WO2002003985A1 - Method of modulating expression of ldl-receptor-related protein and uses thereof - Google Patents

Abstract

Methods for increasing expression of the LDL receptor-related protein (LRP) in cells or animals are disclosed. The methods comprise treating the cells or animals with an HMG-CoA reductase inhibitor (statin). Such treatments are also useful for: reducing activity of LRP ligands in cells or animals, determining whether a particular condition is caused by insufficient or excess expression of an LRP, determining whether a particular protein is inactivitated by an LRP, and other similar applications.

Description

METHOD OF MODULATING EXPRESSION OF LDL-RECEPTOR-RELATED PROTEIN

AND USES THEREOF

Reference to Government Grant

This invention was made with Government support under National Institutes of Health Grant No. AR40661. The Government has certain rights in the invention.

Related Application

This application is a continuation-in-part of Application Serial No. 09/370,738 filed August 9, 1999.

Background of the Invention

(1) Field of the Invention

The present invention generally relates to the modulation of enzyme activity in a cell or animal. More specifically, the invention relates to the enhancement of levels of low- density lipoprotein receptor-related protein (LRP) in cells and animals and uses thereof in treating disorders mediated by excessive levels of proteins which bind to LRP or have receptors which bind to LRP. (2) Description of the Related Art

The low density lipoprotein receptor-related protein (LRP) is a large membrane- bound protein which is highly conserved among vertebrates. Its name is derived from its strong homology to the low density lipoprotein receptor, and belongs to the LDL receptor superfamily (Brown et al., 1997, Nature 388:629). This superfamily consists of endocytotic receptors that primarily participate in the recognition and endocytosis of lipoproteins (Brown et al., 1979, Proc. Natl. Acad. Sci. U.S.A. 7-5:3330). The LRP has a structure similar to other LDL receptor superfamily members, having a single transmembrane domain and numerous ligand-binding domains organized as cysteine-rich repeats arranged in clusters, followed by two EGF-like repeats separated from a third one by a spacer region containing a YWTD consensus sequence, and an NPXY internalization signal in the cytoplasmic domain. The LRP mRNA and protein are both found in many tissues and cell types, including liver, brain, and placenta. Recent reviews of various aspects of the structure and activity of LRP include Krieger et al., 1994, Annu. Rev. Biochem. .53:601-637 and Strickland et al., 1995, FASEB J. :890-898.

LRP was first identified as the α₂-macroglobulin (α₂M) receptor. α₂M is a plasma protein which serves as a circulating inhibitor for all known endoproteinases and other ligands. Upon ligand binding, α₂M undergoes a conformational change which exposes an LRP recognition site. The α₂M ligand complex then binds LRP, after which the complex is internalized (endocytosed) and the ligand degraded. Thus, LRP mediates the endocytosis and degradation of any α₂M ligand, including endopeptidases.

-Other α₂M ligands include coagulation factor Xa (Narita et al., 1998, Blood 91:555- 560), leptin (Birkenmeyer et al., 1998, Eur. J. Endocrinol. 139:224-230), prostate-specific antigen (Bar et al., 1998, J. Urol. 159:291-303), β-amyloid peptide (Narita et al., J. Neurochem. (59:1904-1911) and cytokines such as epidermal grc-wth factor, insulin, transforming growth factor-β, platelet-derived growth factor, interleukin-lβ, interleukin-6, basic fibroblast growth factor, and nerve growth factor (Gonias et al., 1994, Ann. N. Y. Acad. Sci. 737:213-290; Hussaini et al., 1990, J. Biol. Chem. 2-55:19441-19446). Additionally, several ligands are internalized upon either direct binding to the LRP, or where a LRP- binding receptor for the ligand has not been identified. These ligands include ApoE and ApoE-containing lipoproteins such as apoE-enriched β-VLDL and associated chylomϊcron remnants and VLDL remnants, lipoprotein lipase, lactoferrin, tissue plasminogen activator, urokinase plasminogen activator, complement component C3 (Melinger et al., 1999, J. Biol. Chem. 274:38091-38096), exotoxin A from Pseudomonas aeruginosa (Zdanovsky et al., 1996, J. Biol. Chem. 271:6122-6128), saposin precursor (Hiesberger et al., 1998, EMBO J. 17:4611-4625) and thrombospondin (Mikhailenko et al., 1995, J. Biol.Chem. 270:9543-9549). Several other proteins are ligands to their own specific receptor, and the ligand-receptor complex binds to LRP and is internalized. These proteins include Cl inhibitor-Cls complex (Storm et al., 1997, J. Biol Chem. 272:31043-31050), tissue plasminogen activator- plasminogen activator inhibitor- 1 complex, urokinase plasminogen activator-plasminogen activator inhibitor- 1 complex, and neutrophil elastase-αl-antitrypsin-elastase complex. Additionally, copending patent application 09/370,738 establishes that matrix metalloproteinases (MMPs) are ligands for specific receptors and the MMP-receptor complex is internalized by LRP. As disclosed therein, the receptor for collagenase-3 (MMP-13) is a member of the macrophage mannose receptor type C lectin family. See also Nagase et al., 1994, Ann. N. Y. Acad. Sci.732:294-302, discussing the interaction of MMPs with α₂M: Another ligand for LRP is receptor-associated protein (RAP). RAP is a 39kDa protein which copurifies with LRP and inhibits ligand binding by LRP and other members of the low density lipoprotein receptor superfamily (Kounnas et al., 1996, J. Biol Chem 271:6523). It is thought that RAP serves as an intracellular chaperone of LRP, perhaps by regulating LRP transport to the cell surface. RAP is often used in experimental studies to confirm that endocytosis is mediated by LRP. See, e.g., Example 1, demonstrating RAP inhibition of endocytosis of collagenase-3.

The low density lipoprotein receptor-related protein is known to contain multiple independent binding domains. Previously published work has shown that charged residues are required for ligand interactions with the low density lipoprotein receptor-related protein. (Sottrup- Jensen et al., 1986 FEBSLett. 205: 20). As disclosed in copending application 09/370,738, the LRP binding domain identified on collagenase-3 (136-KAFRK-140) (SEQ ID NO: 1) conforms to a published recognition motif of the low density lipoprotein receptor- related protein, consisting of two lysine residues separated by any three amino acids (KXXXK). (Nielson et al., 1996, J. Biol. Chem. 271:12909). This motif is present on α₂- macroglobulin, as well as on RAP (Ellgaard et al, 1997 FEBSLETT. 244:544). Several LRP genes have been cloned. See, e.g., the disclosure of a human LRP gene in Herz et al., 1988, EMBO J. 7:4119, which also can be found as GenBank Accession No. KM002332, and a mouse gene in van der Zee et al., 1994, Genomics 23:256-259 (1994). The promoter region of LRP has also been cloned and characterized. See, e.g., Kϋtt et al., 1989, Biochem. et Bioph.ys.Acta 1009:229-236. Unlike the promoter for the low density lipoprotein (LDL) receptor, the LRP promoter does not contain a sterol-regulatory response element and is thus not significantly repressed in response to cholesterol or other sterols. However, LRP transcription and expression is downregulated in macrophages by lipopolysaccharide, interferon-δ, and estradiol, and is increased somewhat (about 4-fold) by colony stimulating factor-1 (Hussaini et al., 1990, J. Biol Chem. 2-55:19441-19446). Because of the importance of the LRP in regulating levels of a diverse group of circulating and cellular ligands, there is a need for treatments which strongly affect levels of LRP in cells. The need is particularly acute for treatments which can greatly increase expression of LRP, since such a treatment could be used to evaluate postulated LRP binding to other ligands. A treatment which greatly increases LRP expression could also be used to determine the role of LRP ligands in disease, and to treat disorders which are caused by excessive expression of an LRP ligand.

Summary of the Invention

Accordingly, the inventor has succeeded in discovering that LRP levels can be greatly increased by treating cells with HMG-CoA reductase inhibitors, also known as statins. These agents can thus be applied in the treatment of disorders which are mediated by an excessive amount of LRP ligands.

Thus, one embodiment of the present invention is directed toward methods for increasing expression of an LDL receptor-related protein (LRP) in a cell. The methods comprise treating the cell with an effective amount of an HMG-CoA reductase inhibitor.

These methods are useful for increasing LRP binding of an LRP ligand, particularly when the cell is in a vertebrate with a condition associated with excessive levels of an LRP ligand.

The present invention is also directed toward methods of reducing activity .of an LRP ligandin a vertebrate. The methods comprise treating the animal with an HMG-CoA reductase inhibitor. Such treatments could comprise treating cells ex vivo then implanting the treated cells into the vertebrate. Alternatively, the treatments could comprise administering the HMG-CoA reductase inhibitor to the vertebrate in a pharmaceutically acceptable composition.

In an additional embodiment, the present invention is directed to methods for determining whether a particular condition in a vertebrate is caused by insufficient expression of an LRP. The methods comprise treating the vertebrate with an HMG-CoA reductase inhibitor, then assessing the status of the condition. Relief of the condition indicates that the condition is caused by insufficient expression of the LRP. The methods can further comprise the treatment of a second vertebrate with the HMG-CoA reductase inhibitor and receptor- associated protein, then assessing the status of the condition. In those cases, relief of the condition in the first vertebrate but not the second vertebrate further indicates that the condition is caused by insufficient expression of the LRP.

In additional embodiments, the present invention is directed toward methods for determining whether a disorder in a non-human vertebrate is caused by excessive levels of an LRP. These methods comprise treating the vertebrate with a statin and determining whether the treatment causes the disorder. Here, a determination that the treatment causes the disorder indicates that the treatment is caused by excessive levels of an LRP.

The present invention is also directed toward methods for determining whether a protein in a cell is an LRP ligand. The methods comprise comparing active protein levels before and after treating the cell with a statin, where lower levels of the active protein after statin treatment indicate that the protein is an LRP ligand.

Additionally, the present invention is directed toward methods of producing a recombinant protein in a cell in response to statin induction. The methods comprise (a) creating a cell comprising a recombinant polynucleotide encoding the protein operably linked to an LRP promoter; and (b) exposing the cell to a statin and culturing the cell under conditions and for a time sufficient to produce the protein.

Among the several advantages achieved by the present invention, therefore, may be noted the provision of: methods for increasing expression of LRP; methods for reducing activity of an LRP ligand; methods of treating a vertebrate with a disorder mediated by excessive activity of an LRP ligand; methods for determining whether a particular condition is mediated by excessive activity of an LRP ligand; methods for determining whether a particular condition is caused by insufficient expression of an LRP; methods for determining whether a disorder in a nonhuman vertebrate is caused by excessive levels of an LRP; . methods for determining whether a protein in a cell is an LRP ligand; and methods of producing a recombinant protein in a cell in response to statin induction.

Brief description of the Drawings

Figure 1 depicts varying degrees of binding of ¹²⁵I-rat collagenase-3 to the following cell lines: UMR 106-01 rat osteosarcoma cells; ROS 17/2.8 rat osteosarcoma cells; SAOS-2 human osteosarcoma cells; BC-1 rat breast carcinoma cells; NTH 3T3 mouse fibre-blasts; rat fibroblasts (FB); and normal mineralizing rat osteoblasts (NRO).

Figure 2 depicts displacement of rat collagenase-3 binding to UMR cells by human collagenase-3 (open circles) and rat collagenase-3 (closed circles).

