MX2008003558A - Tissue inhibitor of metalloproteinases (timp) linked to glycosylphosphatidylinositol (gpi) -anchors for treatment of cancer and skin lesions - Google Patents

Abstract

The present invention relates to fusion constructs of glycosylphosphatidylinositol (GPI)- anchored tissue inhibitors of metalloproteinases (TIMPs) and their use for the treatment of cancer and in regenerative medicine. By this approach, the GPI-anchored TIMP proteins are incorporated into the surface membrane of tumor cells and render tumor cells sensitive to FAS-induced apoptosis. Furthermore, the fusion constructs of the present invention are effective agents useful in wound healing applications. In one embodiment, the TIMP is linked to mucin followed by GPI in order to enhance surface presentation. The use of GPI to link TIMP renders the resulting fusion protein particularly useful as an anti-cancer agent for the treatment of cancer, and, in particular, any residual cancer following an incomplete surgical resection of primary tumors in an individual.

Description

TISSUE INHIBITOR OF METALOPROTEINASAS LINKED TO ANCHORAGES OF GLICOSILFOSFATIDILINOSITOL FOR TREATMENT OF CANCER AND SKIN INJURIES Field of the Invention The present invention relates to the field of fusion constructions and their use for the treatment of cancer and in regenerative medicine. Specifically, the invention relates to constructs which comprise metalloproteinase tissue inhibitors (TIMPs) anchored to glycosylphosphatidylinositol (GPI). Additionally, the fusion constructs of the present invention are effective regenerative agents useful in the field of wound healing applications. Background of the TIMP Invention in Cancer Research The treatment of cancer remains a demanding task and employs different procedures and therapeutic strategies, which offer varying degrees of success. A known method is to increase the sensitivity of cancer cells to immune-mediated lysis. The sensitivity of tumors to immune-mediated lysis has been linked to the biology of matrix metalloproteinases (MMP by its abbreviations in English), and specifically, to the expression of the cellular surface of MMP by the REF cell. : 190930 tumor objective. Matrix metalloproteinases (MMPs) degrade components of the extracellular matrix (ECM) and have been implicated in tissue remodeling, tumor invasion, and metastasis (Egeblad &Werb, 2002; Itoh &Nagase, 2002) . MMP activity has also been associated with the efficiency of both perforin / granzyme and FAS mediated apoptosis (summarized in Egeblad &Werb, 2002. It has been shown that MMP activity is regulated at many levels including four endogenous inhibitors, the tissue inhibitor of metalloproteinases (TIMP-1, -2, -3 and 4 (Bode &Maskos, 2003) The in vivo equilibrium between MMP and TIMP determines whether resorption or matrix deposition occurs (Nagasa &Woessner, 1999) Metaloproteinase endogenous tissue inhibitors (TIMPs) exhibit diverse physiological / biological functions that include moderation, metastasis, and apoptosis of tumor growth.These diverse biological activities of TIMP have been linked in part to the stoichiometry of interactions. of cellular surface protein / TIMP / MMP The recruitment of cytotoxic lymphocytes represents a potential pathway in the defense against tumors. all T-cell cytotoxic (CTL) and natural killer (NK) cells that infiltrate and recognize tumor cells, a Effective antitumor immunity often fails to develop efficiently. This inefficiency is a reason that prevents the complete elimination of residual tumor cells after complete surgical removal, either due to an advanced stage of disease or local inoperability. The etiology of this functional deficiency in cytotoxic lymphocytes is currently unclear. In general, the anti-tumor effects of CTL and NK cells are mediated through either the perforin / granzyme-mediated apoptosis trajectories or FAS (CD95 / CSD95L) (Kagi et al., 1994). The trajectory of perforin is mediated by secreted cytotoxins during CTL or NK recognition of target cells (Kagi et al., 1994). The CD95 or FAS death receptor belongs to the regulator of the cell death family of proteins and is of central importance in apoptosis mediated by tumor cell immunology (Nagata, 1999). FAS / CD95 / Human Apo-l is a simple transmembrane glycoprotein receptor (325 amino acids, 45-48 kDa). The FAS ligand (FAS ligand, FASL, CD95L) is an integral membrane protein and is a transmembrane glycoprotein of type II. FASL is a member of the dek TNF family, which includes TNFa, a- and ß-lymphotoxin chains (LT), the CD40 ligand and the CD30 ligand. The action of FAS is mediated by FADD (domain of associated death to FAS) / MORTI, an adapter protein that has a death domain at its C terminus and binds to the cytoplasmic death domain of FAS. Many tumors have been found to be resistant to apoptosis mediated through the trajectory of FAS (Frost et al., 2003, summarized in Igney &Kra er, 2002). As a model system for testing the TIMP-BPI constructs of the present invention, renal cell carcinoma cell lines have been used as an example. Renal cell carcinoma (RCC) is the seventh leading cause of cancer. Approximately one third of patients with RCC have metastatic disease on presentation and up to 50% of relapse after nephrectomy (Vogelzang &Stadler, 1998). RCC is difficult to treat and immune therapies such as interferon-alpha and interleukin-2 are generally more effective than chemotherapy or radiation (Vogelzang &Stadler, 1998). Cytotoxic lymphocytes represent a potential component in the defense against tumors which include RCC. A member of the TIMP family, TIMP-1, is an MMP inhibitor which acts extensively (Bode &Maskos, 2003). It is a soluble protein that can be detected on the cell surface only through its association with proteins bound superficially (Bre et al., 2000; Klier et al., 2001). The total role of TIMP-1 in cancer biology remains the subject of conflict reports (Brand, 2002). To date, it is accepted that TIMP-1 plays a role in angiogenesis, cell migration, and proliferation (Brand, 2002). Recently, it is shown that a TIMP-1 protein anchored to GPI exhibits a pronounced suppression of endothelial cell migration in response to bFGF (Djafarzadeh R et al., 2004). Conventional strategies and procedures for cancer therapy still suffer from the problem that tumor cells are difficult to eliminate once the tumor has developed. Primary tumors are usually removed from the patient by surgery. However, in some cases not all regions are available for surgery, and, in this way, the tumor cells remain in the body where they can develop into secondary tumors. This is a result of incomplete surgical excision of the primary tumor. The present invention therefore provides an effective anti-cancer agent and strategies for the purpose of reducing or alleviating the proliferation of tumor cells in an individual, particularly in a patient who is subjected to incomplete surgical removal of a primary tumor. The anti-cancer agents of the present invention are useful for kill tumor cells both in cell lines in vitro and in tissues in vivo. Role of TIMP in Regenerative Medicine The present invention is additionally useful in the field of regenerative medicine. A significant area in the field of regenerative medicine is related to the process of wound healing. Wound healing is related to a natural restorative response to damaged tissue and involves a complex cascade of cellular events that ultimately generates regeneration, reconstitution, and reestablishment of the tensile strength of damaged tissue. This process generally couples the recruitment and proliferation of different cell types, an elaboration of the cell matrix, and an increase in immune surveillance. Wound healing proceeds in a synchronized, sequential fashion and can be divided into four general phases: inflammation, granulation, re-epithelization and tissue remodeling. Each phase of the wound healing process is regulated by special signal transduction paths. During the healing of wounds, an increase in the expression of growth factors and cytosines occurs; in particular, the increase in TNF, IL-1 and IL-6 levels has been described. During the initial inflammation phase, which involves the effector proteins IL-1, TNF-a and CSF, both macrophages and neutrophils are recruited to form a fibrin clot. During the granulation phase, the fibroblasts proliferate, migrate to the wound and secrete ECM. The effector proteins involved in this latter phase include MMP, PDGF, FGF, EGF and VEGF. The third phase in wound healing, re-epithelialization, is characterized by the proliferation of keratinocytes, which migrate towards the wound, and also by an increase in myofibroblasts, which are responsible for the contraction of the wound. The result of this phase, which involves the effector proteins MMP, KGF, TGF, GM-CSF, EGF and uPA and tPA, is the re-epithelization of the wound surface, the dissection of the black coast and the formation of a barrier. Finally, during the tissue remodeling phase, fibroblasts produce a collagenous matrix which leads to the formation of scar tissue, apoptosis of fibroblasts, and a change from activation to differentiation of keratinocytes. The known effector proteins involved in this last phase of wound healing include TGF-bl, MMP and TIMP. In this way, the effector cells responsible for many aspects of wound healing are the fibroblasts and keratinocytes, and the MMPs that play an important role in both the migration of fibroblasts (MMP-1, -2, -3 and -13) and keratinocytes (MMP-1, -2, -3 and -10) (Singer & Clark, 1999) in addition to scar formation. Each of the MMPs has a different substrate specificity within the NDE and plays an important role in the renewal and degradation of NDEs. The MMP family includes, inter alia, collagenases (MMP-1, MMP-8, MMP-13, MMP-18), stromelysins (MMP-3, MMP-10, MMP-11), gelatinases (MMP-2, MMP). -9), matrilysin (MMP-7), metalloelastase (MMP-12) and a series of membrane-bound matrix metalloproteinases (MT-MMP). As the function of MMP is to proteolytically break the surrounding ECM, a balance between this protease activity and ECM deposition during wound healing, ie reconstruction of damaged tissue, needs to be optimally maintained. The control of MMP activity is modulated by TIMP proteins, which are produced by most cells, and act to inhibit MMPs in a 1: 1 ratio. Where this delicate balance between proteolytic cleavage and ECM deposition is disturbed, disorders such as abnormal wound healing may result, for example, chronic wounds, excessive scarring or keloid scarring. Therefore, there is a need to control or influence the physiological balance between protease activity and ECM deposition during the wound healing process. In an additional modality, the constructions of The fusion of the present invention provides an effective regenerative agent for the treatment of conditions defined by a disturbed balance between MMP protease activity and NDE deposition as, for example, in keloid healing or chronic wounds, which are commonly associated with increased of MMP. Additionally, the fusion constructions of the present invention provide a regenerative agent that can reduce, minimize or inhibit scar formation during the wound healing process. Summary of the Invention The present invention provides novel anti-tumor agents and methods for the treatment of cancer. The present invention is based on the surprising discovery that TIMPI anchored to GPI effectively reduces or alleviates the proliferation of cancer cells, and promotes the death of cancer cells in cell lines both in vitro and in vivo. The structural and functional determinants of TIMP have been combined with an anchor of glycosylphosphatidylinositol (GPI) and, optionally, with ucine, in order to generate a highly effective chemotherapy agent. This procedure exploits TIMP proteins anchored by glycosylphosphatidylinositol (GPI) to be incorporated into the surface membranes when they are purified and added to cancer cells. The TIMP-GPI fusion with a mucin domain also increases the presentation of TIMP proteins on the surface cell membrane and makes the fusion construct more effective in leading cancer cells to be sensitive to immune mediated destruction. In the following examples, the present invention demonstrates that TIMP has the potential for inhibition of tumor cell growth and reduction of tumor development in both in vitro cell lines and in vivo tissues. The binding of TIMP to an anchor of GPI and exogenous administration of TIMP anchored to GPI results in an efficient insertion of TIMP protein into the cell membranes of cancer cells. The surface expression of TIMP-1 anchored to GPI induces a variety of biological effects in cancer cell lines with potential therapeutic relevance such as inducing the apoptotic pathway mediated by FAS in cancer cells. As shown in the following examples, the suppression of cancer cell proliferation is observed to be dose dependent. The TIMP-1 protein anchored to GPI also blocks the secretion of proMMP-2 and proMMP-9 and dramatically alters the cellular surface association of various MMPs. More significantly, tumor cell lines normally resistant to FAS apoptosis are brought to be sensitive to death mediated by FAS / CD95. Treatment with GPI-TIMP results in down-regulation of the anti-apoptotic BCL2 protein and a corresponding increase in pro-apoptotic BAX protein. This shift towards a higher concentration of pro-apoptotic proteins may be a reason for the increased sensitivity of FAS-mediated apoptosis of surface-modified cancer cells by TIMP. Using the above procedure, the TIMP proteins anchored to GPI or polypeptides of the present invention have been tested to be particularly useful in therapeutic applications in the treatment of residual cancer after incomplete surgical removal of the primary tumor as in advanced breast cancer, osteosarcoma , renal cell carcinoma or in malignant brain tumors, for example glioblastoma. On the other hand, since tumor cells, including renal cell carcinoma (RCC), are intrinsically resistant to FAS-mediated death, the present invention provides an effective means to bring tumor cells to be susceptible to FAS-mediated apoptosis. In a first aspect, the present invention therefore relates to a fusion construct (TIMP-GPI or TIMP-mucin-GPI) which comprises a sequence of amino acids of a metalloproteinase tissue inhibitor (TIMP) or a biologically active fragment thereof, wherein the TIMP or biologically active fragment thereof is linked to an amino acid sequence of a mucin domain followed by an amino acid sequence of a glycosylphosphatidylinositol anchor (GPI) In a preferred embodiment, the 3 'end of TIMP is fused directly to a GPI linkage sequence and does not contain a mucin domain. The term "mucin" is related to a family of large, highly glycosylated proteins. One class of mucins are bound to the membrane due to the presence of a hydrophobic membrane expansion domain that favors retention in the plasma membrane, while another class of mucins are secreted at the mucosal surfaces. The mucin genes encode mucin monomers that are typically synthesized as rod conformation apomucin nuclei that are post-transducionally modified by exceptionally abundant glycosylation. Two quite different regions are found in mature mucins. One region includes the amino and carboxy terminal regions, which are slightly glycosylated, but rich in cysteine residues, which are probably involved in establishing disulfide bonds within and between the mucin monomers. The second central region is formed of multiple random repeats of 10 to 80 residue sequences, where half of the amino acid residues are serine or threonine. Mucins are generally secreted as massive aggregates of proteins which have molecular masses of approximately 1-10 million Daltons. Within these aggregates, monomers are bound together mostly by non-covalent interactions, although intermolecular disulfide bonds may also play a role in this process. At least 19 human mucin genes have been distinguished including MUC1, 2, 3A, 3B, 4, 5AC, 5B, 6-9, 11-13 and 15-19. Mucin as used in the present invention is preferably a membrane-bound mucin domain and preferably comprises an amino acid sequence selected from the group which consists of MUC1, MUC3A, MUC3B, MUC4, MUC11, MUC12, MUC16, and MUC17, or a variant or portion of it (previous mucins are reviewed in Moniaux N et al., 2004). In another preferred embodiment, a mucin stalk is used that is isolated from the chemokine CXCL16 or fractalkine (CX3CL1) associated with the surface. Fractalkine is a member of the large and complex chemokine gene superfamily, which consists mainly of secreted, proinflammatory molecules. The Typical chemokine core structure is partially maintained by disulfide bonds between positionally conserved cysteine residues. For most chemokine peptides, a familiar structural feature is the distribution of four cysteines within the molecule, that is, a cysteine signature pattern: CXC, CC, and C, where C is a cysteine and X is any residue of amino acids. Four different chemokine families have been identified based on the observation that the chemokine peptides can be distinguished by the organization of the cysteine residues located near the N-terminus of the molecule. Fractalkine by itself defines one of the chemokine families, and is structurally distinguished from other chemokine families since the N-terminal fractalquinaquin cysteines are separated by three residues (ie, a CX3C pattern) as well as being bound to the cell membrane by a transmembrane extended C-terminator anchor that includes a mucin-like domain, or a mucin-like stem. In this way, the mucin and fractalkine domains contained within the fusion constructs of the present invention are suitable for achieving improved anchoring of the TIMP protein in the cell membrane. TIMP as used in the present invention is preferably derived from a mammal; more preferred is a human (the four TIMPs are reviewed in Mannello F et al., 2001). Example of TIMP proteins that can be used according to the present invention comprise TIMP-1, TIMP-2, TIMP-3 or TIMP-4, and their corresponding variants in other organisms such as mouse, rabbit, dog, cat, sheep and cow . The GPI anchor as used in the present invention is preferably derived from the antigen associated with lymphocyte function (LFA-3), or a portion thereof, and includes a GPI signal sequence mediating the membrane association. The present invention also relates to a nucleic acid molecule, such as RNA or DNA, which comprises a nucleic acid sequence encoding the GPI-anchored TIMP construct of the invention. In a further aspect of the present invention, the nucleic acid molecule of the invention is contained in an expression plasmid, a vector or a host cell for expression of the nucleic acid molecule of the invention. The present invention also relates to the use of TIMP-GPI or TIMP-mucin-GPI fusion constructs of the invention for the treatment of cancer, particularly residual cancer after the surgical removal of a primary tumor.

