMP genes on or off, as a group, in response to various agents, but that it has the capacity for fine control over which MMPs are expressed and the degree to which each is expressed. Changes in MMP protein expression induced by PMA treatment are correlated with changes in specific mRNA levels in passaged cultures. The kinetics of mRNA accumulation suggest that the MMP genes can respond to multiple intracellular signals initiated in a temporal cascade by PMA. It is the combined effects of the individual signals on the accumulation of specific mRNAs that must determine the ultimate pattern of MMP protein expression. The distinct patterns of MMP expression produced by primary and passaged cell cultures may be analogous to patterns of expression that might occur under particular in vivo conditions.

Metalloproteinases secreted by tumor cells play an important role in metastasis. In the present study, we determined whether an inhibitor of these proteinases could inhibit the ability of tumor cells to degrade collagen and to metastasize. Metalloproteinases with degradative activities for type I collagen, type IV collagen, gelatin, and casein were secreted by a highly metastatic rat embryo cell line (4R) transfected by c-Ha-ras1 (also known as HRAS1). These metalloproteinases were identified by sodium dodecyl sulfate substrate-polyacrylamide gel electrophoresis as 92-kilodalton and 68-kilodalton gelatinolytic enzymes and 48-kilodalton and 45-kilodalton caseinolytic proteinases. A recombinant human tissue inhibitor of metalloproteinases (rTIMP) completely inhibited the proteolytic activities of these enzymes and was also a potent inhibitor of the proteolytic degradation of collagen by intact c-Ha-ras1-transfected cells. The ability of these cells to colonize the lungs after intravenous injection into nude mice was inhibited by 83% when rTIMP was repeatedly injected intraperitoneally into the animals. These data demonstrate that rTIMP is a potent inhibitor of the metalloproteinase activities of these cells and can also inhibit their metastatic potential.

We report herein that cartilage proteolytic activity increased in bovine and rabbit articular cartilage after treatment with a purified staphylococcal culture medium or intraarticular infection with Staphylococcus aureus. Staphylococcal culture medium increased the release of gelatinolytic, collagenolytic, and caseinolytic activity into the medium of isolated chondrocytes or cartilage organ culture. The proteolytic activities were determined in assays using radiolabeled substrate and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Staphylococcal culture medium was proteolytically inactive by both assay techniques. RNA synthesis of isolated chondrocytes was unaffected by staphylococcal culture medium, whereas overall protein synthesis was inhibited by 84%. An analysis of extracts of Staphylococcus aureus-infected rabbit knee cartilage by substrate gels showed increased gelatinolytic and caseinolytic activity compared with extracts of uninfected knee cartilage. Our data suggest that the rapid loss of proteoglycan and persistent degradation of cartilage in staphylococcal septic arthritis is due to the production and activation of chondrocyte proteases.

In a study of the vectorial secretion of proteins by bovine aortic arch endothelial cells, we found that the extracellular matrix macromolecules collagen and fibronectin as well as several matrix-degrading metalloproteinases were secreted selectively in the basal direction. In contrast, the tissue inhibitor of metalloproteinases showed only a weak preference for the basal direction. Three proteins at 18-35 kDa were secreted with preference apically, counter to the basal secretion of approximately 70% of the total secreted protein. As expected, rabbit synovial fibroblasts, which were used as a control, secreted proteins, including collagen, gelatin-degrading proteinases, and casein-degrading proteinases, equally in apical and basal directions. The basal secretion of collagen, fibronectin, gelatinases, and tissue inhibitor of metalloproteinases by bovine aortic arch endothelial cells suggests that the structural and functional polarity of these cells is manifested, in part, at the level of polarized secretion of matrix-related proteins.

Matrix metalloproteinases constitute the major proteolytic enzyme group degrading the extracellular matrix components. They are involved in the remodeling of the connective tissue during normal biological processes. Their abnormal expression leads to pathological situations.

The ability of human neutrophil elastase and cathepsin G to activate matrix metalloproteinase 3 (MMP-3 = stromelysin) and MMP-2 ('gelatinase') purified from human rheumatoid synovial fibroblasts in culture was examined. The zymogen of MMP-3 (proMMP-3) was activated to full activity with elastase and cathepsin G by limited proteolysis of the molecule into two active forms of Mr approximately 45,000 and Mr approximately 25,000. In contrast, proMMP-2 was not activated at all by these neutrophil serine proteinases, although it was degraded into small fragments. These data suggest that neutrophil elastase and cathepsin G may play an important role in the activation of proMMP-3 in vivo in various inflammatory conditions, but proMMP-2 may be activated in different ways.