Figure 3 depicts specific binding of collagenase-3 to its receptor and the absence of such binding when Ca²⁺ is not present. Figure 4 depicts binding to electrophoresed UMR 106-01 membranes by: ^l25I-rat collagenase-3 using a ^I25I-rat collagenase-3 probe without added unlabeled rat collagenase-3 (left lane); ^I25I-rat collagenase-3 with unlabeled rat collagenase-3 added, showing displacement of radioactive binding (center lane); low density lipoprotein receptor-related protein antibodies by Western blot (right lane).

Figure 5 depicts northern hybridizations with low density lipoprotein receptor-related protein (LRP) mRNA (upper panel) and β-actin mRNA (lower panel), showing the presence of low density lipoprotein receptor-related protein mRNA in rat osteoblastic cells (UMR 106-

01) and MEF-1 mouse fibroblastic cells but not MEF-2 mouse fibroblastic cells. Figure 6 depicts the binding of ¹²⁵I-rat collagenase-3 to both low density lipoprotein receptor-related protein and rat collagenase receptors, showing electrophoresed cell membranes from UMR 106-01, MEF-1 and MEF-2 probed with: ^I25I-rat collagenase-3 in the presence or absence of unlabeled rat collagenase-3 (Panels 1 and 2, respectively); ¹²⁵I- ' receptor-associated protein (RAP) in the presence or absence of unlabeled RAP (Panels 3 and 4, respectively); and anti-low density lipoprotein receptor-related protein antibodies (Panel 5). Figure 7 depicts the equivalent binding of collagenase-3 to UMR 106-01 cells and mouse embryo fibroblasts (MEF-1 and MEF-2) which have (MEF-1) and which lack (MEF-

2) the low density lipoprotein receptor-related protein.

Figure 8 depicts the inhibition of binding of ^12SI-rat collagenase-3 to UMR 106-01 cells by excess non-radioactive collagenase-3, and the lack of such an effect by receptor- associated protein (RAP).

Figure 9 depicts the time course of internalization of rat collagenase-3 by MEF-1 cells (closed circles) and the lack of such internalization by MEF-2 cells (open circles).

^"Figure 10 depicts the inhibition of ^,25I-rat collagenase-3 internalization in UMR 106- 01 cells by increasing concentrations of receptor-associated protein (RAP).

Figure 11 A depicts the inhibition of internalization of ^I25I-rat collagenase-3 by receptor-associated protein (RAP) in UMR 106-01 cells.

Figure 1 IB depicts the inhibition of internalization of ¹²⁵I-rat collagenase-3 by receptor-associated protein (RAP) in normal rat osteoblasts. Figure 12A depicts the time course of ^12SI-rat collagenase-3 internalization in UMR

106-01 cells in the presence (open circles) or absence (closed circles) of receptor-associated protein (RAP).

Figure 12B depicts the time course of ¹²⁵I-receptor-associated protein (RAP) in the presence (open circles) or absence (closed circles) of rat collagenase-3. Figure 13 A depicts the total, nonspecific, and specific binding of varying concentrations of ^,25I mouse collagenase-3 to UMR 106-01 rat osteosarcoma cells.

Figure 13B depicts a Scatchard analysis of the data shown in Figure 13 A, indicating binding by two receptors. Figure 14 shows schematic representations of chimeric collagenase molecules, where human collagenase-1 (MMP-1) sequences are shaded, and mouse collagenase-3 (MMP-13) sequences are unshaded.

Figure 15A depicts electrophoresis of various chimeric collagenase proteins where lane 1 is full-length mouse collagenase-3 (1-472); lane 2 is C-terminally truncated mouse collagenase-3 (1-265); lane 3 is MH(213-267) M (exon 5 replacement); lane 4 is HM/M(141- 472); lane 5 is M(l-228)/H; lane 6 is HM(141-228)/H; lane 7 is HM(166-228)/H; and lane 8 is H/M(229-472).

Figure 15B depicts gelatin zymography performed on each protein from Figure 15A, showing enzymatic activity of the chimeric collagenases. Figure 16 depicts collagenase chimera binding displacement of ¹²⁵I collagenase-3 in

UMR 106-01 cells by chimeric collagenases M(l-228)/H, HM(141-228)/H, HM(141-472), MH(213-267)/M, and M(l-472), but not H/M(229-472).

Figure 17 depicts reduced chimera binding displacement of ¹²⁵I collagenase-3 in UMR 106-01 cells by HM(166-228) H (closed circles) when compared to wild-type mouse collagenase-3.

Figure 18 depicts an amino acid sequence comparison of receptor binding domains between receptor-binding MMP-13 homologs and MMPs which do not bind to receptors on UMR cells (human MMP-1, -2, -3, and -9).

^~ Figure 19A depicts the binding of ^I25I labeled collagenase-3, and the reduced binding of M(l-228) H, and HM(141-228)/H to UMR 106-01 cells. The inset graph depicts the internalization of ¹²⁵I labeled collagenase-3, and the reduced internalization of M(l-228)/H, and HM(141-228)/H in UMR 106-01 cells.

Figure 20 illustrates the phenotyping of cultured normal and osteoarthritic human cells by RT-PCR using primers which amplify genes specific for synoviocytes (collagen I) and chondrocytes (aggrecan and collagen II), showing that collagenase-3 was amplified only from osteoarthritic chondrocytes.

Figure 21 depicts the binding of ¹²⁵I collagenase-3 to non-arthritic (control) cells and the reduced binding to osteoarthritic (OA) cells. Figure 22 depicts the internalization of collagenase-3 by non-arthritic human chondrocytes and the reduced internalization by osteoarthritic (OA) chondrocytes, as well as the inhibition of internalization by receptor-associated protein (RAP).

Figure 23 depicts the excretion of degraded collagenase-3 from non-arthritic (control) human chondrocytes and synoviocytes and the reduced excretion from osteoarthritic (OA) chondrocytes and synoviocytes.

Figure 24 depicts the recovery of collagenase-3 receptor activity by osteoarthritic chondrocytes following treatment without (open circles) or with (closed circles) pravastatin. Figure 25 depicts total (closed circle), specific (open circle), and non-specific (triangle) binding of ¹²⁵I collagenase-3 to T/AC-62 immortalized human chondrocytes^ showing concentration-dependent specific binding.

Figure 26 depicts northern hybridization of low density lipoprotein receptor-related protein (LRP) mRNA (upper panel) and β-actin mRNA (lower panel) in total mRNA from osteoarthritic chondrocytes treated (right side) and untreated (left side) with pravastatin.

Description of the Preferred Embodiments

The present invention is based on the discovery that transcription and expression of low density lipoprotein receptor-related protein (LRP) is greatly increased by treatment with HMG-CoA reductase inhibitors. HMG-CoA reductase inhibitors, also known as statins, are compounds which are widely used to lower serum cholesterol. However, statins have not previously been reported to affect expression of LRP. Non-limiting examples of statins useful for the present invention include lovastatin, pravastatin, simvastatin, atorvastatin, mevastatin, fluvastatin, and cerivastatin. It is believed that all HMG-CoA reductase inhibitors are effective in increasing expression of LRP. Statins inhibit HMG-CoA reductase, which catalyzes the rate-limiting step in cholesterol biosynthesis. Several reviews on these widely used compounds have been published. See, e.g., pp. 884-888 in Witztum, "Drugs used in the treatment of hyperlipoproteinemias", Chapter 36 of Goodman & Gilman's The Pharmacological Basis of Therapeutics, Ninth Ed.. Hardman, Limbird, Molinoff, Ruddon, Gilman eds., McGraw-Hill, New York, 1996. Statins competitively inhibit biosynthesis of cholesterol, reducing synthesis of cholesterol in the liver. This lowers the sterol pool in hepatocytes, where cholesterol is .made. Consequently, LDL receptor levels increase due to the sterol-regulatory response element present in the LDL receptor promoter. However, the LRP promoter does not have a sterol-regulatory response element. Therefore, the skilled artisan would not expect LRP levels to increase in response to statin treatment. Aside from their ability to inhibit cholesterol biosynthesis, statins have recently been shown to activate the promoter for bone morphogenetic protein-2 and, as a consequence, increase bone formation in rodents (Mundy et al., 1999, Science 286: 1946-1949). However, this finding would also not suggest to the skilled artisan that statins would have any effect on _ LRP expression.

As discussed in copending patent application 09/370,738, the effect of statins on LRP was discovered while determining the mechanism for the statin-induced increase in endocytosis of collagenase-3, a matrix metalloproteinase. It was discovered in those studies that collagenase-3 is endocytosed (internalized) after binding to a specific collagenase-3 receptor, which was discovered to be an LRP ligand. See examples 1-3 therein and irithis application. Based on the findings disclosed in copending application 09/379,738, it was believed that the mechanism of statin-induced increases in endocytosis of collagenase-3 was due, at least in part, to increased transcription of LRP. That belief is confirmed herein. See, e.g., example 5, demonstrating that pravastatin treatment of osteoarthritic chondrocytes greatly increases LRP mRNA levels. The skilled artisan would also expect these results to be repeatable using any HMG-CoA reductase inhibitor or any cell type that has the LRP gene. It would also be expected that treatment in vivo of any vertebrate with an HMG-CoA reductase inhibitor would induce an increase in transcription and expression of LRP.

As previously discussed, treatment of collagenase-3-containing cells with statins increases endocytosis of collagenase-3, at least in part by increasing LRP levels. Since LRP binds and internalizes a wide array of ligands, it would be expected that statin treatment would also increase endocytosis of any of these ligands which encountered cells expressing the increased levels of LRP. Non-limiting examples of ligands which are bound and- interrlalizedby LRP include endoproteinases, coagulation factor Xa, leptin, prostate-specific antigen, β-amyloid peptide, epidermal growth factor, insulin, transforming growth factor-β, platelet-derived growth factor, interleukin-lβ, interleukin-6, basic fibroblast growth factor, nerve growth factor, ApoE, ApoE-containmg lipoproteins, lipoprotein lipase, lactoferrin, tissue plasminogen activator, urokinase plasminogen activator, complement component C3, exotoxin A from Pseudomonas aeruginosa, saposin precursor, thrombospondin, Cl inhibitor- Cl s complex, tissue plasminogen activator-plasminogen activator inhibitor-1 complex, urokinase plasminogen activator-plasminogen activator inhibitor-1 complex, neutrophil elastase-αl-antitrypsin-elastase complex, and matrix metalloproteinases including collagenase-3. The skilled artisan would also expect that there will be other unknown LRP ligands which will be discovered in the future. It would also be expected that binding and 5 endocytosis of those ligands would also be increased by statin treatment. Thus, in one aspect of the invention, methods are provided for increasing expression of an LRP in a vertebrate cell. The methods comprise treating the cell with an effective amount of an HMG-CoA reductase inhibitor (statin). The cells can be any cells that are capable of expressing LRP. The cells can be treated in vitro, such as in cell culture. Ex vivo applications are also envisioned, such as in cells which are removed from the vertebrate, treated, then implanted into the same or a different vertebrate animal. Additionally, the cells can be treated in vivo. Depending on the cell type, cell source, and statin utilized, statin concentrations of 1-100 μM are preferred. More preferred concentrations are 2-50 μM; even more preferred are concentrations of 5-20 μM; the most preferred statin concentration is 10 μM. For any particular cell type and statin, optimal statin concentration for increasing LRP expression can be readily discerned without undue experimentation. See, e.g., Example 5, where expression of LRP is evaluated by northern hybridization. The skilled artisan would know that other methods for evaluating LRP expression, such as dot blot hybridization, could also be used to evaluate the effects of any particular statin treatment. These methods encompass any increase in LRP expression which is above that of untreated cells. Preferably, the statin treatment increases expression of the LRP by at least two-fold. More preferably, expression is increased at least five fold. In the most preferred embodiment, the methods involve an increase in LRP expression of about 20-fold. These methods would be useful with any vertebrate cell which is capable of expressing an LRP. As used herein, a vertebrate is any member of the subphylum Vertebrata, a primary division of the phylum Chordata that includes the fishes, amphibians, reptiles, birds, and mammals, including humans, all of which are characterized by a segmented boriy or cartilaginous spinal column.