In a primary aspect of the present invention, the TIMP-GPI or TIMP-mucin-GPI fusion constructs of the invention are contained in a pharmaceutical composition or medicament. In a further embodiment, the TIMP-GPI or TIMP-mucin-GPI fusion constructs of the invention are suitably used as anticancer agents or drugs. In a preferred embodiment, the anti-cancer drugs of the invention are administered and applied locally to the side of tumor mass removal in high-risk tumor patients with a high risk of residual cancer cells and increased incidence of a local relapse, and in those patients who have an obvious residual tumor due to advanced stage disease or local inoperability. Preferably, the fusion construct is administered by wound dispersion and / or injection in regions that are not available for surgery. The present invention also relates to an in vitro method for inhibiting cancer cell proliferation which comprises the steps of subjecting a cancer cell line to an effective amount of the TIMP-mucin-GPI or TIMP-GPI fusion construct. In a further embodiment, the present invention provides novel agents and methods for the treatment of conditions defined by a disturbed equilibrium between the activity of normal physiological MMP protease and the NDE deposition, which results in abnormal healing of the wound. In one embodiment, the present invention provides agents and methods suitable for the treatment of keloid or hypertrophic healing and chronic wounds commonly associated with increased levels of MMP. Additionally, the present invention also provides effective agents and methods for reducing, minimizing or inhibiting scar formation during the wound healing process. Definitions The term "TIMP" as used herein means an inhibitor of endogenous tissue of metalloproteinases, which is known to be involved in physiological / biological functions which include the inhibition of active matrix metalloproteinases, activation regulation pro MMP, cell growth, and modulation of angiogenesis. The human "TIMP family" contains four members, TIMP-1, TIMP-2, TIMP-3 and TIMP-4. A preferred member used in the present invention, TIMP-1, is a secreted protein that can be detected on the cell surface through its interaction with surface proteins (Bode &Maskos, 2003). The term "fusion construct" or "TIMP fusion construct" as used herein refers to both the nucleic acid molecule and the amino acid molecule encoded herein. The invention relates specifically to nucleic acids which encode a sequence of nucleotides that include the sequence defined by SEQ ID NO: 1-5, or a homologue thereof, or unique fragments thereof. In the present invention, the sequence of a nucleic acid molecule encoding the resulting protein is considered homologous to a second nucleic acid molecule if the nucleotide sequence of the first nucleic acid molecule is at least about 70% homologous, preferably at least about 80% homologous, and more preferably at least about 90% homologous to the sequence of the second nucleic acid molecule. The homology between two nucleic acid sequences can be. easily determined using the known BLASTN algorithm (Altschul, et al., 1990) with default parameters. As a further example, another known test for determining the homology of two nucleic acid sequences is whether they hybridize under normal hybridization conditions, preferably under conditions of astringent hybridization. Given the nucleic acid sequence described herein, the skilled person can easily design nucleic acid structures which have particular functions in various types of applications. For example, him skilled artisan can construct oligonucleotides or polynucleotides for use as primers in nucleic acid amplification procedures, such as polymerase chain reaction (PCR), ligase chain reaction (LCR), reaction repair chain (RCR), PCR oligonucleotide ligation assay (PCR-OLA), and the like. Oligonucleotides useful as probes in hybridization studies, such as in situ hybridization, can be constructed. Numerous methods are known for labeling such probes with radioisotopes, fluorescent tags, enzymes, and linker portions (eg, biotin), in this way the probes of the invention can be easily adapted for easy detection capability. Oligonucleotides can also be designed and manufactured for other purposes. For example, the invention allows the design of antisense oligonucleotides, and triplex formation oligonucleotides for use in the study of structure / function relationships. Homologous recombination can be implemented by adapting the nucleic acid described previously for use as a means of objectification. The protein encoded by the nucleic acid of the present invention further includes functional homologs. A protein is considered a functional homologue of another protein for a specific function, as described later, if the homologue has the same function as the other protein. The homolog may be, for example, a fragment of the protein, or a substitution, addition, or mutant deletion of the protein. The determination of whether two amino acid sequences are substantially homologous is, for the purpose of the present invention, based on the FASTA searches according to Pearson & Lipman (1988). For example, the amino acid sequence of a first protein is considered homologous to that of a second protein if the amino acid sequence of the first protein has at least about 70% amino acid sequence identity, preferably at least about 80% of identity, and more preferably at least about 95% identity, with the sequence of the second protein. The possibility of replacing an amino acid in a sequence with an equivalent amino acid is well known. Groups of amino acids known to be equivalent include: (a) Ala (A), Ser (S), Thr (T), Pro (P), Gly (G); (b) Asn (N), Asp (D), Glu (E), Gln (Q); (c) His (H), Arg (R), Lys (K); (d) Met (M), Leu (L), He (I), Val (V); Y (e) Phe (F), Tyr (Y), Trp (W). Substitutions, additions, and / or deletions in the amino acid sequences can be made as long as the protein encoded by the nucleic acid of the invention continues to satisfy the functional criteria described herein. An amino acid sequence which is substantially the same as another sequence, but which differs from the other sequence by means of one or more substitutions, additions, and / or deletions, is considered to be an equivalent sequence. Preferably, less than 20%, more preferably less than 10%, and still more preferably less than 5%, of the number of amino acid residues in a sequence are replaced by, added to, or deleted from the protein encoded by the nucleic acid of the invention. By the term "MMP" as used herein is meant a matrix metalloproteinase belonging to the MMP superfamily as represented by at least 26 extracellular matrix degrading metalloendopeptidases that are acting during tissue development and differentiation, cellular infiltration , wound healing, and as moderators of the immune response. By the term "GPI" as used herein is meant glycoinositol phospholipids, in particular, glycosylphosphatidylinositol as described in Medof et al., nineteen ninety six. These anchors similar to phospholipids have a common structure for membrane binding regardless of the protein function. The GPI anchor units are composed of a linear glycan which contains a phosphoethanolamine, three residues of mannose, and a non-acetylated glucosamine linked to an inositol phospholipid. The GPI sequence contains the signals that direct the GPI anchor. By the term "mucin" or "mucin domain" as used herein is meant a non-membrane or membrane-linked glycoprotein component. Usually, membrane-bound mucins exhibit hydrophobic sequences or transmembrane domains responsible for their anchoring in the lipid bilayer and, optionally, contain one or more domains similar to von Willebrandt factor, which function in the oligomerization of mucin monomers and in the packaging in secretion vesicles. The term "mucin" or "mucin domain" as used herein also comprises mucin stems or mucin-like domains, such as the mucin stems typically found in the CXCL16 chemokines or in fractalkine (CX3CL1). A "TIMP-GPI" fusion construct as used herein relates to TIMP that is directly fused to a GPI link sequence. The construction TIMP-GPI fusion is designed by replacing the 3'-end sequence of mRNA or cDNA from GPI-anchored proteins (ie, a sequence containing the signals that direct the GPI anchor) to the 3'-end sequence mRNA or TIMP cDNA. A "TIMP-mucin-GPI" or "TIMP-muc-GPI" as used herein is related to a TIMP that is fused directly to a mucin domain followed by a GPI linkage sequence. The TIMP-mucin-GPI fusion construct is designed as described for TIMP-GPI but includes the amino acid sequence of a mucin domain between the amino acid sequences of TIMP and GPI. By analogy, "TIMP-fractalkine-GPI" or "TIMP-frac-GPI" is related to TIMP that is directly fused to a fractalkine domain followed by a GPI link sequence. With the term "RCC" is meant a renal cancer carcinoma which is considered to be a progressive tumor with limited therapeutic options due to the tumor resistance to current chemotherapeutic agents and radiation. The RCC serves as a model system in the present invention to show the anti-tumor activity of TIMPI anchored to GPI. The model cell lines used in the present invention are the RCC-25 and RCC-53 cell lines that are established from patients with stage I and stage IV cell carcinomas.