^~ Where in vivo applications of the present invention are utilized, compositions comprising the statin can be administered to the vertebrate by any suitable route known in the art including, for example, intravenous, subcutaneous, intramuscular, transdermal, intrasynovial, intrathecal, or intracerebral. The compositions can also be administered to target cells directly in ex vivo treatment protocols. Administration can be either rapid as by injection or over a period of time as by slow infusion or administration of a slow release formulation. For treating cells in the central nervous system, administration can be by injection into the cerebrospinal fluid.

It is contemplated that the statin compositions useful for the present invention are usually employed in the form of pharmaceutical preparations. Such preparations are made in a manner well known in the pharmaceutical art. One preferred preparation utilizes a vehicle of physiological saline solution, but it is contemplated that other pharmaceutically acceptable carriers such as physiological concentrations of other non-toxic salts, five percent aqueous glucose solution, sterile water or the like may also be used. It may also be desirable that a suitable buffer be present in the composition. Such solutions can, if desired, be lyophilized and stored in a sterile ampoule ready for reconstitution by the addition of sterile water for ready injection. The primary solvent can be aqueous or alternatively non-aqueous._ The carrier can also contain other pharmaceutically-acceptable excipients for modifying or maintaining the pH, osmolarity, viscosity, clarity, color, sterility, stability, rate of dissolution, or odor of the formulation. Similarly, the carrier may contain still other pharmaceutically-acceptable excipients for modifying or maintaining release or absorption or penetration across the blood-brain barrier. Such excipients are those substances usually and customarily employed to formulate dosages for parenteral administration in either unit dosage or multi-dose form or for direct infusion by continuous or periodic infusion.

It is also contemplated that certain formulations comprising the agent are to be administered orally. Such formulations are preferably encapsulated and formulated with suitable carriers in solid dosage forms. Some examples of suitable carriers, excipients, and diluents include lactose, dextrose, sucrose, sorbitol, mannitol, starches, gum acacia, calcium phosphate, alginates, calcium silicate, microcrystalline cellulose, polyvinylpyrrolidone, cellulose, gelatin, syrup, methyl cellulose, methyl- and propylhydroxybenzoates, talc, magnesium, stearate, water, mineral oil, and the like. The formulations can additionally include lubricating agents, wetting agents, emulsifying and suspending agents, preserving agents, sweetening agents or flavoring agents. The compositions may be formulated so as to provide rapid, sustained, or delayed release of the active ingredients after administration to the vertebrate by employing procedures well known in the art. The formulations can also contain substances that diminish proteolytic and nucleic acid degradation andor substances which promote absorption such as, for example, surface active agents.

The agent is administered to vertebrates in an amount effective to increase LRP expression in target cells within the vertebrate. The specific dose is calculated according to the approximate body weight or body surface area of the patient or the volume of body space to be occupied. The dose will also be calculated dependent upon the particular route of administration selected and the amount of increased LRP expression desired. Further refinement of the calculations necessary to determine the appropriate dosage for treatment is - routinely made by those of ordinary skill in the art. Such calculations can be made without undue experimentation by one skilled in the art in light of the activity disclosed herein in LRP assays. Exact dosages are determined in conjunction with standard dose-response studies. It will be understood that the amount of the composition actually administered will be determined by a practitioner, in the light of the relevant circumstances including the condition or conditions to be treated, the choice of composition to be administered, the age, weight, and response of the individual animal, the severity of the symptoms, and the chosen route of administration. Dose administration can be repeated depending upon the pharmacokinetic parameters of the dosage formulation and the route of administration used.

These methods are useful for increasing binding and endocytosis of LRP ligands. As used herein, the term "LRP ligand" refers to any molecule which specifically binds to LRP, either directly or through another molecule such as β₂-macroglobulin (β₂M) or a specific receptor for the LRP ligand. In some cases, the treated cell may also produce a receptor for the LRP ligand. However, for many applications, the presence of the ligand receptor on the treated cell is not necessary to achieve increased binding and endocytosis of the ligand, since many LRP ligands bind to LRP directly, and other ligands bind to circulating receptors such as β₂M.

Thus, in an additional embodiment, the present invention is directed to methods of reducing activity of an LRP ligand. The methods comprise treating a cell which is capable of expressing LRP with a statin as described above. As in the previous method, the cell can be treated in vitro, ex vivo, or in vivo.

Reduction of an LRP ligand is particularly useful in a vertebrate which suffers from a condition characterized by excessive production of the ligand. Nonlimiting examples of such conditions include: Alzheimer's disease, where increased endocytosis of LRP ligands β- amyloid peptide (Narita et al., Id) and apoE could be useful treatments (See, e.g., Mahley, 1996, Isr. J. Med. Sci. 32:414-429); excessive thrombosis, where increased endocytosis of coagulation factor Xa could be useful; excessive complement activation, where increased endocytosis of complement component C3 could be useful; and osteoarthritis, where increased endocytosis of collagenase-3 could be useful. Even where relief of a particular condition is not established, these invention methods can be useful in research and development of effective treatments.

Thus, the present invention is also directed to methods of treating a vertebrate with a disorder associated with excessive activity of an LRP ligand. The methods comprise treating the vertebrate with a statin as described above.

Similarly, the present invention is directed toward methods for determining whether a particular condition is caused by insufficient expression of an LRP. In this embodiment, the method comprises treating a vertebrate afflicted by the condition with a statin. The status of the condition is then evaluated. Relief of the condition coincident with statin treatment indicates that the condition is caused by insufficient expression of LRP. To further confirm LRP involvement, the vertebrate is treated with RAP. RAP treatment would inhibit the LRP- induced relief of the condition; thus, RAP treatment which prevents statins from relieving the condition further confirms that insufficient expression of LRP is a cause of the condition. In an additional embodiment, the present invention is directed toward methods for determining whether a disorder in a non-human vertebrate is caused by excessive levels of an LRP. These methods comprise treating the vertebrate with a statin and determining whether the treatment causes the disorder. Here, a determination that the treatment causes the disorder indicates that the treatment is caused by excessive levels of an LRP.

The present invention is also directed toward methods for determining whether a protein in a cell is an LRP ligand. The methods comprise comparing active protein levels before and after treating the cell with a statin, where lower levels of the active protein after statin treatment indicate that the protein is an LRP ligand. Such a finding can be confirmed by including RAP in some treatments, which would inhibit the effect of the statin if LRP is involved. In this embodiment, the treatment of cells by statins is performed as disclosed above.

In additional embodiments, the invention is directed toward methods of producing a recombinant protein in a cell in response to statin induction. The methods comprise (a) creating a cell comprising a recombinant polynucleotide encoding the protein operably linked to an LRP promoter; and (b) exposing the cell to a statin and culturing the cell under conditions and for a time sufficient to produce the protein. In these embodiments, preferred cells are eukaryotic cells; more preferred cells are vertebrate cells; most preferred cells are mammalian cells. The recombinant polynucleotide in these embodiments can be on an extrachromosomal element, such as (a) normal cellular sites of extrachromosomal genes such as the mitochondria; (b) a recombinant virus; and (c) a self-replicating nucleic acid such as a plasmid. Alternatively, the recombinant polynucleotide can be present on a chromosome. The protein encoded by the recombinant gene is not narrowly limited, and may include any protein which is compatible with the statin-induced cell. Additionally, the recombinant polynucleotide can include other genes, such as genes which assist in the transformation of the recombinant cell and/or maintenance of the recombinant polynucleotide in the cell, for example an antibiotic resistance gene or a gene complementing a deletion in a required cellular gene.

Preferred embodiments of the invention are described in the following examples. Other embodiments within the scope of the claims herein will be apparent to one skilled in the art from consideration of the specification or practice of the invention as disclosed herein. It is intended that the specification, together with the examples, be considered exemplary only, with the scope and spirit of the invention being indicated by the claims which follow the examples.

The procedures disclosed herein which involve the molecular manipulation of nucleic acids are^' known to those skilled in the art. See generally Fredrick M. Ausubel et al. (1995), "Short Protocols in Molecular Biology", John Wiley and Sons, and Joseph Sambrook pt al. (1989), "Molecular Cloning, A Laboratory Manual", second ed., Cold Spring Harbor Laboratory Press, which are both incorporated by reference.

Example 1 This example describes the characterization of the collagenase-3 endocytotic receptor system using procedures which are applicable to identifying agents useful for increasing endocytosis of matrix metalloproteinases, as well as identifying LRP ligands.

The following reagents were used in this example. From Sigma Chemical Co., St. Louis, MO: ascorbic acid, bovine serum albumin (BSA), chloramine T, proteinase E (Pronase®-E), sodium iodide, sodium metabisulfite, Tween 20 and Tween 80, isopropylthio- β-D-galactoside, glutathione and glutathione-agarose, thrombin inhibitor, 3-(( 3- cholamidopropyl) dimethyl-ammonio)-l-propanesulfonate (CHAPS), insulin, transferrin. Also used were Na¹²⁵I and ECL immunoblotting detection kit from Amersham, Kalamazoo, MI, bovine serum thrombin from Boehringer Mannheim, Basel, Switzerland, SDS-PAGE materials and non-fat dry milk from Bio-Rad, Hercules, CA, and cell culture media, fetal bovine serum (FBS) and other cell culture reagents from the Washington University Tissue Culture Support Center, St. Louis, MO.