By the term "FAS" is meant a member of the tumor necrosis factor / nerve growth factor receptor family that induces apoptosis independent of TNF-a. Other abbreviations known in the art for FAS are Apol (= Apoptosis induced by protein 1) and CD95. The term "regeneration" generally refers to restoring the integrity of traumatized or otherwise damaged tissue. This term may include the processes of wound healing, tissue repair, and other types of restorative activities which occur at the location where a physiological insult has occurred and then tissue damage. Brief Description of the Figures Figure 1A-1C. Incorporation of TIMP-1-GPI in the cell membranes of RCC. Fig. 1A To demonstrate the reincorporation of the GPI-TIMP-1 protein into the cell membranes, purified TIMP-1-GPI or control rhTIMP-1 is added to the RCC-26, RCC-53 and A498 cells. TIMP-1 is detected on the cell surface by FACS analysis. Gray histograms are isotope control staining, solid line histograms represent TIMP-1 antibody staining. Fig. IB To demonstrate GPI binding following incubation with 200 ng / ml or 700 ng / ml of TIMP-1-GPI or rhTIMP-1, cells are treated with 60 ng / ml of PCL and subjected to FACS analysis. The gray histograms represent isotype control. Fig. 1C ELISA of human TIMP-1 is used to determine the amount of TIMP-1 protein released from RCC cells treated with TIMP-1-GPI (as shown in B) after digestion with PLC. Figure 2. TIMP-1-GPI inhibits the release of proMMP-2 and proMMP-9 from RCC-53 cells. Zymography is used to study the secretion of MMP-2 and MMP-9 proteins from RCC-53. The cells are treated with increased amounts of TIMP-1-GPI, or control rhTIMP-1, and after 48 hours the serum-free culture supernatant is removed and analyzed by gelatinase zymography. Figures 3A-3B. Surface expression of MMP after treatment with TIMP-1-GPI. After incubation of the RCC-53 cells with 700 ng / ml of TIMP-1-GPI protein for 24 hours, FACS was performed using specific antibodies directed against: (Fig.3A) TIMP-1, MMP-1, MMP-2 , MMP-3, MMP-7, MMP-8 and MMP-9, or (Fig.3B) MMP-12, MMP-13, MMP-14, MMP-16, HLA-A2 (HB82), bread HLA Class I (W6 / 32) and ICAM-1. As an additional control, TIMP-1-GPI is cleaved from the surface after one hour by PLC treatment (see Figures 1A-1C). The gray histograms are the isotype control staining, the solid line histograms represent samples treated with TIMP-1-GPI. Fig.3C The secretion of a series of MMPs from RCC53 is tested using Western blot and monoclonal antibodies directed against MMP-1, MMP-3, MMP-7, MMP-8, MMP-12 and MMP-13. The culture medium (serum free) is taken 24 hours after the treatment of RCC-53 with 700 ng / ml of either rhTIMP-1 or TIMP-1-GPI and compared to untreated control cells. Fig.3D The effect of sequestration of MMP on the cell surface in invasion of RCC-53 through Matragel The base model membrane is evaluated. The optimal migration of RCC-53 cells to VEGF (4 ng / ml) is set as a baseline or "zero" and the 100% inhibition value is set at the migration value seen for untreated RCC-53 cells. VEGF signal. The RCC-53 cells are pretreated with 350 ng / ml or 700 ng / ml of either rhTIMP-1 or TIMP-1-GPI. After one hour the cells are washed and applied to the migration chamber. The effect on migration is then evaluated. The data presented represent an average of four wells and two experiments. Figures 4A-4C. The effect of rhTIMP-1 and TIMP-1-GPI protein on the proliferation of RCC lines. The effect of increasing the levels of TIMP-1-GPI or rhTIMP-1 control protein in the proliferation of RCC-53 (Fig.4A), A498 (Fig.4B) and RCC-26 (Fig.4C) is measured using an MTT assay. MTT is added after 24 hours, 48 hours or 72 hours as indicated. Figures 5A-5B. TIMP-1-GPI does not influence the susceptibility of RCC to perforin-mediated apoptosis. There are RCC cells without treatment (g), treated with 700 ng / ml of TIMP-1-GPI (0) or rhTIMP-1 protein (o) by 24 hours and incubate with either CTL lines JB4 (Fig.5A) or NK (Fig.5B) (NKL for RCC-53 and A498, or NK-92 for RCC-26). Representative examples of three independent experiments with similar results are shown. Figures 6A-6C. TIMP-1-GPI does not increase FAS expression but leads cells to be sensitive to FAS-induced apoptosis Fig.6A RCC-53, RCC-25 or A498 cells are treated or not treated with 700 ng / ml of TIMP -1-GPI or rhTIMP-1 for 24 hours, and stained with anti-human FAS (L-958) and analyzed by surface expression of FAS by flow cytometry. Monoclonal antibody isotype control stains are shown as gray histograms. The three RCC cell lines are treated with TIMP-1-GPI or rhTIMP-1 followed by incubation of L-957. The binding of annexin V-fluoroisothiocyanate (FITC) is used to detect viable and early apoptosis by flow cytometry.

The low level of apoptosis in RCC-53 cells in (Fig. B) is verified using a more sensitive cytoplasmic ELISA test. Increased levels of apoptosis are detected after incubation of L-957 with increased levels of TIMP-1-GPI but not with rhTIMP-1. Figures 7A-7C. Expression of BCL-2 and BAX in RCC after treatment with TIMP-1-BPI or rhTIMP-1, analyzed by internal FACS staining and Western blot. RCC-53 cells (Fig.7A), RCC-26 (Fig.7B) and A498 (Fig.7C) are preincubated with TIMP-1-GPI or rhTIMP-1 for 24 hours, then treated with FAS antibody that activates L-957 for an additional 16 hours. The cells are then analyzed with anti-BCL-2 and anti-BAX monoclonal antibodies using flow cytometry. In parallel, proteins are extracted and measured by Western blot. The signals derived from BAX and BC1-2 are normalized to β-actin levels after densitometry. The results of FACS are presented as histograms with values in parentheses corresponding to MFI of either BCL-3 or BAX or corresponding isotype antibodies. Figure 8: General study of tissue remodeling and fibrosis A schematic diagram representing the complex interaction of factors involved in the delicate balance of NDE production and change during the process of healing of wounds. Figure 9: Effect of TIMP fusion constructs on fibronectin production of fibroblasts in the presence of rhTIMP-1 Confluent fibroblasts are cultured in the presence or absence of rhTIMP-1 and TIMP-1-GPI; The expressed and secreted fibronectin is quantified by Western blot analysis using anti-fibronectin antibodies (β-actin served as a control). RhTIMP-1 (at 700 ng / ml) does not lead to any significant decrease in fibronectin expression, whereas TIMP-1-GPI (at 700 ng / ml) strongly reduces the fibronectin that is secreted by fibroblasts. Figures 10A-10B. Effect of TIMP fusion constructs on fibronectin production of fibroblasts in the presence of TNF-α Fibroblasts are cultured in the presence (Figure 10A) or absence (Figure 10B) of 10 ng / ml of TNF-a which activates fibroblasts together with 350 ng / ml of TIMP-1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. At a concentration of 350 ng / ml of TIMP-1-GPI, the transcribed fibronectin RNA is significantly reduced, regardless of whether TNF-a is present in the medium.