Previous work has shown that the rat collagenase-3 endocytotic receptor system is present on UMR 106-01 rat osteosarcoma cells (Omura et al, 1994, J. Biol. Chem. 2ό^"P:24994). To determine whether the receptor is present in other cells, several cell types were tested for their ability to bind rat collagenase-3. These cell types were normal rat osteoblasts, rat embryo fibroblasts, the rat epithelial breast carcinoma cell line BC-1, mouse NTH 3T3 fibroblasts, the human osteosarcoma cell line SAOS-2, and the rat osteosarcoma cell lines ROS 17/2.8 and UMR 106-01. The human osteosarcoma cell line SAOS-2 (ATCC HTB 85) and the mouse embryo fibre-blast cell line NTH 3T3 (ATCC CRL 1658) were cultured according to recommendations of the American Type Culture Collection, Rockville, MD. •UMR 106-01 rat osteosarcoma cells were cultured as described in Roswit et al., 1992, Arch. Biochem. Biophys. 292:402, but 5% FBS was used instead of 10% FBS. The rat breast carcinoma BC-1 cell line was cultured in 1:1 DME:Ham's F12 medium with 25 mM HEPES, pH 7.1, 5 μg/ml insulin, 1 μg ml transferrin, 5 mg/ml BSA, 10 units penicillin/ml and 10 μg streptomycin/ml. The rat osteosarcoma cell line ROS 17/2.8 was cultured in Ham's F12 medium with 5% FBS, 1% glutamine, 10 units penicillin/ml, 10 μg streptomycin/ml, 80 mM CaCl₂, 25 mM HEPES. Normal rat osteoblasts (NRO) were isolated from newborn rat calvariae as described in Shalhoub et al., 1992, J. Cell. Biochem. 50:425, and cultured in Eagle's minimal essential medium (MEM) containing 10% FBS, nonessential amino-acids, 10 units penicillin/ml, 10 μg streptomycin/ml. After cells reached confluence, the culture medium was changed to BGJ_b medium containing 10% FBS, 10 units penicillin/ml, 10 μg streptomycin/ml, 50 μg/ml ascorbic acid and 2.16 mg/ml β-glycerophosphate to allow differentiation and mineralization. For all binding experiments, cells were seeded into 2.0 cm² wells. After the cells reached approximately 95% confluence, the medium was replaced with fresh medium containing 1 mg/ml BSA and the cells were assayed for binding 4 h later. The cells were first washed with maintenance medium, then incubated in the same medium with 0.01% Tween 80 containing ^,25I-rat collagenase-3 or other iodinated ligands at 4°C for 2 h. Non-specific binding was assessed by adding a 50-100-fold excess of cold ligand to half the wells, while an equivalent volume of buffer was added to the remaining wells. After incubation, the wells were washed three times with ice-cold MEM (0.5 ml). The cells were then lysed with 500 μl of 1 M NaOH and the lysates were counted on a gamma counter. Protein labeling with ¹²⁵I was done using the chloramine T method (Greenwood et al., 1963, Biochem. J. 89: 114). The proteins had specific activities ranging from 9 to 27 μCi/μg.

Figure 1 shows the results of the binding assay. The binding of rat collagenase-3 to normal rat osteoblasts and normal rat embryo fibroblasts was higher than binding to the UMR 106-01 cells. Very low levels of binding were observed in rat epithelial breast carcinoma cells, BC-1, mouse NTH 3T3 fibroblasts and human osteosarcoma cells, SAOS-2. Osteoblastic cells in vitro can secrete a number of matrix metalloproteinases including collagenase-3 (Partridge et al, 1987, Endocrinology 72,0:1956; Heath et al., 1984, Biochem. Biophys. Ada 802:151; Varghese et al., 1994, Endocrinology 134:2438; Meikle et al., 1992, J. Cell Sci. 103: 1093), 72-kDa and 92-kDa gelatinase (Meikle et al., Id.; Rifas et al., 1989, J. Clin. Invest. 84:686; Lorenzo et al., 1992, Matrix 12:282; Thomson et al., 1987, Biochem. Biophys. Res. Commun. 148:596) and stromelysin-1 (Meikle et al., Id.). These enzymes are thought to play an active role in extracellular matrix remodeling in bone tissue. Competition experiments have shown that various proteins are not able to compete with rat collagenase (Omura et al., 1994,7. Biol. Chem. 269:24994). However, human collagenase-1 (MMP-1) was the only matrix metalloproteinase which had been tested in those studies. In order to evaluate the specificity of the rat collagenase receptor in UMR 106-01 rat osteosarcoma cells, the ability of these cells to bind other matrix metalloproteinases was investigated. Ligand binding assays were performed using rat collagenase-3 (rat MMP-13) (isolated from media of cultures of post-partum rat uterine smooth muscle cells as described in Roswit et al., 1983, Arch. Biochem. Biophys. 225:285), human fibroblast collagenase - 1 (MMP-1) (purified by Dr. Howard Welgus, Washington University, St. Louis, MO), human stromelysin - 1 MMP-3) (from Dr. Paul Cannon, Syntex, Palo Alto, CA), human collagenase - 3 (human MMP-13) (produced by Dr. Carlos Lopez-Otin in a vaccinia virus based expression system as described in Freije et al., 1994, J. Biol. Chem. 269:16166), human 92 kDa gelatinase (MMP-9) and human 72 kDa gelatinase (MMP-2). Human 92 kDa and 72 kDa gelatinases were kind gifts from Dr. Howard Welgus.

As shown in Table 1, only human collagenase-3 was comparable to rat collagenase-3 in binding to UMR cells. This was expected since human collagenase-3 has 86% homology to rat collagenase-3 (Freije et al., 1994, J. Biol. Chem. 269:16166). Human collagenase-3 also competes effectively with ¹²⁵I-rat collagenase-3 for binding to the collagenase receptor (Figure 2). This result argues for the existence of a specific receptor for collagenase-3 on osteoblastic cells, in contrast to collagenase- 1, which has never been observed to be produced by these cells, nor to bind or compete for binding in these cells.

Table 1. Analysis of ^12SI-labeled proteinases binding to UMR cells. The displayed values represent means ±SEM for triplicate wells.

The binding assay described above was utilized next on UMR 106-01 cells using ¹²⁵I- rat collagenase in the presence and absence of Ca²⁺in order to investigate the requirements of ligand-receptor interaction for this ion. The results (Table 2, Figure 3) show that Ca²⁺' is necessary for rat collagenase-3 binding to its receptor. Table 2. Binding of collagenase-3 to its receptor requires Ca²⁺. The values displayed represent means ±SEM.

To determine the molecular weight of the rat collagenase-3 receptor, ligand blotting was performed using partially purified UMR 106-01 cell membranes. Cell membranes were prepared by differential centrifugation of homogenized cells at 1,000 x g for 10 min, 10,000 x g for 10 min, 100,000 x g for 40 min in buffer containing 20 mM Tris-HCl, pH 7.5, 2 mM MgCl₂, 0.25 M sucrose, 1 mM PMSF. The 100,000 x g membrane pellet was then resuspended in buffer containing 50 mM Tris-HCl, pH 8.0, 2 mM CaCl₂, 80 mM NaCl. The samples of cell membranes were subjected to 4- 15% SDS-P AGE under non-reducing conditions at 50 V for 3 h and then electrotransferred to PVDF filters in transfer buffer containing 10% methanol, 192 mM glycine, 56 mM Tris at 15 V for 16 h at 4°C. The filters were blocked with 5% non-fat dried milk in buffer containing 50 mM Tris-HCl, pH 8.0, 80 mM NaCl, 2 mM CaCl₂ and 0.1% Triton X-100 (binding buffer) for 1 h at room temperature. The filters were then incubated for 16 h at 4°C in the same buffer supplemented with 1% nonfat dried milk in the presence of 20 pmol ¹²⁵I-rat collagenase-3 in the presence or absence of the same unlabeled ligands (30-40-fold excess of rat collagenase-3). The filters were then washed with the same buffer, dried and subjected to autoradiography. For Western blot analysis, the filters were wetted with methanol for 2 sec, rinsed with H₂0 and equilibrated with buffer containing 20 mM Tris-HCl, pH 7.6, 137 mM NaCl, 0.1% Tween 20. The filters were then incubated 2 h at room temperature in the same buffer containing 5% non-fat dried milk. Subsequently, the filters were incubated with anti-LRP antibodies (1:2,000) (rabbit polyclonal antibody - generously provided by Dr. Dudley Strickland, American Red Cross, Rockville, MD) in the same buffer containing 1% non-fat dried milk for 16 h at 4°C. A 1 : 10,000 dilution of HRP-conjugated goat anti-rabbit IgG in the same buffer containing 1% non-fat dried milk was incubated with the filters for 1 h at room temperature to detect the primary antibodies. Detection was performed using an ECL kit.

These ligand blot studies show that ^12SI-labeled rat collagenase-3 binds to two proteins with molecular weights of about 600 kDa and 170 kDa (the closed circle and star, respectively, in Figure 4, left lane). ¹²⁵I-collagenase binding was highly specific, since a 40- fold excess of unlabeled rat collagenase abolished binding to both proteins (Figure 4, center lane).

As described previously (Walling et al., 1998, J. Cell Physiol. 177:563), rat collagenase-3 undergoes a process of binding, internalization and degradation following secretion from UMR 106-01 cells. It was hypothesized that the mechanism might be similar to the internalization of the members of the LDL receptor superfamily (see, e.g., Brown et al., 1983, Cell 32:663). Therefore, it was proposed that one of the proteins which showed collagenase-3 binding on ligand blot analysis might be a member of the LDL receptor superfamily. Among members of this superfamily, only two have molecular weights around 600 kDa: LRP and gp330/megalin. None of the members of the LDL receptor superfamily has a molecular weight of about 170 kDa. Western blotting with anti-LRP antibodies shows that the 600 kDa protein is the large subunit of the LRP (Figure 4, right lane, closed circle). Anti-LRP antibodies also detected the small subunit of the LRP (Figure 4, right panel,'open circle). In order to exclude the possibility that the collagenase-3 receptor is the LRP, two cell lines of mouse embryo fibroblasts were used: wild-type (MEF-1) and LRP-null (MEF-2) (both generously provided by Dr. Joachim Herz). These cells were cultured in Dulbecco's MEM with 10% FBS, 10 units penicillin/ml, 10 μg streptomycin/ml.

To determine whether MEF-2 cells had LRP mRNA Northern blot analysis was performed as follows. Poly^^-R A was isolated from 2 x 10⁸ of each of UMR 106-01, MEF-1 and MEF-2 cells using the mRNA purification kit from Invitrogen, Carlsbad, CA. Five μg of mRNA from each of UMR 106-01, MEF-1 and MEF-2 cells was separated by electrophoresis in 0.5% agarose formaldehyde (2.2 M) gel. The RNA was UV-crosslinked to a Zeta-=Probe GT membrane (Bio-Rad) after upward capillary transfer. The 5.99 kb fragment of LRP in pGEM-4 vector (ATCC 65430) was used as a probe for identification of LRP mRNA. The plasmid with insert was labeled using the nick-translation kit from Promega. β- actin cDNA was labeled by random priming using the Promega (Madison, WT) Prime-a-Gene kit. Prehybridization and hybridization of both LRP and β-actin probes was carried out at 42°C in 50% formamide, 5 x SSC, 0.2% each of BSA, Ficoll and PVP, salmon sperm DNA (250 μg ml), 0.1% SDS and Na pyrophosphate, pH 6.5 (50 mM) with 10⁶ cpm/ml of each probe for 16 h. The filter was washed in 2 x SSC, 0.1% SDS for 4 x 5 min at room temperature, followed by 0.1 x SSC, 0.1% SDS for 15 inin at 50°C. The Northern blot analysis shows that both UMR 106-01 and MEF-1 cells express LRP, while MEF-2 cells do not (Figure 5). Ligand blot and Western blot analyses further show that ¹²⁵I-rat collagenase-3 specifically binds to the large subunit of the LRP in UMR 106-01 and MEF-1, but not MEF-2 cell membranes (Figure 6, panels 1,2 and 5, closed circles). Additionally, ¹²⁵I-RAP binds to only the large subunit of the LRP in UMR 106-01 and MEF-1 cell membranes (Figure 6, panels 3, 4 and 5). However, all three of these cell lines show binding of ^I25I-collagenase-3 to the 170 kDa protein (Figure 6, panel 1, star). Also, both MEF-1 and MEF-2 cells have an additional protein with molecular weight of approximately 200 kDa which specifically binds ¹²⁵I-rat collagenase-3 (Figure 6, panel 1, triangle).

¹²⁵I-collagenase-3 binding assays were performed with MEF-1, MEF-2 and UMR 106-01 cells. The results show no significant difference in binding between wild-type and LRP-deficient cells, suggesting that the LRP is not required for collagenase-3 binding to these cells (Figure 7). RAP also does not inhibit ¹²⁵I-rat collagenase binding to the UMR cells (Figure 8), although it is known to inhibit binding of most ligands for the LRP. These data indicate that the 170 kDa protein is a specific receptor for collagenase-3 in UMR 106-01 cells. Although the LRP is not required for rat collagenase-3 binding to the cell, it might be required for ligand internalization. Therefore, internalization assays were performed with ¹²⁵I-rat collagenase-3 using MEF-1 and MEF-2 cells as follows. After binding ¹²⁵I-labeled proteins as above, the cells were washed three times with cold modified Eagle's media (0.5 ml) to remove unbound ligand. The cells were then warmed to 37°C by the addition of prewarmed modified Eagle's media (0.25 ml), and incubated at 37°C for selected intervals. At each time point, the media were collected, and the cells were washed once with ice-cold MEM, then incubated with 0.25% Pronase®-E in modified Eagle's media for 15 min at 4°C to strip cell surface proteins. The cell suspension was then centrifuged, and the radioactivity associated with cell pellets (defining internalized ¹²⁵I-proteins) was measured at each time point.

- The results of the ligand internalization studies show that despite equal binding, MEF-2 cells cannot internalize rat collagenase-3 (Figure 9). This suggests that the LRP is required for collagenase-3 internalization. It is known that RAP inhibits internalization of ligands by the LRP (Kounnas et al., 1996, J. Biol. Chem 271:6523). Therefore, internalization assays were performed using ^,25I-labeled rat collagenase-3 as a ligand and receptor-associated protein as a competitor. Human receptor-associated protein from the pGEX- RAP expression vector (provided by Dr. Joachim Herz, University of Texas Southwestern Medical Center, Dallas, TX) was expressed in bacteria and prepared as described in Herz et al., 1991, J. Biol. Chem. 266:21232. Those assays show that internalization of ^I25I-rat collagenase-3 is inhibited by RAP by approximately 70% in UMR 106-01 cells (Figure 10). The ability of RAP to inhibit internalization of ¹²⁵I-rat collagenase-3 in UMR 106-01 osteoblastic cells and normal rat osteoblasts was determined next. The presence of 100 mM RAP in binding medium reduced the intracellular accumulation of ¹²⁵I-collagenase by 79% in UMR 106-01 cells and by 43% in normal mineralizing rat osteoblasts (Table 3). Inhibition of collagenase-3 internalization by RAP in both transformed osteoblastic cells and normal osteoblasts suggests that the same type of receptor operates in both cell types.

Table 3. Inhibition of ¹²⁵I-rat collagenase-3 internalization by RAP in osteosarcoma cells and normal osteoblasts. The values displayed represent means ±SEM for triplicate wells.

To investigate the mechanism by which RAP regulates internalization of collagenase- 3, an experiment was performed where excess unlabeled RAP or rat coιlagenase-3_.was prebound to UMR 106-01 cells. Binding and internalization of ^,25I-labeled rat collagenase-3 and RAP were then allowed to proceed. In that experiment, while prebound RAP inhibited rat collagenase-3 internalization, prebound rat collagenase-3 had almost no effect on RAP internalization (Figure 10 A, B).

This example describes collagenase-3 interaction with the cell and shows that it involves two receptors: the specific collagenase-3 receptor acts as the primary binding site, while the LRP is required for internalization.

Binding assays show that the collagenase-3 receptor is present mostly in osteoblasts and fibroblasts. Interestingly, cell-surface binding of collagenase-3 does not necessarily correlate with expression of collagenase-3 by these cells. For example, ROS 17/2.8 cells do not express collagenase-3, but the binding of the enzyme to ROS 17/2.8 cells was comparable to that of UMR 106-01 cells. At the same time, the binding to BC-1 cells, which secrete collagenase-3 at a high constitutive level, was very low. Based on these data, this receptor may bind enzyme secreted by neighboring cells, or play other roles in addition to regulation of the extracellular abundance of collagenase-3.

UMR 106-01 cells were assayed for their ability to bind different metalloproteinases. Although the members of the metalloproteinase family share a number of general functional and structural features, the collagenase receptor is highly specific for rat collagenase-3 and human collagenase-3, with almost no binding of other matrix metalloproteinases. The mouse collagenase-3 also binds equally as well as the rat enzyme. Nevertheless, the possibility cannot be ruled out that the receptor may have ligands other than collagenase-3. Ligand and Western blot analyses show that rat collagenase-3 can specifically, bind to the LRP and a protein with a molecular weight of approximately 170 kDa which is present in membranes of UMR 106-01, MEF-1 and MEF-2 cells. Equal levels of rat collagenase-3 binding to UMR 106-01, wild-type (MEF-1) and LRP-null (MEF-2) cells suggest that the collagenase-3 receptor is present in all of these cell lines and that the LRP does not participate in primary binding of collagenase-3 to the cell surface. Although MEF-1 and MEF-2 cells bind rat collagenase-3 equivalently, our experiments show that MEF-2 cells cannot internalize the bound ligand. Also, rat collagenase-3 internalization by UMR 106-01 cells is abolished in the presence of RAP. Therefore, it appears that collagenase-3 interaction with the cell is a two step process. First, a specific collagenase receptor of 170 kDa acts as a primary binding site for collagenase-3 on the cell surface. Interaction between the LRP and the enzyme- receptor complex then occurs, resulting in internalization of collagenase-3. A similar process has been reported for uPA/PAI-1, tPA/PAI-1 and uPA/rPN-1 complexes (Andreasen et al., 1994, FEBSLett. 338:239; Conese et al., J. Biol. Chem. 269:25668). In those cases, the serine protease binds to a specific receptor as a primary event. The inhibitor then binds to the receptor/ligand complex which leads to its rapid internalization and degradation by the LRP. This latter process is inhibited by RAP, which implicates the LRP.

The ligand blot studies showed that mouse embryo fibroblasts have an additional protein with a molecular weight of approximately 200 kDa which also specifically binds ¹²⁵I- rat collagenase-3. In these cells, then, three membrane proteins might be involved in collagenase-3 clearance, indicating that our proposed mechanism might vary somewhat in different cell types.

The inhibition studies showed that RAP abolishes rat collagenase-3 internalization in UMR 106-01 cells, while collagenase-3 does not change the level of RAP internalization. Thus, collagenase-3 does not compete for binding to RAP sites on the LRP. In addition, RAP may be a physiological modulator of collagenase-3 internalization by the LRP. It has been shown that RAP is coexpressed with either LRP or gp330 (Zhend et al., 1994, J. Histochem. Cytochem. 42:531). However, it is still unknown whether RAP is expressed in osteoblastic cells. Further experiments may show the presence of RAP in bone tissue.

Example 2

This example describes the identification of receptor binding domains on collagenase- 3.

The following reagents were used in this example. Tissue culture media and reagents from Fisher Scientific Co., Itasca, EL. Econopac desalting columns from BioRad, Hercules, CA. HisTrap nickel columns for purification of recombinant proteins were purchased from Pharmacia. Isopropyl β-D-thiogalactoside from Boehringer-Mannheim (=Hoffman-LaRoche, Basel, Switzerland). Other reagents were as specified in Example 1, or were purchased from Sigma Chemical Co. or from Fisher Scientific Co.

Recombinant mouse collagenase-3 (mMMP-13, residues 1-472) and various chimeric collagenases were produced as follows. Recombinant mouse collagenase-3 was subcloned into the pET30 expression plasmid (containing an N-terminal 6xHis purification tag) using restriction endonucleases (Ncol and BamHT) which flank the cDNA sequence. Four of the chimeric constructs (amino acid residues in parentheses) used herein, H/M(228-472), HM/M (141-472), M(l-228)/H, and HM(141-228) H, were described previously (Krane et al., 1996, J. Biol. Chem. 271:28509). In each case, human MMP-1 sequences are represented as "H", and mouse MMP-13 sequences are represented as "M." The MH(213-267)M construct was generated in Dr. Rrane's laboratory by replacing the exon 5 of MMP-13 with exon 5 of MMP-1. Each of these constructs was subcloned into the pET30 expression vector as above. A sixth construct, HM(166-228)/H was generated from the HM(141-228)/H construct using an EcoR F restriction site that is conserved between MMP- 1 and MMP- 13.

Plasmids bearing the constructs of interest were transformed into BL21 E. coli cells. These cells were grown (to OD₆oo= 0.6) in 500 ml of LB broth containing kanamycin (30 μg/ml) and expression of recombinant protein was induced with the addition of EPTG (0.4 mM). After 4 h, cells were centrifuged, and the pellet was washed in 50 mM Tris buffer (containing 5 mM CaCl₂ and 200 mM NaCl, pH 7.6) and stored overnight at -20°C. The bacterial pellet was then resuspended (1 ml per 25 ml culture broth) in PBS containing 6 M -guanidine HCI (pH 7.6, lysis buffer). The lysate was passed through an 18 g needle 8-10x and was then centrifuged (18,000 rpm at 4°C for 30 min). All subsequent steps were performed at 4°C. Purification of protein was performed using a HisTrap nickel column according to the manufacturer's protocol, sequentially washing the column with lysis buffer containing 10 mM and 40 mM imidazole prior to elution in lysis buffer containing 500 mM imidazole. Refolding and dialysis of purified collagenase-3 and collagenase chimeras was performed as described in Zhang and Gray, J. Biol. Chem. 271:8015, 1996 and the protein was immediately frozen at -70°C. The C-terminally truncated collagenase-3, M(l-265), (Knauper et al., J. Biol. Chem. 272:1608, 1997) was generated by exploiting the natural autocatalytic activity of the enzyme, through overnight dialysis followed by denaturation, repurifϊcation on nickel columns to separate the N-terminal protein, and dialysis. The purified proteins were enzymatically active (as determined by gelatin zymography). Collagenase-3 has a molecular mass of 62 kDa. The H/M (229-472) and HM/M(141-472) constructs also have molecular masses of 62 kDa; the M(l-228)/H, HM(141-228)/H, and HM(166-228)/H constructs are slightly truncated (58 kDa) due to a secondary BamHI site in the C-terminal region of the human MMP-1 sequence (these truncations are found in an area of the molecule which is not important to this work).