Figures 11A-11B. Effect of TIMP fusion constructs on fibroblast IL-6 production Fibronectin RNA is assayed as transcribed by fibroblasts by Northern blot analysis, using a probe for IL-6 RNA. In this way, the fibroblasts are cultured in the presence (Figure HA) or absence (Figure 11B) of 10 ng / ml of the fibroblast that activates TNF-a together with 350 ng / ml of TIMP-1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. At a concentration of 350 ng / ml of TIMP-1-GPI, the transcribed IL-6 RNA is strongly reduced, regardless of whether TNF-a is present in the medium. Figures 12A-12F. Effect of TIMP fusion constructs on collagen production of fibroblasts. The fibronectin RNA as transcribed by fibroblasts is evaluated by Northern blot analysis, using probes for collagen 1A1 (Figure 12A and 12B), collagen 4A2 (Figure 12C and 12D), and Collagen 16A1 (Figure 12E and 12F), respectively. In this way, the fibroblasts are cultured in the presence (Figures 12A, 12C, 12E) or absence (Figure 12B, 12D, 12F) of 10 ng / ml of TNF-a which activates fibroblasts together with 350 ng / ml TIMP- 1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. At a concentration of 350 ng / ml of TIMP-1-GPI, all three transcribed collagen RNAs are significantly reduced, regardless of whether TNF-a is present in the medium. Figures 13A-13B. Effect of TIMP fusion constructs on TGF-β production of fibroblasts Fibronectin RNA as transcribed by fibroblasts is evaluated by Northern blot analysis, using probes for TGF-β. In this way, the fibroblasts are cultured in the presence (Figure 13A) or absence (Figure 13B) of 10 ng / ml of TNF-a which activates fibroblasts together with 350 ng / ml of TIMP-1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. At a concentration of 350 ng / ml of TIMP-1-GPI, almost no TGF-β RNA can be detected, regardless of whether TNF-a is present in the medium, and independently of the FCS content of the medium. Detailed Description of the Invention The TIMP Family and Protein Modification of Cell Surfaces Metalloproteinase tissue inhibitors (TIMPs) are known as the major cellular inhibitors of the matrix metalloproteinase subfamily.

(MMP), which exhibit various degrees of efficacy against different members of MMP, as well as different patterns of tissue expression and modes of regulation. TIMPs typically modulate the activity of soluble MMPs, bound to matrix and cellularly associated. All four mammalian TIMPs have many broad similarities, but exhibit different structural characteristics, biochemical properties and expression patterns, suggesting that each TIMP has a particular function in vivo. The TIMP-1 protein is the most widely expressed and studied member of the TIMP family. Other members of the TIMP family include TIMP-2, TIMP-3 and TIMP-4. TIMP proteins not only share common structural features, which include a series of conserved cysteine residues that form disulfide bonds essential for the conformation of native protein (Brew et al., 2000), but also have extensive biological overlapping activities. The conserved N-terminal region of TIMP proteins is necessary for functional inhibitory activities, whereas it is thought that divergent C-terminal regions modulate the selectivity of inhibition and efficiency of binding of agents to MMPs (Maskos &Bode, 2003). However, apart from their ability to act as MMP inhibitors, the various members of the TIMP family may also exhibit additional biological activities, including the regulation of proliferation and apoptosis in addition to the modulation of angiogenic and inflammatory responses. TIMP-1 has been found to inhibit most MMPs (except MMP-2 and -14), and preferably inhibits MMP-8. TIMP-1 is produced and secreted in soluble form by a variety of cell types and is widely distributed throughout the body. It is an extensively glycosylated protein with a molecular mass of 28.5 kDa. TIMP-1 inhibits the active forms of MMP and forms a complex with the proform of MMP9. Like MMP9, the expression of TIMP-1 is sensitive to many factors. The increased synthesis of TIMP-1 is caused by a wide variety of reagents including: TGF beta, EGF, PDGF, FGFb, PMA, all transretinoic acid (RA), IL 1 and IL-11. TIMP-2 is a 21 kDa glycoprotein that is expressed by a variety of cell types. It forms a stoichiometric, non-covalent complex with both latent and active MMPs. TIMP-2 shows a preference for inhibition of MMP-2. TIMP-3 is typically linked to ECM and inhibits the activity of MMP-1, -2, -9 and -13. TIMP-3 shows 30% amino acid homology with TIMP-1 and 38% homology with TIMP-2. TIMP-3 has been shown to promote the disunion of transformed cells from ECM and to accelerate the morphological changes associated with cell transformation. Because of its high affinity link to ECM, TIMP-3 is unique among TIMPs. TIMP-3 has been shown to promote the disunion of transformed cells from ECM and to accelerate the morphological changes associated with the cellular transformation. TIMP-3 contains a glycosaminoglycan binding domain (GAG) which comprises six amino acids (Lys30, Lys26, Lys22, Lys42, Arg20, Lys45) which are thought to be responsible for an association with the cell surface. TIMP-3 is the only TIMP that normally inhibits TACE (TNF-a converting enzyme), another metalloprotease that releases soluble TNF and is responsible for the processing of the IL-6 receptor in order to play a central part in the process of wound healing. TIMP-4 inhibits all known MMPs, and preferably inhibits MMP-2 and -7. TIMP-4 shows 37% amino acid identity with TIMP1 and 51% homology with TIMP2 and TIMP3. TIMP4 is secreted extracellularly, predominantly in coronary and cerebral tissue and appears to function in a tissue-specific manner with respect to extracellular matrix homeostasis (ECM). Protein modification of cell surfaces is a potentially powerful technology through which the surface protein composition of cells can be manipulated without gene transfer. By substituting the cDNA sequence derived from mRNA from a GPI-linked protein containing the GPI signal domain by the carboxyl terminal region of a protein of interest, it is possible to generate a fusion construct that encodes a protein bound to GPI. This procedure offers multiple advantages over traditional gene transfer procedures. For example, the method is applicable to cells that are difficult or impossible to transfect (eg, primary microvascular endothelium, primary target cells, etc.). The amount of protein added to the cell surface can be controlled and quantified (by FACS or immunofluorescence). In addition, multiple GPI-anchored proteins can be inserted sequentially or concurrently into the same cells. Through molecular modification it is possible to express an additional epitope tag that aids in protein purification as well as in reagent monitoring during experiments. The agent can be injected directly into the tumor or peritumor area and the effect of selective leukocyte recruitment on tumor growth or FAS-induced apoptosis is determined. TIMP fusion constructs for the treatment of cancer The prognosis of malignant tumors is mostly dependent on their clinical and pathological stage. While most carcinomas (for example, primary and secondary tumors) can be completely removed surgically in most cases, the Removal of advanced stage cancer is often not possible and associated with early disease recurrence and increased disease related mortality. Particularly in breast cancer, advanced stage disease with extended tumor size (> 2 cm) is associated with the occurrence of distant metastasis and limited survival. A large tumor volume is also associated to be a critical parameter for the presence of residual cancer, which also presents a high risk for local relapse and the spread of distant metastasis. Similarly, brain tumors such as glioblastoma (grade IV astrocytoma) is another conceivable target for tumor surveillance, since complete surgical removal is almost always impossible and local relapse occurs in 95% of all cases within one year of primary surgery. In order to solve the problem that is linked to residual cancer, it is necessary to identify a novel therapy option for cancer that is particularly useful for the treatment of residual cancer after incomplete surgical excision. As a solution, the present invention provides TIMP anchored to GPI, which can be applied locally at the excision margins to attract immune cells and focus residual tumor monitoring on the residual cancer cells. For this purpose, TIMP proteins are anchored by GPI, and when they are purified and added to cancer cells they are incorporated into their surface membranes and are fully functional. By substituting the 3 'end sequences - mRNA from GPI-anchored proteins naturally (i.e., a sequence containing the signals directing the GPI anchor) by the endogenous 3' end-mRNA sequence, virtually any of the proteins TIMP can be expressed as a derivative anchored to GPI. In the present invention, the incorporation of the purified GPI-TIMP protein into the surface membranes of tumor cell lines is demonstrated by incubation of the cell lines with TIMP-1-GPI, purified TIMP-1-mucin-GPI or control protein ( rh) Recombinant human TIMP-l. As detailed below, surface expression with TIMP-1 anchored to GPI results in a strong surface signal for TIMP-1. As used herein, the terms "isolated and / or purified" refers to the in vitro isolation of a DNA or polypeptide molecule from its natural cellular environment, and from association with other components of the cell, such as nucleic acid or polypeptides, such that they can be sequenced, replicated, and / or expressed.

For example, "isolated GPI-binding sequence" is RNA or DNA which contains more than 9, preferably 36, and more preferably 45 or more, sequential nucleotide bases that encode at least a portion of a linking sequence, or a variant thereof, or an RNA or DNA complementary thereto, which is complementary or hybridizes, respectively, to RNA or DNA encoding the binding sequence and remains stably bound under stringent conditions, as defined by methods well known in the art. technique. In this way, the RNA or DNA is "isolated" since it is free of at least one contaminating nucleic acid with which it is normally associated in the natural source of RNA or DNA and is preferably substantially free of any other RNA or DNA from mammal. As used herein, the term "recombinant nucleic acid" for example, "recombinant DNA sequence" refers to a nucleic acid, eg, DNA, that has been derived or isolated from any appropriate tissue source, which may be subsequently altered chemically in vitro, such that its sequence is not found naturally, or corresponds to naturally occurring sequences that are not positioned since they can be positioned in a genome which has not been transformed with exogenous DNA . An example of DNA "derived" from a source can be a DNA sequence that is identified as a useful fragment within a given organism, and which is then synthesized chemically in essentially pure form. An example of such "isolated" DNA from a source can be a useful DNA sequence that is cleaved or removed from the source by chemical means, for example, by the use of restriction endonucleases, such that it can to be further manipulated, for example, amplified, for use in the invention, by known methodology of genetic modification. Unlike conventional polypeptide anchors, which have different membrane sequences and connect to specific cytoplasmic extensions, these phospholipid-like anchors use a common structure as a general mechanism for membrane binding regardless of protein function. The GPI anchor units are composed of a linear glycan which contains a phosphoethanolamine, three mannose residues, and a non-acetylated glucosamine linked to an inositol phospholipid. These are prefabricated in the endoplasmic reticulum (ER) and are added to primary translation products at the time of their translocation between the ER membrane. The GPI-modified products are then glycosylated in the ER and Golgi apparatus, and subsequently transported to the cell surface.

The preferred GPI binding sequences that can be used in the present invention are derived from GPI anchors that are isolated from, for example, enzymes such as alkaline phosphatase, acetylcholinesterase, 5 'nucleotidase (C073); adhesion molecules such as antigen associated with lymphocyte function (LFA-3); CD58); Neural cell adhesion molecule (NCAM); complementary regulatory proteins such as decomposition acceleration factor (DAF or CD55); or others such as the Fey type III B receptor (Fc-y-RIII or CD16b), Thy-1 (CD90), Qa-2, Ly-6A. The reactive lysis membrane inhibitor (MIRL or CD59). For the purpose of the present invention, the lymphocyte-associated antigen (LFA-3) is preferred. The skilled person will recognize that also any other known GPI anchors may be used for the practice of the present invention. For the construction of TIMP-GPI, either the total length sequence of TIMP can be used in the fusion construct or a functionally active portion thereof, which retains TIMP activity. Similarly, a portion of the GPI sequence can also be used as long as the portion allows incorporation of the TIMP protein into the surface cell membrane of the cancer cells. Next, a plurality of modalities that are related to TIMP fusion constructions, where constructions are produced and provided for the treatment of cancer and as agents in the field of regenerative medicine. In a first embodiment, the TIMP molecule is selected from the group which consists of TIMP-1, TIMP-2, TIMP-3, and TIMP-4 and is preferably fused to a GPI sequence. In another preferred embodiment, the GPI sequence has 36 amino acids in length. In yet another embodiment, the TIMP molecule is selected from the group which consists of TIMP-1, TIMP-2, TIMP-3 and TIMP-4 and is fused to a mucin or fractalkine domain followed by the GPI sequence . In a further embodiment, the TIMP protein that is selected from the group which consists of TIMP-1, TIMP-2, TIMP-3 and TIMP-4 and fused to the GPI sequence, or fused to a mucin domain or Fractalkine followed by the GPI sequence, or fused to a mucin domain or fractalkine domain followed by the GPI sequence, is truncated at the C (carboxyl) terminal. In a preferred embodiment, the TIMP molecule is truncated in the first 50, 50-100 or 50-152 N-terminal amino acid residues (amino terminal). More preferably, the TIMP molecule is truncated at the first 152 N terminal amino acid residues and is the TIMP-1 molecule. The term "truncated" refers to the TIMP nucleic acid or amino acid sequence that contains less than the total number of nucleic acid bases or amino acid residues found in a nucleic acid sequence of TIMP or native protein or to a nucleic acid sequence or sequence of nucleic acids. amino acid that has been removed from the unwanted sequences. In yet a further embodiment, the TIMP fusion construct is defined by a sequence selected from the group which consists of SED ID Nos: 1, 2, 3, 4 and 5. The obtained construct can then be expressed in any suitable cell line or host cell to obtain the functional TIMP protein or polypeptide. For this purpose, any of the suitable known vectors or plasmids can be used to express the TIMP proteins anchored to GPI of the present invention. As described in more detail below, the target cancer cells treated with TIMP-GPI protein (and as a control with the rhTIMP-1 protein) are recognized by the protein constructs and, as a consequence, die due to FAS-mediated apoptosis. In a preferred embodiment, and by the exemplary form, a vector used for expression of the fusion constructs of the present invention contains the promoter for the human alpha-1 factor of lengthening followed by a site of multiple cloning and an internal ribosomal binding site which allows bicistronic expression of the construct and dihydrofolate reductase (DHFR) used as a selection marker (Mack, et al., P.N.A.S. USA 92: 7021, 1995). The 3 'end (carboxyl terminal) of the TIMP protein is either directly fused to a GPI-binding sequence (e.g. derived from antigen-3 associated with lymphocyte function (LFA-3)) or the mucin-like domain isolated of CXCL16 or fractalkine (CX3CL1) followed by the GPI signal. As indicated above, these mucin regions are fairly composed of serine / threonine / glycine / proline residues shown to facilitate cell-cell interactions. The resulting fusion constructs are transfected into Chinese hamster ovary (CHO) cells deficient in dihydrofolate reductase (DHFR) and the screening is performed as described (Mack, et al., PNAS USA 92: 7021-7025, 1995). In a preferred embodiment, the transfectants can be exposed to methotrexate to increase the rate of expression by gene amplification. In a further embodiment, the TIMP-GPI construct can also be fused to a mucin domain to increase the efficiency of membrane incorporation of TIMP-GPI proteins. Mucins are components of glycoprotein bound to membrane or non-membrane that are first identified in secreted mucus that lines the surfaces of glandular epithelium. The membrane-bound mucins exhibit hydrophobic sequences or transmembrane domains responsible for their anchoring in the lipid bilayer. At the moment, a total of 21 genes have received the name MUC: MUC1-2, MUC3A, MUC3B, MUC4, MUC5AC, MUC5B, MUC6-13, MUC15-20 (Moniaux N, et al., 2004). The five common characteristics of a mucin are: (1) secretion in the mucosal layer. (2) High molecular weight O-glycoprotein, (3) presence of a random repeat arrangement encoded by a centrally positioned and unique large exon, (4) presence of a predicted peptide domain which contains a high percentage of residues of serine and threonine, and (5) a complex pattern of mRNA expression. With one exception (MUC7), secretory mucins (MUC2, MUC5AC, MUC5B and MUC6) have one or several domains similar to von Willebrandt factor, peptides rich in cysteine, which function in the oligomerization of mucin monomers and in packaging in secretion vesicles. Typically, secreted mucins are expressed exclusively by specialized epithelial cells, secreted in the mucus, and exhibit a restricted expression pattern within the human body. The four secretory mucins, also referred to as the gel-forming mucins, have an architecture common with a high level of similarity to the pro von Willebrand factor. They are also known to host five D domains due to their homology to the D domains of von Willebrand factor. The membrane bound mucins are composed of MUC1, MUC3A, MUC3B, MUC4, MUC11, MUC12, MUC16 and MUC17. The membrane-anchored mucins contain a SEA module (hedgehog sperm protein, Enterokinase and Agrin), with the exception of MUC4. MUC3A-B, MUC4, MÜC11-12 and MUC17 contain two to three domains similar to epidermal growth factor (EGF). Examples of membrane-bound mucins that can be used in the present invention are MUC1, MUC3, MUC4 and MUC12. In a preferred embodiment of the invention, the mucin stalk of the chemokine CXCL16 or fractalkine (CX3CL1) associated with the surface is used. CXCL16 is a member of the CXC chemokine subfamily. Unlike other members of this subgroup, CXCL16 is structurally different and has four distinct domains: a chemokine domain bound to the cell surface by means of a mucin-like stem, which in turn binds to the transmembrane domains and cytoplasmic Fractalkine (CX3CL1) has a structure similar to that of CXCL16, and both CXCL16 and fractalkine act as adhesion molecules when expressed on the cell surface, and before cleavage from the cell surface, The soluble chemokines act as chemoattractants. Preferably, the mucin domain is fused between the 3 'end of the TIMP sequence and the 5' end of the GPI anchor sequence by any of the known conventional genetic modification methods. The TIMP-mucin-GPI fusion construct obtained from the invention can then be transfected and expressed in any known cell line or suitable host cell. The skilled person will recognize that any other mucins or mucin domains are suitable for the purpose of the present invention. In a preferred embodiment, the fractalkine fused to the TIMP molecule comprises amino acids 100-342 of CX3CL1 followed by the GPI sequence. In even a more preferred embodiment, the TIMP molecule is the TIMP-1 molecule truncated to the first 152 N-terminal amino acids (SED ID NO: 5). Although the use of TIMP-1 (Bode &Maskos, 2003) for the preparation of the TIMP protein anchored to GPI is preferred in the present invention, the person skilled in the art will recognize that also other TIMP proteins can be used for the practice of the present invention. Additional examples of human TIMP that are useful are TIMP-2, TIMP-3 and TIMP-4 (Mannello F, et al., 2001). The TIMPs used are derived from human sources and administered to treat the human cancer cells. The skilled person will probably recognize that also TIMP-1 homologs, in particular TIMP-1, in organisms other than humans will have a similar effect on killer tumor cells. For example, in some embodiments, the TIMP-1 sequence derived from an animal such as dog, cat, mouse, rabbit, cow or sheep, and birds can be used for the construction of a TIMP-GPI fusion construct of the present invention. The TIMP-GPI chimera will subsequently be applied to the tumor site in a similar manner as described for a human individual. Tumor and cancer cells that can be treated with GPI-anchored TIMP include the following but not limiting cancers: breast cancer, kidney cancer, prostate cancer, leukemias, seminomas, melanomas, teratomas, lymphomas, neuroblastomas, gliomas, rectal cancer , endometrial cancer, kidney cancer, adrenal cancer, thyroid cancer, blood cancer, skin cancer, brain cancer, cervical cancer, intestinal cancer, liver cancer, colon cancer, stomach cancer, intestinal cancer, gastrointestinal cancer, cancer of the lymph node, esophageal cancer, colorectal cancer, pancreatic cancer, cancer of the ear, nose and throat (ENT), cancer of the uterus, ovarian cancer and lung cancer and its metastasis.

For the treatment of residual cancer, TIMP-GPI can be administered locally and applied next to the removal of tumor mass in patients with high-risk tumors with a high risk of residual cancer cells and an increased incidence of a local relapse and in those patients with an obvious residual tumor due to advanced stage disease or local inoperability. Preferably, the fusion construct is administered in a concentration of 0.5 to 5 μg / ml protein, more preferred in 0.5 to 1 μg, or 1 to 2 μg / ml. A concentration of approximately 1 μg / ml of TIMP-GPI or TIMP-mucin-GPI is more preferred. The protein can be administered to the individual by any applicable routes of administration. It is preferred that the treatment be performed during surgery in such a way that the fusion construct is dispersed in the wound or is injected into regions that are not available in the surgery. For this purpose, the GPI-anchored TIMP fusion construct of the present invention may be a constituent of a pharmaceutical composition or medicament which further comprises one or more conventionally known carriers, diluents and excipients. As it is concluded from the following examples, TIMP anchored to GPI seems to induce its anti-tumor activity, ie the death of tumor cells, not by apoptosis induced by CTL and NK cells (lytic path mediated by perforin / granzyme) but rather by the second trajectory, which implies apoptosis mediated by FAS / CD95 (for further details see examples 5 and 6, figures 5 and 6). Additionally, while many tumor cells are resistant to FAS-mediated apoptosis, treatment with TIMP-1-GPI, but without control rhTIMP-1, leads to cell lines to be sensitive to FAS-mediated apoptosis. It was also found that treatment with TIMP-1-GPI protein reduces BCL2 and increases the expression of BAX protein. BCL2 proteins represent a family of proteins involved in the control of apoptosis. Some members of this family (such as BCL2 and BCL-XL) are anti-apoptotic, while others (such as BAD or BAX) are pro-apoptotic. The sensitivity of cells to apoptotic stimuli depends on the balance between members of the BCL2 pro and anti-apoptotic family. In addition, the effect of treatment with TIMP-1-GPI on the expression of BCL2 and BAX is determined and it is shown that the treatment of cancer cells with TIMP-1-GPI increases the expression of pro-apoptotic BAX, and decreases the expression of BCL2 anti-apoptotic. Similar results are obtained for the TIMP fusion constructs encoded by SEQ ID NO: 1, 2, 3, 4 and 6, respectively.