The recombinant mouse MMP-13 has essentially equivalent catalytic, kinetic, and binding activity as the rat MMP- 13 homolog utilized in Example 1. For example, this protein displays comparable receptor binding activity compared to the purified rat uterine collagenase. As shown in Figure 13A, ^,25I-labeled mMMP-13 binds to UMR 106-01 rat osteosarcoma cells specifically and saturably. Scatchard analysis of these data (Figure 13B) indicates the presence of two receptor populations. The high affinity site (ostensibly the collagenase-3 receptor) was determined to have a K_d of 3.9 nM and a B_max of73.9 fmol / 10⁶ cells (computer analysis with the GraphPad InPlot program yields a K_d of 3.9 nM and a B_max of 75 fmol / 10⁶ cells). The lower affinity site (ostensibly the LRP) was determined to have a Kd of 46.2 nM and B^ of 660 fmol / 10⁶ cells (computer analysis yields a K of 52.8 nM and B_ma_x-of 834 fmol / 10⁶ cells). Chimeric Collagenase Constructs. Since this receptor system is entirely specific for collagenase-3 in UMR 106-01 cells, chimeric collagenase constructs were next utilized to investigate the interaction of collagenase-3 with the collagenase-3 dual receptor system (Figure 14). Each of these recombinant proteins has been expressed and purified (Figure 15 A). Each protein has the expected mobility on SDS-PAGE. Each also has functional activity, as demonstrated by gelatin zymography (Figure 15B). In gelatin zymography experiments, samples of chimeric proteins were subjected to non-reducing SDS-PAGE on a 12% acrylamide gel containing 0.09% gelatin. The gel was run at 100 mV for 4 h at 4°C, rinsed for 30 min in 0.01% Triton X-100, and incubated overnight at room temperature in 50 mM Tris buffer, pH 7.4, containing 10 mM CaCl₂, 100 mM NaCl, and 10 mM ZnCl₂. The gel was then stained for 2 h in Coomassie brilliant blue and fixed and destained for 4 h in 50% methanol / 10% acetic acid. Activity of the enzyme (5 μg) is determined by zones of clearing, indicating gelatinolytic activity. The multiple bands seen on zymography represent active fragments produced through the autocatalytic activity of these enzymes.

These proteins were used as competitors in a series of binding assays in UMR cells using ¹²⁵I mouse collagenase-3. Radioactive labeling and binding assays were performed as in Example 1. As shown in Figure 16, the MH(213-267)M construct and the truncated MMP- 13, M(l-265) construct compete in an essentially equivalent manner compared to full-length MMP-13, indicating that each of these constructs retains all receptor-binding domains possessed by full-length collagenase-3. The M(l-228)/H construct competes with greater than 90% efficiency compared to full-length collagenase-3; the H/M (229-472) construct competes for binding only slightly (<17%), demonstrating no significant difference in binding efficacy compared to an unrelated protein (bovine serum albumin). These data indicate that the essential collagenase-3 bmding domain(s) are present to the amino-terminal side residue 228, within the pro- (residues 1-104) or catalytic (105-235) domains of the full-length enzyme.

The other constructs compete with intermediate efficiency, reflecting the presence or loss of binding domains (perhaps as well as minor structural differences between the constructs resulting in slightly altered three-dimensional presentation of binding domains). The HM(141-228) construct competes with 90% efficiency compared to the full-length enzyme; this level of competition was not significandy different from the binding efficacy observed with a construct that also contains the entire C-terminus of collagenase-3: HM(141- 472). These data suggest that the essential receptor-interacting domains are encompassed in collagenase-3 amino acids 141-228, within the catalytic domain. An additional construct that further-subdivides this region, HM(166-228)/H exhibits an impaired ability to compete for receptor binding (Figure 17). The results are also presented in Table 4.

Table 4. Summary of receptor binding data for collagenase molecules. Binding data are presented as percent of mouse collagenase-3 binding ± S.E.M (at 200-, 400- and 800-fold excess over ¹²⁵I mouse collagenase-3) for seven pooled experiments (UMR cells) or from triplicate wells at each concentration of competitor (MEF cells).

LRP = low density lipoprotein receptor-re ate prote n

Having localized the receptor-binding activity to a portion of the collagenase-3 catalytic domain, the catalytic domain sequence of MMP-13 homologues (human, rat, and mouse) was next compared with other MMPs that were determined in Example 1 to not interact with receptors on UMR cells, including MMP-1, MMP-2 (72-kDa gelatinase), MMP- 3 (stromelysin-1), and MMP-9 (92 kDa gelatinase). This comparison (Figure 18) revealed several discrete regions in which the sequence was entirely conserved among receptor-binders but highly divergent among receptor non-binders. Three domains in particular (residues 131- 140, 209-212, and 250-258) containing conserved stretches of charged amino acids are potential receptor binding domains. Serendipitously, the disparate organization of the chimeric molecules allows nearly independent evaluation of the contribution of each domain to receptor binding.

To determine whether these putative receptor-binding regions are likely to be exposed to solvent, a hydrophilicity plot was generated from the mMMP-13 catalytic domain sequence; other physicochemical parameters (hydrophobicity, polarity, side chain volume) were also analyzed for these three domains using published values for each individual residue. (Chechetkin and Lobzin,J. Theor. Biol. 198:191, 1999). These results indicate that these three charged domains possess high hydrophilicity and low hydrophobicity, indicating that these regions are likely to be exposed to solvent and thus accessible for receptor binding. These assumptions were supported by the recently published crystal structure of collagenase- 3 (Lovejoy et al, Nature Struct. Biol. 6:211, 1999). The binding competition data presented above clearly show that constructs containing MMP-13 residues 209-SSSK-212 (SEQ ID NO:2) retain the capacity to bind components of the collagenase-3 receptor system. This region thus appears to be the high-affinity binding domain recognized by the collagenase-3 receptor, and the other two domains (SEQ ED's NO: 1 and 3) may be recognized by the lower-affinity LRP or may alternatively stabilize the , interaction with the collagenase-3 receptor. To investigate the contribution of each receptor, two mouse embryo fibroblast (MEF) cell lines were utilized. The MEF-1 cell line is known to express both the collagenase-3 receptor and the LRP, while the MEF-2 cell line has been rendered LRP-null. The chimeric molecules containing LRP-binding domains would thus exhibit impaired binding to MEF-2 cells. This hypothesis is confirmed by the data on'Table 4.

The use of the chimeric collagenases in ^I2SI-collagenase-3 binding competitions performed in the MEF-1 cell line yielded results which were comparable to the findings in UMR cells. However, when the experiment is performed on MEF-2 (LPR-null) cells, the HM/M (141-472) construct demonstrates a significantly impaired ability to compete for receptor binding (p < 0.05 compared to competition in MEF-1 cells) (Table 4). This construct lacks the conserved sequence, 136-KAFXK-140 (SEQ ID NO: 1), suggesting that the impaired binding activity in MEF-2 cells is due to a lost interaction with the LRP. This sequence corresponds to a published LRP consensus binding sequence. Interestingly, the M(l-228)/H construct, which lacks the conserved domain from residues 250-258, also demonstrates impaired binding in MEF-2 cells, suggesting that this site contains a novel secondary or low- affinity LRP recognition motif.

The overall model of the interaction of collagenase-3 with this dual receptor system involves a requisite collagenase-3 receptor interaction mediated by the binding domain 209- SSSK-212 (SEQ U) NO:2). Interaction with the LRP, (either independently or upon transfer from the primary receptor) is then mediated through domains 136-KAFRK-140 (SEQ ID NO:l) and 250-GKSHXMXPD-258 (SEQ ID NO:3), with the 136-KAFRK-140 (SEQ ID NO:l) domain being the more critical. The moderate differences in binding of the two constructs containing only the collagenase-3 receptor recognition domain (i.e., HM(141- 228)/H and HM(166-228)/H) may simply derive from structural differences.

This model predicts that chimeric collagenases lacking LRP recognition domains would fail to be internalized. To evaluate this prediction, two of the chimeric constructs, M(l-228)/H and HM(141-228)/H, were iodinated. Both of these constructs retain the putative collagenase-3 receptor recognition motif. As expected, both are capable of binding specifically to UMR cells (Figure 19 A). However, both constructs demonstrate significantly reduced endocytosis (performed as described in Example 1) compared to full-length collagenase-3 (Figure 19B). This indicates that these constructs have an impaired interaction with the LRP, either through the absence of required sequence domains or through an altered three-dimensional structure secondary to chimeric manipulations. In this example, chimeric collagenase proteins were used to identify a discrete collagenase-3 region required for interaction with cell-surface receptors. These constructs were designed to replace critical regions of collagenase-3 with homologous regions of collagenase-1 (which does not bind to surface receptors in UMR 106-01 cells).

The chimeric collagenases were used as competitors to ^,25I collagenase-3 binding in cells known to express both the collagenase-3 receptor and the LRP, and in a cell line lacking the LRP. Using the experimental data and MMP sequence comparison, a collagenase-3 receptor-binding domain (209-SSSK-212) (SEQ ID NO:2) and a LRP-binding domain (136- KAFRK-140) (SEQ ID NO:l) are identified. These regions lie within the catalytic domain of the enzyme, a teleologically appealing site for a receptor-binding (and hence inactivating) activity. Chimeric constructs containing both of these domains retain 85-99% of the cell- surface, binding capacity of full-length collagenase-3. While removal of the LRP-binding domain does not result in significant detriment to binding in cells expressing both receptors, elimination of both receptor binding domains essentially abolishes binding activity.

The LRP is expressed in a wide variety of human tissues and is known to contain multiple independent binding domains. Previously published work has shown that charged residues are required for ligand interactions with the LRP. (Sottrup- Jensen et al., 1986 FEBS Lett. 205: 20). The low density lipoprotein receptor-related protein binding domain identified on collagenase-3 (136-KAFRK-140) (SEQ JJD NO:l) conforms to a published recognition motif of the LRP, consisting of two lysine residues separated by any three amino acids (KXXXK). (Nielson et al., 1996, J. Biol. Chem. 271:12909). This motif is present on ₂-macroglobulin, as well as on the 39-kDa receptor-associated protein. (Ellgaard et al, 1997 FEBS LETT. 244:544).

The close proximity of the receptor recognition domains for the collagenase-3 receptor and the LRP raises the question of whether these two receptors compete for ligand binding. The collagenase-3 receptor appears to have ~10-fold higher affinity for collagenase- 3 compared to the LRP, but it appears to be considerably less abundant at the cell-surface. - However, constructs containing only the putative collagenase-3 receptor recognition sequence still retain up to 90% of their receptor binding capacity. Thus, kinetic and affinity considerations may explain the preferential binding to the collagenase-3 receptor, given that our experiments were performed using concentrations of ^{1 5}I-collagenase-3 (3 nM) slightly below the calculated K of the collagenase-3 receptor (3.9 nM). Alternatively, receptor binding may be dependent upon the folding state of the ligand, and different folding states may present one binding domain to greater advantage.

Also of interest is the process through which ligand bound to the collagenase-3 receptor is transferred to the LRP. As each receptor appears to be independently capable of binding collagenase-3, the collagenase-3 receptor could conceivably serve merely as a molecular docking station for this ligand, resulting in functional inactivation of the enzyme. Collagenase-3 subsequently released from this receptor would then be susceptible for endocytosis upon binding the LRP. However, a more compelling (and physiologically parsimonious) model involves a direct interaction between the collagenase-3 receptor , and the LRP, perhaps analogous to the interaction between the urokinase plasminogen activator receptor and the LRP (Conese et al., 1995, J. Cell Biol. 131:1609). In the latter example, urokinase plasminogen activator binds to the urokinase plasminogen activator receptor and is internalized in an LRP-dependent fashion only upon forming a complex with its specific inhibitor (PAI-1 or protease nexin-1) (Conese et al., 1994, J. Biol. Chem. 269:11886).

Urokinase plasminogen activator receptor is subsequently recycled to the cell surface (Nykjasr et al., 1997, EMBOJ. 16:2610). While the LRP is not essential for cell-surface binding of collagenase-3, it is required for ligand endocytosis (see Example 1). This work suggests a complex interaction, as chimeric constructs capable of binding the collagenase-3 receptor alone, or both the collagenase-3 receptor and the LRP, display impaired endocytosis compared to full-length collagenase. This may suggest the presence of a cryptic collagenase- 3 domain that mediates interaction between the two receptors. Alternatively, binding of collagenase-3 to either receptor may induce conformational changes required for endocytosis; the chimeric proteins may lack sequence or structural determinants necessary to allow ligand internalization.

Example 3 This example illustrates the role of impaired collagenase-3 endocytosis in osteoarthritis and describes treatments to improve the ability of osteoarthritic cells to endocytose collagenase-3.

The following reagents were used in this example. Pravastatin sodium (Bristol-Myers Squibb Company, 10 mg tablets) was dissolved in 50% methanol (1 mM stock solution); cells were treated b.Ld. at a final concentration of 10 μM. Other reagents were as specified in previous examples, or were purchased from Sigma Chemical Co. or from Fisher Scientific Co. Patients (17 osteoarthritic and 9 nonarthritic) were recruited. Classification of patients as osteoarthritic was based upon criteria established by the American College of Rheumatology (Hochberg et al., 1995, Arthritis and Rheumatism 35:1535). All patients in the experimental group showed clinical and radiographic evidence of osteoarthritis and were _ undergoing primary total knee or hip arthroplasty. All patients in the control group had no previous history of arthritis or joint trauma at the site. No evidence of arthritis was observed in the control tissues at collection. Informed consent was obtained from all surgical patients. Patients with systemic infection, autoimmune disease, previous joint surgery or trauma at the site were excluded from the study. Patients receiving treatment with corticosteroids, bisphosphonates, or intraarticular hyaluronan were also excluded.

Human tissue (articular cartilage and synovium) was obtained at surgery or autopsy and was prepared as follows. Tissue was minced in a laminar flow hood, then incubated in sterile filtered serum-free Dulbecco's modified Eagle's medium (50 ml) containing trypsin (0.25%) for 1 h at 37°C on an orbital shaker. Tissues were then centrifuged at 2000 rpm for 10 min at 4°C, rinsed in Dulbecco's modified Eagle's medium, and incubated in sterile-filtered Dulbecco's modified Eagle's medium containing 10% heat-inactivated fetal bovine serum (10% DMEM) and bacterial collagenase (Sigma; 0.7 mg/ml) at 37°C on an orbital shaker for 4 h (synovial fibroblasts) or overnight (cartilage). Undigested tissue was removed, and the cells were centrifuged at 2000 rpm at 4°C for 10 min, then rinsed and resuspended in 10% DMEM. Cells were enumerated with a hemocytometer and plated at 5 x 10⁴ cells/well of 24- well plates (for binding and degradation assays), or at 1-5 x 10⁶ cells per 175 cm² flask (for RNA collection). Media were changed thrice weekly until confluence was reached (typically 3-5 weeks), at which time experimentation or RNA collection was performed.

. Table 5 characterizes the patients recruited from this study. There was no significant difference in mean age, weight, or height between the control and osteoarthritis groups. The distribution between hip and knee was relatively even in both groups. A greater proportion of female subjects was in the osteoarthritis group (12/17 osteoarthritis vs. 3/9 control), but no gender-specific differences in the data were detected.

Table 5. Patient characteristics.

Abbreviations in Table 1: R, right; L, left; THA, total hip arthroplasty; TKA, total knee arthroplasty; amp, amputation; hp, hemipelvectomy (secondary :o metastatic small cell carcinoma); itf, intertrochanteric fracture; aut, autopsy. ^!Ages, heights, and weights in Table 1 are presented as means ± SD.

To confirm phenotypes of cultured human tissues and to analyze the expression of collagenase-3 in these cells, reverse transcriptase-polymerase chain reaction was performed on RNA isolated from confluent cultured cells as follows. Total RNA and poly (A)⁺ mRNA from human cells were isolated using the Tri-Reagent (Sigma) and FastTrack kit (Invitrogen), respectively. Reverse-transcriptase polymerase chain reaction (RT-PCR) was perforrhed to detect marker transcripts (collagen types I, II, and aggrecan) as well as collagenase-3. Reverse transcription was carried out at 42°C for 60 min in the presence of patient RNA (100 ng) and forward- and reverse-orientation primers to each gene of interest. The primer sequences (with modification) used were as follows. Collagen a (T) (amplifying a 261 bp fragment): forward (5 '- GCG GAA TTC CCC CAG CCA CAA AGA GTC- 3 ')(SEQ ID NO:4); reverse (5'-CAG TGC CAT CGT CAT CGC ACA ACA CCT) (SEQ JD NO:5), T_m = 79°C. Collagen _t(Ω), amplifying a 307 bp fragment: forward (5'-GTC CCC GTG QOC TCC CCG-3')(SEQ ID NO:6); reverse (5'-CAG TGC CAT CCA CGA GCA CCA GCA CTT-3')(SEQ ID NO:7), T_m = 62°C. Aggrecan, amplifying a 297 bp fragment: forward (5'- CCA TGC AAT ITG AGA ACT-3')(SEQ ID NO:8); reverse (5'-CAG TGC CAT ACA AGA AGA GGA CAC CGT-3 ')(SEQ ID NO:9), T_ra = 50°C. Collagenase-3, amplifying a 392 bp fragment: forward (5'-CCT CCT GGG CCA AAT TAT GGA G-3')(SEQ ID NO:10); reverse (5 '-CAG CTC CGC ATC AAC CTG CTG-3 ')(SEQ ID NO: 11), T_m = 64°C. β-actin (purchased from Stratagene, Menasha WT), amplifying a 661 bp fragment: forward (5'- TGA CGG GGT CAC CCA CAC TGT GCC CAT CTA -3')(SEQ ID NO:12); reverse (5' - CTA GAA GCA TTT GCG GTG GAC GAT GGA GGG- 3')(SEQ ID NO:13), T_m = 60°C,. Standard cycling conditions were as follows: initial denaturation (94°C, 30 s), annealing at optimal temperature (49-60°C, 1 min), and elongation (72°C for 2 min). Thirty cycles of amplification were performed per assay.

As expected, cartilage-specific transcripts (aggrecan and collagen αι(ιT)) were amplified only from chondrocyte RNA, and the fibroblast marker transcript, collagen α.ι(i), was amplified from synovial fibroblast RNA (Figure 20). In each case, transcripts were seen in both osteoarthritis and control samples. Inappropriate synthesis of type I collagen by osteoarthritis chondrocytes has been noted previously (Zlabinger et al., Rheumatol. Int. 6:63). Collagenase-3 was amplified from osteoarthritis chondrocytes, which supports the published ELISA results of others (Wolfe et al., Arthritis Khum. 36: 1540).

Binding of ^s I collagenase-3 to osteoarthritis and control human cells. To ^* investigate the ability of human chondrocytes and synoviocytes to immobilize collagenase-3 at cell-surface receptors, binding assays were performed with ^{, 5}I collagenase-3, as described in Example 1. Excess unlabeled collagenase-3 was used to determine nonspecific binding. Compared to the binding in nonarthritic tissues, osteoarthritic tissues showed significantly reduced levels of ^I25I collagenase-3 binding, with a 16.4% decrease in osteoarthritis chondrocytes and a 75.5% decrease in osteoarthritis synoviocytes (determined by integrating the area under the curves) (Figure 21). Differences between osteoarthritis and control binding are statistically significant for all data points (1-20 nM) in the chondrocyte assays_. (p < 0.001) and for all data points beyond 2 nM (3-20 nM) in the synoviocyte assays (p < 0.004). Scatchard and computer analysis of these data suggest the decreased binding in osteoarthritis cells is due to decreased receptor number rather than an alteration in receptor affinity (as the B-_max is reduced without significant change in K_d). These findings indicate a decreased abundance of the collagenase-3 receptor in osteoarthritis tissues, which may explain the reported high levels of this enzyme in osteoarthritis synovial fluids.

To determine whether the decreased collagenase-3 binding in osteoarthritis tissues correlated with reduced receptor function, the amount of ^{1 S}I collagenase which was internalized by osteoarthritis and control chondrocytes was determined as described in Example 1. A 66.4% decrease was observed in collagenase-3 internalization in osteoarthritis cells compared to control cells (p < 0.001)(Figure 22). This correlates well with the binding data above.

The internalization assays were also performed in the presence of 39 kDa RAP. Collagenase-3 internalization was inhibited by 88.2% (p < 0.001) in control chondrocytes in the presence of RAP (Figure 22), indicating that the LRP is involved in collagenase-3 internalization in human chondrocytes (analogous to the findings with osteoblasts disclosed in Example 1). Incubation of osteoarthritis chondrocytes with RAP also reduced collagenase-3 internalization, by 74.3% (pO.OOl). Results with synoviocytes were similar.

As receptor-mediated processing of collagenase-3 culminates in lysosomal degradation and extracellular release of the ligand, binding and degradation assays as above were performed and measured the presence of degraded (TCA-soluble) ¹²⁵I-collagenase-3 in the media overlying cells. Compared to nonarthritic tissues, osteoarthritis chondrocytes and synoviocytes demonstrated significantly reduced (p < 0.001 for the 15-90 min data points) excretion of ¹²⁵I collagenase-3, by a proportion of 69% and 71.2%, respectively (determined by integrating the area under each curve) (Figure 23). Excretion of collagenase-3 from these cells was also inhibited by RAP. These findings indicate that processing of collagenase-3 is impaired in osteoarthritis, ostensibly as a direct consequence of the impaired binding activity.

The cholesterol-lowering drugs, HMG-CoA reductase inhibitors (statins) are well- known to increase hepatic expression of LDL receptors. Since collagenase-3 internalization and degradation are dependent upon an LDL receptor superfamily member, it was hypothesized that treatment with statins would also increase collagenase-3 processing. Osteoarthritis chondrocytes were treated without or with pravastatin (10 μM b.i.d. for three days) prior to performing binding and degradation assays as above. Remarkably, excretion of degraded collagenase-3 was indeed enhanced (by over 320%; p < 0.02) in the presence of pravastatin, to levels approaching those seen for non-arthritic tissues (Figure 24). This was seen despite only a modest (30%; p < 0.05) increase in binding. Similar results were obtained using atorvastatin. Results were similar but less pronounced in osteoarthritis synoviocytes. Statin treatment produced no significant changes in collagenase-3 binding or degradation in control cells.

In this example, evidence is presented for collagenase-3 receptor dysfunction in human osteoarthritic tissues. Also presented are data that indicate that collagenase' processing is improved in those tissues upon treatment with HMG CoA reductase inhibiting agents. Since high levels of collagenase-3 have been found in the synovial fluid of patients with osteoarthritis, it was hypothesized that receptor-mediated removal of this enzyme was impaired. These data indicate that specific binding of collagenase-3 is drastically reduced in osteoarthritic tissues. The observed decrease in collagenase-3 binding by osteoarthritis tissues is paralleled by proportionate decreases in internalization and degradation of the enzyme. These data indicate a pathophysiological model for the development and progression of osteoarthritis, whereby a primary or secondary dysfunction of the collagenase-3 receptor system leads to increased levels of this destructive enzyme in synovial fluid and the consequent erosion of articular cartilage.