In summary, the availability of the methodology to produce quantities of micrograms to milligrams of TIMP proteins anchored to recombinant GPI together with the capacity of incorporation of these molecules on the surfaces of cancer cells provides an effective tool for the treatment of cancer. TIMP fusion constructs for use in regenerative medicine Additionally, the fusion constructs of the present invention are suitable for use in regenerative medicine, particularly in the area of wound healing. As described above, TIMP proteins that are fused to GPI or mucin-GPI or fractalkine-GPI are efficiently incorporated into the cell surface membrane, where they focus functional domains on those cell surfaces independently of protein-protein interactions. The TIMP fusion constructs of the present invention are typically very stable and exhibit novel and amplified bioactivities. As described above, both MMP and TACE play a crucial role in the wound healing process. Increased MMP levels are associated with various wound healing disorders, inter alia occurrence of chronic wound. Since TIMPs are natural inhibitors of MMP, the fusion constructs of the present invention they can also be used as effective therapeutic agents to control the wound healing process, for example, in regenerative medicine and are suitable for treating disorders characterized by an increase in MMP levels. In this way, the present invention provides agents and methods suitable for use in regenerative medicine and / or for treating disorders characterized by an increase in MMP levels. In a preferred embodiment, the fusion constructs of the present invention are used to treat or prevent excessive scarring, and healing of abnormal wounds including keloid or hypertrophic scarring and / or chronic wounds. In a further preferred embodiment, the fusion constructs of the present invention are used to inhibit or prevent the formation of scar tissue. A typical wound healing response is characterized by the movement of specialized cells at the site of the wound. Platelets and inflammatory cells are usually the first cells that reach the site of damage and these molecules provide important functions and chemical signals, which include cytosines that are necessary for the flow of connective tissue cells and other healing factors. The term "wound" means an alteration of normal physiological structure and function.

In this way, the process of wound healing refers to the complex and dynamic sequence of events that ultimately result in the restoration of physiological continuity and function. When a wound heals, scarring usually develops instead. During the course of normal wound healing, simple tissues such as fat, connective tissue, and epithelium are regenerated. However, since the skin is a more complex organ that is derived from the two germ layers, it is cured by the formation of a predominantly fibrous tissue, that is, a scar. In the normal healing of a wound, the proteolytic activity of MMP is controlled by several mechanisms which include gene transcription, enzyme production, and local secretion of endogenous TIMP inhibitors. During the repair of the wound, there is a physiological balance between the activities of MMP and TIMP. However, it is known that matrix metalloproteases are at high levels in chronic wounds and such high concentrations of MMPs are known to damage the healing process of the wound. Multiple cell types, which include macrophages, fibroblasts, neutrophils, epithelial cells, and endothelial cells, synthesize MMP in the presence of biochemical signals specific such as inflammatory cytosines. MMPs are able to digest almost all the components of the extracellular matrix, which challenges the balance required between the protein degrading activities of MMP and other cellular activity that synthesizes and deposits the protein components of tissue. Figure 8 provides a summary of the process of tissue remodeling, fibrosis and those factors involved in modulating this process. Before an acute and chronic inflammatory reaction to a particular damage, a parasitic infection, or an autoimmune response, fibrogenic factors are expressed and secreted in this way leading to the activation of fibroblasts and keratinocytes. These fibrogenic factors include, inter alia, TGF-β, IL-1β, IL-α, MOB, TGF-α, IL-4, IL-13, bFGF, TNF-α and PDGF-BB. Perhaps the two most important of these cytosines are: TNF, which is mitogenic for fibroblasts and promotes angiogenesis and is secreted by macrophages, mast cells, and T lymphocytes, and TGF-a, which is mitogenic for keratinocytes and fibroblasts, stimulates migration of keratinocytes, and is secreted by macrophages, T lymphocytes, and keratinocytes. The stage which involves secretion of TNF and TGF marks the transition from the inflammatory phase of the wound healing process in the process of tissue reconstruction, ie, the proliferative phase.

Upon activation, fibroblasts secrete IL-6 which, in combination with TGF-β, leads to a proliferation of fibroblasts. TGF-β also promotes the differentiation of fibroblasts. The result of these processes of proliferation and differentiation is a total increase in collagen, fibronectin, TIMP, MMP, as well as other ECM proteins, which lead to an increase in ECM production and a decrease in ECM change. The present invention is based on the unexpected discovery that the described fusion constructs, ie the TIMP proteins or mutants thereof, fused to a GPI anchor, a mucin-GPI anchor or a fractalquina-GPI anchor can be used as a powerful agent to influence the level of expression and / or activity of cytosines and other important enzymes involved in the wound healing process. In this way, the present invention offers effective regenerative agents and methods for controlling the wound healing process (eg influencing, inhibiting or preventing the formation of healing tissue) and for treating other known dysfunctions associated with the wound healing process. . Abnormal healing of wounds Keloid and hypertrophic scarring is characterized by an accumulation of excess collagen and are distinguishable from each other by their physical appearance. Both Keloid and hypertrophic scars are wounds that heal externally in an exaggerated way to the surface of the skin. A keloid scar typically continues to enlarge beyond the size and shape of the wound, while hypertrophic scarring enlarges within the physical confines of the original wound. Hypertrophic scarring is usually observable soon after wound damage, while keloid scarring can be formed as late as one year after the time of injury. However, almost all examples of abnormal scarring are associated with physiological insults including tattoos, burns, injections, bites, shots, trauma, surgery, or infection. Hypertrophic and keloid scars can both be described as variations of the typical wound healing process. In a typical wound, the anabolic and catabolic processes achieve a balance approximately 6-8 weeks after the original damage. During the maturation of healing, the tensile strength of the skin improves as the collagen fibers are progressively crosslinked. At this point, the healing is usually hyperemic and can be thickened. However, the initial healing tissue tends to be sublateral over a period of months that produce more mature healing that is typically flat, white, flexible and possibly extended in appearance. Where there is an imbalance between the anabolic and catabolic phases of the wound healing process, more collagen is produced than that which degrades, and healing can therefore grow in all directions. An optimal, simple method to treat hypertrophic and keloid scar tissue has not been developed, so the rate of recurrence of these abnormal cicatrizations is significant. In summary, specific cytosines and enzymes, which include MMP and TIMP, play a crucial role in the wound healing process and in the formation of healing tissue. Additionally, abnormal expression of MMP and cytosine levels are often associated with healing of abnormal wounds. In this way, the present invention offers effective regenerative agents and methods for controlling the wound healing process and / or for treating dysfunctions commonly associated with wound healing. Specifically, the fusion constructs of the present invention can be used to effectively control, alter, inhibit or even prevent these undesirable processes. The fusion constructions of the present invention can be formulated as pharmaceuticals and administered at the site of damage. In one modality, the damage site is created by surgery, a burn, an injection, a bite, a vaccine, trauma, surgery or infection. In another embodiment, the fusion construct used for the preparation of the medicament to be applied to the damage site is selected from the group which consists of TIMP-1-GPI, TIMP-2-GPI, TIMP-3-GPI, TIMP-4 -GPI, TIMP-1-muc-GPI, TIMP-2-muc-GPI, TIMP-3-muc-GPI and TIMP-4-muc-GPI or mutants thereof. Formulation of TIMP constructs and modes of administration The pharmaceutical compositions based on the TIMP constructs of the present invention can be formulated in a conventional manner using one or more physiologically acceptable carriers or excipients. Techniques and formulations can generally be found in Remrnington's Pharmaceutical Sciences, Meade Publishing Co., Easton, Pa. For injection purposes, the compounds of the invention can be formulated in a liquid solution, preferably in a physiologically compatible buffer such as Hank's solution or Ringer's solution. In addition, the compounds may be formulated in solid form and re-dissolved, or suspended immediately before use. The lyophilized forms are also suitable. In addition to these formulations, the compounds can also be formulated as a preparation of deposition. These deposition formulations that act for a long time can be administered by implantation (eg, subcutaneously or intramuscularly) or by injection. In this way, the compounds can be formulated with suitable polymeric or hydrophobic materials (for example, as an emulsion in an acceptable oil) or as an ion exchange resin, or as a slightly soluble derivative, such as a slightly soluble salt. Other suitable delivery systems include microspheres, which offer the possibility of a non-invasive and local delivery of drugs over an extended period of time. This particular technology uses microspheres which have a precapillary size which can be injected by means of a coronary catheter into any selected part of a tissue, for example the heart or other organs, without causing a resultant inflammation or ischemia. The therapeutic agent administered is slowly released from these microspheres and readily taken up by the cells present in the surrounding tissue (eg, damaged or cancerous cells). For topical administration, the oligomers of the invention can be formulated into ointments, ointments, gels, or creams generally known in the art. A washing solution which contains the oligomer can be used locally to treat damage or inflammation or to accelerate generally the healing process. The TIMP constructs of the present invention can be combined when a medicament is prepared, such that the resulting medicament comprises more than one, preferably two, and even more preferably three different TIMP constructs. With this procedure, the amplified and novel bioactivities of the different members of the TIMP family can be combined and objectified preferably to the cell surface, which leads to a synergistic effect. For example, fusion constructs of TIMP-1 inhibit most MMPs, except MMP-2 and MMP14. Therefore, any of the TIMP constructs can be combined with any of the TIMP-2 or TIMP-4 constructs which preferentially inhibit MMP-2. Therefore, by means of this combination, a more complete inhibition of the MMP family can be achieved. In one embodiment, the formulations of the present invention therefore comprise a TIMP-1 construct, or a TIMP-2 and / or TIMP-4 construct. In a preferred embodiment, the formulation comprises a TIMP-1 construct selected from the group which consists of truncated TIMP-1-GPI as encoded by SEQ ID NO: 1, truncated TIMP-1-frac-GPI as encoded by SEQ. ID NO: 5, and TIMP-1-muc-GPI truncated as encoded by SEQ ID NO: 2 and a construct of TIMP-2 and / or TIMP-4. Preferably, the construction of TIMP-2 is encoded by SEQ ID NO: 3. In a further embodiment, the formulation comprises a TIMP-3 construct of the present invention, preferably that is encoded by SEQ ID NO; 4, which inhibits TACE together with at least one of the TIMP constructs selected from the group which consists of truncated TIMP-1-GPI as encoded by SEQ ID.

NO: l, truncated TIMP-1-frac-GPI, TIMP-1-muc-truncated GPI, a TIMP-2 and TIMP-4 construct. Preferably, the TIMP-1 and TIMP-2 constructs are encoded by SEQ ID NO: 1, 2, 3 and 5, respectively. EXAMPLES In the following examples, the anti-tumor effects of TIMPI anchored to GPI in cancer cells are described in more detail. Although the described experiments are carried out with human TIMP-1, the invention should not be limited in this type of TIMP. In the following examples, an anchor of GPI to TIMP-1 is fused to focus defined concentrations of this inhibitory protein on the surface of three renal carcinoma cell lines (RCC) (RCC-26, RCC-53 and A498) independently of protein-protein interactions on the cell surface. As shown in the following, the TIMP-1-GPI added exogenously inserted efficiently into the cell membrane of RCC alters dramatically alters the association of MMP with the cell surface. Treatment with TIMP-1-GPI inhibits the proliferation of RCC and leads RCC cells normally resistant to FAS to be sensitive to FAS-induced apoptosis but do not alter perforin-mediated lysis by cytotoxic effector cells. The increased sensitivity to FAS-mediated apoptosis correlates with an alteration in the balance of proteins of the BCL-2 pro and anti-apoptotic family. The cell lines RCC-26 (Schendel et al., 1993) and RCC-53 are established from local patients with clear cell carcinomas of stage I and stage IV, respectively. For this reason, they represent the two clinical extremes of RCC. The CTL that infiltrate the tumor are isolated from the tumor of both patients. Although these naturally occurring effector cells are unable to control tumor growth in vivo, the surface marker staining of RCC-26 and RCC-53 reveals good surface expression of MHC class I, and both lines are shown to induce Allo-specific and anti-tumor CTL in vitro ((Schendel et al., 2000) and DJS, unpublished observation). A498 was originally isolated from the tumor of a 52-year-old man and is a well-studied example of RCC (Girad et al., 1973).

Example 1 Incorporation of exogenously added TIMP-1-GPI on the surface of RCC-53 The TIMP-1 protein anchored to GPI is generated and isolated as previously described (Djafarzadeh et al., 2004). The incorporation of purified GPI-TIMP-1 protein in the surface membranes of the RCC cell lines RCC-53, RCC-26 or A498 is demonstrated by incubation of the cell lines with 700 ng / ml of purified TIMP-1-GPI or recombinant human TIMP-1 control protein (rh) for one hour. The surface-associated TIMP-1 protein is then detected using FACS. (Figure 1A). The addition of control rhTIMP-1 does not lead to change in the change of FACS, however, TIMP-1 anchored to GPI results in a strong surface signal for TIMP-1. To demonstrate that the exogenously added protein is anchored to GPI, the RCC-53 cells are first incubated with the TIMP-1-GPI protein (200 or 700 ng / ml), and then treated with 60 ng / ml phospholipase C ( PLC). The FACS analysis demonstrates the complete loss of cellular surface signal TIMP-1 following the digestion of PLC (Figure IB). To measure the integration efficiency of TIMP-1-GPI, the TIMP released from the membrane is collected in the wash buffer, and quantified using ELISA specific for TIMP-1 (Figure 1C). The results show that 66% of the starting TIMP-1 antigen is recovered from the 200 ng / ml mixture, while 31% can be recovered from incubation of 700 ng / ml. Example 2 The TIMP-1-GPI protein blocks the release of proMMP-2 and proMMP-9 from RCC-53. An increased expression of MMP-2 and MMP-9 correlates with a poor RCC prognosis (Hemmerlein et al., 2004). The clear cell carcinoma cell line of stage IV, RCC-53 constitutively secretes both proMMP-2 and proMMP-9 (Figure 2). The effect of increasing the superficial TIMP-1 levels on the constitutive release of the MMP-2 and MMP-9 proteins is tested using gelatinase zymography assays (Djafarzadeh et al., 2004; Klier et al., 2001). The rhTIMP-1 protein at 600 or 1200 ng / ml has no effect on the secretion of proMMP-2 or proMMP-9. In contrast, starting at 10 ng / ml, treatment with TIMP-1-GPI shows a concentration-dependent decrease in both the release of proMMP-2 and proMMP-9 in the growth medium. Example 3 Treatment with TIMP-1-GPI leads to an increase in the surface expression of matrix metalloproteinases. Based on the results of the gelatinase zymography experiments, it is possible that TIMP-1-GPI may act by sequestering MMP on the cell surface. TIMP-1 binds most active forms of MMP, the exceptions being MMP-14 and MMP-16 (Brew et al., 2000; Lang et al., 2004). After incubation of RCC-53 with 700 ng / ml of TIMP-1-GPI protein for 24 hours, FACS analyzes using antibodies specific to MMP-1, MMP-2, MMP-3, MMP-7, MMP -8, MMP-9, MMP-12, MMP-13, MMP-14, MMP-15 and MMP-16 show, with the exception of MMP-14, an increase in mean channel fluorescence intensity (MFI) for each one of the MMP. The rhTIMP-1 control protein has no obvious effect on the FACS signal (data not shown). The surface expression of other proteins which include class I MHC (class I bread and HLA-A2) and ICAM-1 is not effected by treatment with TIMP-1-GPI. Digestion of RCC53 treated with TIMP-1-GPI with PLC after one hour (as done in Figures 1A-1C) shows no increase in MMP (Figure 3A and 3B). The accumulation of MMP on the cell surface reflects the reduction in secretion of proMMP-2 and proMMP-9 shown in Figure 2. To further test this apparent blocking of MMP release, Western blot experiments are performed using monoclonal antibodies directed against MMP-1, MMP-3, MMP-7, MMP-8, MMP-12 and MMP-13 in media (serum-free) derived from control RCC53 cells or cells treated 24 hours with 700 ng / ml already be rhTIMP-1 or TIMP-1-GPI (Figure 3C). The presence of each of the MMPs is detected in the media from the control RCC-53 cells. Incubation with rhTIMP-1 does not eliminate the secretion of MMPs. In contrast, TIMP-GPI appears to completely block the release of each MMP studied. To evaluate the effect of this surface accumulation of MMP on the ability of cells to invade ECM, the migration / invasion capacity of the cells is evaluated using a modified Boyden chamber assay with ECM coated membranes. The total capacity of RCC-53 cells to invade ECM is not very pronounced (data not shown). To increase the invasion, it is applied to increase the levels of vascular endothelial growth factor (VEGF) to the lower wells and the experiments are run for 48 hours. An optimal invasion response is observed at 4 ng / ml VEGF (data not shown) and this response is set as baseline invasion, or zero% inhibition (Figure 3D). RCC-53 cells that are pretreated for 30 minutes with 350 or 700 ng / ml of either rhTIMP-1, or TIMP-1-GPI are then washed, and applied to the upper well of the Boyden chamber. The relative increase or decrease in migration / invasion is then determined (see Materials and Methods). While treatment with rhTIMP-1 partially blocks the invasion of the cells, TIMP-GPI in 700 ng / ml completely blocks the invasion of the cells RCC-53 (Figure 3D). Example 4 TIMP-1 anchored to GPI affects the proliferation of RCC To evaluate the effect of surface modification of TIMP-1 on the proliferation of RCC, MTT assays are performed. The exogenously added TIMP-1-GPI protein is found to produce a dose-dependent decrease in proliferation of RCC-53 and A498 in 24, 48 and 72 hours (Figure 4B). The RCC-25 cells proliferate extremely slowly (48+ hours doubled speed) and the general trend suggests a suppression of proliferation (Figure 4C). Additional controls using phosphatidylinositol at a molar concentration equal to the TIMP-1-GPI reagent (Sigma, Germany, Nr. P6636) do not lead to significant changes in cell proliferation (data not shown). Examples Cell mediated cytotoxicity MMP activity has been linked to the sensitivity of target cells to apoptosis induced by cytotoxic T cell activity (both perforin / granzyme-mediated apoptosis and FAS) (Egeblad &Werb, 2002). The effect of TIMP-1-GPI on the death of RCC dependent cells is tested using apoptosis induced by the CTL and allogeneic NK cells. In allogeneic CTL-mediated assays, target cells RCC-53, A498 and RCC-26 are treated with 700 ng / ml of rhTIMP-1 or TIMP-1-GPI, then labeled with Cr51 and incubated with CD8 + CTL JB4 allogeneic (Figure 5A) or NK cell lines (Figure 5B). RCC tumor cell lines are recognized and killed effectively by both CTL and NK cells. Treatment with TIMP-1 or TIMP-1-GPI does not alter the susceptibility of the three RCC lines to either apoptosis mediated by CTL or NK cells. Example 6 Treatment with TIMP-GPI leads to RCC to be sensitive to death mediated by FAS The CTL / NK experiments show that the treatment with TIMP-1-GPI does not influence the lytic trajectory mediated by perforin / granzyme measured in the assay of Chromium release. This trajectory acts rapidly using the secretion of stored cytotoxins to initiate apoptosis and represents one of the mechanisms of cell death initiated mainly by immunity (Trapani et al., 2000). The second path involves FAS / CD95 linkage. The effect of TIMP-1-GPI treatment on FAS-mediated apoptosis is then determined. The expression of FAS in the RCC lines is first evaluated by flow cytometry using mAB anti-FAS not activating (L-958) (H. Engelmann, unpublished results). Untreated cells and treated cells are stained with 700 ng / ml of TIMP-1-GPI or rhTIMP-1 control protein for 24 hours with L-958 and analyzed. The FAS protein is strongly expressed by all three RCC lines. Treatment with TIMP-1-GPI or rhTIMP-1 does not affect the cellular surface expression of FAS (Figure 6A). MAB L-957 anti-FAS can induce apoptosis in cells expressing FAS (H. Englemann, unpublished results). The binding of V-fluoroisothiocyanate annexin (FITC) and the incorporation of propidium iodide (PI) into RCC cells after treatment with L-957 is used to detect apoptosis by flow cytometry . As shown in Figure 6B, RCC-26 and RCC-53 are quite resistant to FAS-mediated apoptosis. The fluorescence intensity (MFI) of untreated cells treated with L-957 is similar. A slight increase in MFI is observed in RCC-53 in response to treatment with L-957. These observations are in line with previous reports that RCC is generally resistant to FAS-mediated apoptosis (Frost et al., 2003). Treatment with TIMP-1-GPI (L-957 + TIMP-1-GPI), but without control rhTIMP-1 (L-957 + rhTIMP-1), leads to cell lines to be sensitive to FAS-mediated apoptosis ( Figure 6B). It is found that A498 is more sensitive to mAB anti-FAS activator (detected by annexin-MFI increased in samples treated with L-957), treatment with TIMP-1-GPI, but without rhTIMP-1, significantly increases the apoptosis of A498 cells. While RCC-26 and A498 cells show a dramatic increase in FAS-induced apoptosis after treatment with TIMP-1-GPI, the RCC-53 cell line shows a less pronounced increase in sensitivity (change in MFI from 62 93). To confirm the apoptosis induced by TIMP-1-GPI / FAS of RCC-53, a second ELISA assay is used in the detection of cytoplasmic chromatin. The advantages of this assay include the lack of subjectivity in interpreting the results and its increased sensitivity relative to annexin V staining. Chromatin ELISA is able to detect as little as 300 apoptotic cells and measures apoptosis events considerably downstream of the early presence of annexin V. As found in the annexin V FACS analysis (Figure 6B), the treatment of RCC-53 with L-957 alone induces a slight increase in apoptosis (Figure 6C). However, treatment of RCC-53 cells with TIMP-1-GPI dramatically increases the sensitivity of the cells to anti-FAS induced apoptosis in a dose-dependent manner. The results show that the treatment of cancer cells with TIMP anchored to GPI effects death of cancer cells by apoptosis induced by FAS. Therefore, TIMP anchored to GPI is useful as an antitumor agent. Example 7 Treatment with TIMP-1-GPI of RCC reduces BCL-2 and increases expression of BAX protein. BCL-2 proteins represent a family of proteins involved in the control of apoptosis (summarized in Igney &Krammer, 2002) . Some members of this family (such as BCL-2 and BCL-XL) are anti-apoptotic, while others (such as Bad or BAX) are pro-apoptotic. The sensitivity of cells to apoptotic stimuli may depend on the balance between members of the BCL-2 pro and anti-apoptotic family (Igney &Krammer, 2002). The effect of treatment with TIMP-1-GPI on the expression of BCL-2 and BAX is then determined. After a 24 hour preincubation with 700 ng / ml of TIMP-1-GPI or control rhTIMP-1, the cells are stimulated with 1 lig / ml of L-957 (or control mAB) for an additional 16 hours. The level of BCL-2 and BAX protein is then determined using intracellular FACS and Western blot. In all three cell lines, treatment with TIMP-1-GPI increases the expression of pro-apoptotic BAX, and decreases the expression of anti-apoptotic BCL-2. A similar pattern is observed in Western blot assays (Figure 7A, B and C).