Components of the collagenase receptor system may be subject to reduced expression (due to multifactorial causes), or to reduced activity (attendant to mechanical joinf degeneration). Alternatively, local factors in the arthritic joint space may lead to reduced or dysregulated receptor expression. It is also possible that genetic variation in the collagenase receptor predisposes to slowly progressive dysfunction.

Although this is the first report correlating dysfunction of an endocytotic receptor with osteoarthritis, others have described reduced expression of integrin (adhesion) receptors in osteoarthritis. Specifically, decreased levels of integrin αl subunits have been found within moderately to heavily damaged osteoarthritis cartilage compared to minimally damaged osteoarthritis cartilage (Lapadula et al., 1998, Clin. Exper. Rheumatol. 15:241. Chondrocytes normally express the α₅βl integrin (fibronectin receptor) (Durr et al., 4993, Exper. Cell Res. 207:235), and engagement of α₅ βi increases collagenase expression (Arner et al., 1995, Arthritis and Rhematism 38:1304; Huhtala et al., 1995, J. Cell. Biol. 129:861). Accordingly, reduced integrin expression may stem from a feedback attempt to limit collagenase synthesis. In contrast, reduced activity of collagenase-3 receptors is likely to represent a primary or exacerbating derangement in osteoarthritis. The collagenase-3 receptor is a distinct cell-surface receptor which is unlikely to belong to the integrin family, as treatment of nonarthritic chondrocytes (or UMR- 106-01 rat osteosarcoma cells) with an integrin blocking agent does not result in a significant decrement in collagenase-3 binding. Moreover, it is unlikely that disease progression results in a generalized loss of cell-surface components, as the expression of other integrins and adhesion molecules is unaltered or increased in osteoarthritis (Loeser et al., 1995, Exper. Cell Res. 217:248).

It is becoming apparent that HMG-CoA reductase inhibitors (statins) have pleiotropic effects extending beyond the lowering of serum cholesterol. These agents are well known to increase cell-surface expression of LDL receptors in hepatocytes. Effects of these agents on the expression of LDL-related receptors at other sites are less well-characterized. Recent reports indicate that statins may have clinical utility in limiting bone loss in animal models of osteoporosis, with efficacy comparable to bisphosphonates (Mundy et al., 1998, Bone 23:S183). In those studies, statins increased transcription of bone morphogenetic protein-2, which in turn is known to suppress levels of collagenase-3. Others have shown that statins prevent experimental osteonecrosis induced by steroids (Cui et al., 1997, Clin. Orthop. Rel Res. 344:8) and that statins and bisphosphonates inhibit osteoclast activity (Fisher et al., 1999, PNAS 96 133). Thus, in joint tissue, statins may have multiple activities, which culminate in a restored balance between the synthesis and degradation of matrix proteins.

Example 4

This example describes the binding of collagenase-3 to an immortalized chondrocyte cell line.

The immortalized human chondrocyte cell line T/AC-62 was generously provided by Dr. Mary Goldring. Rat collagenase-3 was purified and labeled with ¹²⁵I as described in previous examples. A binding assay was performed by adding varying concentrations of ¹²⁵I collagenase-3 to confluent T/AC-62 cells at 4°C. A 200-fold excess of unlabeled mouse collagenase-3 was added to replicate wells to account for nonspecific binding. Specific binding is derived as the difference between total and nonspecific binding. Results are shown in figure 25 as mean values ± S.E.M. for triplicate wells of a representative experiment.

As shown in figure 25, ¹²⁵I-collagenase-3 binds to the T/AC immortalized human chondrocytes specifically and saturably, indicating the presence of the collagenase-3 receptor on these cells. This is the first demonstration of collagenase binding to an immortalized chondrocyte cell line. The T/AC cell line is thus useful for studying collagenase-3 binding to its receptor on chondrocytes.

Example 5

This example describes the induction of LRP transcription by statins in osteoarthritic chondrocytes.

Osteoarthritic chondrocytes were isolated as in example 3. Cells were then either treated with 10 μM pravastatin twice daily for three days, or untreated. Levels of mRNA specific for LRP was then determined by northern hybridization as described in example 1, where 18S rRNA was used as a load normalization. The results are shown in figure 26. Pravastatin induced expression of LRP mRNA in these cells. Phosphorimager analysis established that the treated cells expressed about 20-fold the levels of LRP mRNA as the untreated cells. This establishes that statins induce transcription of the LRP gene.

All references cited in this specification are hereby incorporated by reference. The discussion of the references herein is intended merely to summarize the assertions made by the authors and no admission is made that any reference constitutes prior art. Applicants reserve the right to challenge the accuracy and pertinence of the cited references.

In view of the above, it will be seen that the several advantages of the invention are achieved and other advantages attained.

As various changes could be made in the above methods and compositions without departing from the scope of the invention, it is intended that all matter contained in the above description and shown in the accompanying drawings shall be inteφreted as illustrative and not in a limiting sense.

SEQ IDs

SEO ID NO: SEQUENCE DESCRIPTION 1 KAFXK amino acids - low density lipoprotein receptor-related protein binding site* on mouse collagenase-3

SSSK amino acids -C3R binding site on mouse collagenase-3

GKSHXMXPD amino acids - secondary low density lipoprotein receptor-related protein binding site on mouse collagenase-3

GCGGAATTCCCCCAGCCACAAAGAGTC

Collagen αι(T) forward primer

CAGTGC CATCGTCATCGCACAACACCT

Collagen αι(I) reverse primer

GTCCCCGTGGCCTCCCCG

Collagen αι(II) forward primer

CAGTGCCATCCACGAGCACCAGCACTT

Collagen ai(n) reverse primer

CCA TGC AAT ITG AGA ACT

Aggrecan forward primer

CAGTGCCATACAAGAAGAGGACACCGT

Aggrecan reverse primer

10 CCT CCT GGG CCA AAT TAT GGA

Collagenase-3 forward primer CAGCTCCGCATCAACCTGCTG

Collagenase-3 reverse primer

TGACGGGGTCACCCACACTGTGCCCATCTA β-actin forward primer

CTAGAAGCATTTGCGGTGGACGATGGAGGG β-actin reverse primer

Claims

What is claimed is:

1. A method for increasing expression of an LDL receptor-related protein (LRP) in a cell, comprising contacting the cell with an effective amount of an HMG-CoA reductase inhibitor.

2. The method of claim 1, wherein the increased expression of the LRP causes increased LRP binding of an LRP ligand.

3. The method of claim 2, wherein the ligand is selected from the group consisting of an endoproteinase, coagulation factor Xa, leptin, prostate-specific antigen, β-amylόid peptide, epidermal growth factor, insulin, transforming growth factor-β, platelet-derived growth factor, interleukin-lβ, interleukin-6, basic fibroblast growth factor, nerve growth factor, ApoE, ApoE-containing lipoproteins, lipoprotein lipase, lactoferrin, tissue plasminogen activator, urokinase plasminogen activator, complement component C3, exotoxin A from Pseudomonas aeruginosa, saposin precursor, thrombospondin, Cl inhibitor-Cls complex, tissue plasminogen activator-plasminogen activator inhibitor-1 complex, urokinase plasminogen activator-plasminogen activator inhibitor-1 complex, neutrophil elastase-αl-antitrypsin- elastase complex, and a matrix metalloproteinase.

4. The method of claim 3, wherein the ligand is selected from the group consisting of coagulation factor Xa, β-amyloid peptide, ApoE, complement component C3, and the matrix metalloproteinase collagenase-3.

5. The method of claim 1, wherein the cell is in a vertebrate with a condition associated with excessive levels of an LRP ligand.

6. The method of claim 5, wherein the LRP ligand circulates in the vertebrate.

7. The method of claim 5, wherein the LRP ligand is made by the cell.

8. The method of claim 5, wherein the LRP ligand is selected from the group consisting of an endoproteinase, coagulation factor Xa, leptin, prostate-specific antigen, β- amyloid peptide, epidermal growth factor, insulin, transforming growth factor-β, platelet- derived growth factor, interleukin-lβ, interleukin-6, basic fibroblast growth factor, nerve growth factor, ApoE, ApoE-containing lipoproteins, lipoprotein lipase, lactoferrin, tissue plasminogen activator, urokinase plasminogen activator, complement component C3, exotoxin A from Pseudomonas aeruginosa, saposin precursor, thrombospondin, Cl inhibitor- Cls complex, tissue plasminogen activator-plasminogen activator inhibitor-1 complex, urokinase plasminogen activator-plasminogen activator inhibitor-1 complex, neutrophil elastase-αl-antitrypsin-elastase complex, and a matrix metalloproteinase.

9. The method of claim 5, wherein the condition is selected from the group consisting of Alzheimer's disease, excessive blood clotting, excessive complement activation, and osteoarthritis.

10. The method of claim 9, wherein the vertebrate is a human and the LRP ligand is selected from the group consisting of coagulation factor Xa, β-amyloid peptide, ApoE, complement component C3, and collagenase-3.

11. A method of reducing activity of an LRP ligand in a vertebrate, comprising treating the animal with an HMG-CoA reductase inhibitor.

12. The method of claim 11, wherein the vertebrate is treated with the HMG-CoA reductase inhibitor by treating cells with the HMG-CoA reductase inhibitor then implanting the treated cells into the vertebrate.

13. The method of claim 11, wherein the vertebrate is treated with the HMG-CoA reductase inhibitor by administering to the vertebrate the HMG-CoA reductase inhibitor in a pharmaceutically acceptable composition.

14. The method of claim 11, wherein the LRP ligand is"selected from the group consisting of an endoproteinase, coagulation factor Xa, leptin, prostate-specific antigen, β- amyloid peptide, epidermal growth factor, insulin, transforming growth factor-β, platelet- derived growth factor, interleukin-lβ, interleukin-6, basic fibroblast growth factor, nerve growth factor, ApoE, ApoE-containing lipoproteins, lipoprotein lipase, lactoferrin, tissue plasminogen activator, urokinase plasminogen activator, complement component C3, exotoxin A from Pseudomonas aeruginosa, saposin precursor, thrombospondin, Cl inhibitor- Cls complex, tissue plasminogen activator-plasminogen activator inhibitor-1 complex, urokinase plasminogen activator-plasminogen activator inhibitor-l complex, neutrophil elastase-α 1 -antitrypsin-elastase complex, and a matrix metalloproteinase.

15. The method of claim 11, wherein the vertebrate is a human with a condition associated with excessive levels of an LRP ligand.

16. The method of claim 15, wherein the condition is selected from the group consisting of Alzheimer's disease, excessive blood clotting, excessive complement activation, and osteoarthritis.

17. The method of claim 16, wherein the LRP ligand is selected from the group consisting of coagulation factor Xa, β-amyloid peptide, ApoE, complement component C3, and collagenase-3.

18. A method for determining whether a particular condition in a first vertebrate is caused by insufficient expression of an LRP, comprising treating the first vertebrate with an HMG-CoA reductase inhibitor, then assessing the status of the condition, wherein relief of the condition indicates that the condition is caused by insufficient expression of the LRP.

19. The method of claim 18, wherein the method further comprises treatment of a second vertebrate with the HMG-CoA reductase inhibitor and receptor-associated protein, then assessing the status of the condition, wherein relief of the condition in the first vertebrate but not the second vertebrate further indicates that the condition is caused by insufficient expression of the LRP.

20. The method of claim 18, wherein the first vertebrate is a human.