Example 8 Incorporation of exogenously added TIMP-1-mucin-GPI on the surface of RCC-53 The TIMPl-mucin-GPI protein is generated, and isolated as described in example 1. The incorporation of the GPI-TIMP-1 protein Purified on the surface membranes of the RCC cell lines RCC-53, RCC-26 or A498 is demonstrated by incubation of the cell lines with 700 ng / ml purified TIMP-1-mucin-GPI or control protein (rh) TIMP- Human recombinant for 1 hour. The superficially associated TIMP-1-mucin-GPI protein is then detected using FACS. The construction of TIMP-1-mucin-GPI is efficiently incorporated into the surface membrane and effectively promotes anti-tumor activity. Example 9 TIMP-GPI for the treatment of residual cancer in an individual TIMP-1-GPI or TIMP-mucin-GPI reagent is applied at 1 μg / ml locally in the excision area after surgical tumor excision. An inoperable tumor, glioblastoma (grade IV WHO astrocytoma) is surgically removed and reagents are installed before wound closure.

Example 10 TIMP-GPI for the treatment of residual cancer in an individual An amount of 1 μg / ml of TIMP-1-GPI or TIMP-mucin-GPI reagent is locally applied in the excision area after the surgical tumor excision. A tumor, advanced stage breast cancer, is removed surgically and reagents are installed before wound closure, particularly if there is a clinical risk of local relapse. Example 11 Evaluation of TIMP-GPI in models of tumor metastasis The effect of TIMP-1-GPI on tumor metastasis is evaluated. Using a murine model, a T cell lymphoma is pretreated that efficiently metastasizes to the liver with either TIMP-1-GPI or rh-TIMP-1 control protein. The resulting tumor is then administered via the vein of the tail, and the distribution of the tumor in the liver is determined three and seven days later. The results demonstrate that cells treated with TIMP-1-GPI show a significantly reduced level of micrometastases relative to control cells treated with TIMP. Example 12 Matrigel invasion assays The effect of TIMP-1-GPI on the lines is tested tumor cells of Example 11 in a series of Matrigel experiments. The results confirm that TIMP-1-GPI has a profound effect on the ability of cells of the T cell tumor line to undergo Matrigel invasion in relation to rhTIMP-1. Example 13 Effect of TIMP fusion constructs on the fibronectin production of fibroblasts Confluent fibroblasts are cultured in the presence or absence of rhTIMP-1 and TIMP-1-GPI. The expressed and secreted fibronectin is quantified by Western blot analysis using anti-fibronectin antibodies (β-actin serves as a control). Figure 9 represents the results of this experiment and clearly shows that rhTIMP-1 (at 700 ng / ml) does not lead to any significant decrease in fibronectin expression, whereas TIMP-1-GPI (in 700 ng / ml) strongly reduces the fibronectin that is secreted by fibroblasts. Additionally, the fibronectin RNA transcribed by the fibroblasts is evaluated by Northern blot analysis. The fibroblasts are cultured in the presence (Figure 10A) or absence (Figure 10B) of 10 ng / ml of the fibroblast that activates TNF-a together with 350 ng / ml TIMP-1-GPI or 700 ng / ml TIMP-1- GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. Additionally, either 0%, 1%, 5%, or 10% of FCS are present in the culture medium. These results clearly demonstrate that at a concentration of 350 ng / ml of TIMP-1-GPI, the transcribed fibronectin RNA is significantly reduced, regardless of whether TNF-a is present in the medium or not, and independently of the FCS content of the medium. In 700 ng / ml of TIMP-1-GPI, fibronectin RNA is negligibly detectable. In this way, the TIMP-GPI fusion construct efficiently inhibits both fibronectin synthesis and secretion of fibroblast growth factor. Example 14 Effect of TIMP fusion constructs on IL-6 production of fibroblasts The Northern blot analysis of Example 13 is repeated using a probe for IL-6 RNA. In this way, the fibroblasts are cultured in the presence (Figure HA) or absence (Figure 11B) of 10 ng / ml of TNF-a that activates fibroblasts together with 350 ng / ml of TIMP-1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPU and rhTIMP-1-GPI, respectively. Additionally, either 0%, 1%, 5% or 10% of FCS are present in the culture medium (Figures 11A-11B). The results clearly demonstrate that already at a concentration of 350 ng / ml of TIMP-1-GPI, the transcribed IL-6 RNA is strongly reduced - regardless of whether TNF-a is present in the medium, and independently of the FCS content of the medium. In this way, the TIMP-GPI fusion construct efficiently inhibits the synthesis and secretion of yet another important cytosine involved in wound healing, ie IL-6, in fibroblasts. Example 15 Effect of TIMP fusion constructs on collagen production by fibroblasts Northern blot analysis performed in the Examples 13 and 14 is repeated using probes for Collagen 1A1 (Figure 12A and 12B), collagen 4a2 (Figure 12C and 12D), and collagen 16A1 (Figure 12E and 12F), respectively. In this way, the fibroblasts are cultured in the presence (Figures 12A, 12C, 12E) or absence (Figure 12B, 12D, 12F) of 10 ng / ml of TNF-a that activates fibroblasts together with 350 ng / ml of TIMP-1. -GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. Additionally, either 0%, 1%, 5% or 10% FCS are present in the culture medium. The results clearly show that already at a concentration of 350 ng / ml of TIMP-1-GPI, all three transcribed collagen RNAs are significantly reduced, regardless of whether TNF-a is present in the medium, and independently of the FCS content. medium.

In this way, the TIMP-GPI fusion construct efficiently inhibits the synthesis and secretion of collagen, which is an essential protein in the production of ECM and tissue remodeling. Example 16 Effect of TIMP fusion constructs on the production of TGF-β by fibroblasts The Northern blot analysis of Examples 13-15 is repeated using probes by TGF-β. In this way, the fibroblasts are cultured in the presence (Figure 13A) or absence (Figure 13B) of 10 ng / ml of TNF-a that activates fibroblast together with 350 ng / ml of TIMP-1-GPI or 700 ng / ml of TIMP-1-GPI, denatured TIMP-1-GPI and rhTIMP-1-GPI, respectively. Additionally, either 0%, 1%, 5% or 10% FCS is present in the culture medium. The results clearly demonstrate that already at a concentration of 350 ng &ml of TIMP-1-GPI, almost no TGF-β RNA can be detected can be detected regardless of whether TNF-a is present in the medium or not, and independently of the FCS content of the medium. Therefore, as illustrated by the above Example, the TIMP-GPI fusion construct efficiently inhibits the synthesis and secretion of yet another important cytosine involved in wound healing, ie TGF-β, in fibroblasts.

Example 17 Generation of additional TIMP fusion constructions In addition, the fusion constructs are generated and purified according to the section "Materials and Methods" given below. Specifically, a truncated TIMP-1-GPI fusion construct (SEQ ID NO: 1), a truncated TIMP-1-muc-GPI fusion construct (SEQ ID NO: 2), a TIMP-2-GPI construct (SEQ) ID NO: 3), a TIMP-3-GPI construct and a mutated form of TIMP-3-GPI (SEQ ID NO: 4) and a truncated TIMP-1-fractalkine-GPI fusion construct (SEQ ID NO; 5) are constructed, expressed and purified. The truncated TIMP-1-GPI fusion construct (SEQ ID NO: l) comprises the first 152 amino acids of the human TIMP-1 protein (ie the terminal C-amino acids 126-207 are deleted) fused to a GPI anchor of 36 amino acids in length. The truncated TIMP-1-muc-GPI fusion construct (SEQ ID NO: 2) contains the first 152 amino acids of the human TIMP-1 protein fused to amino acids 256-380 of human CXCR16 (Mucin) fused in addition to an anchor of GPI of 36 amino acids in length. The resulting fusion construct contains 295 amino acids and has a molecular weight of 32,111 kDa. By analogy to the full-length TIMP-1-GPI, the TIMP-2-GPI (SEQ ID NO: 3) and TIMP-3-GPI consist of the human TIMP-2 and TIMP-3 protein, respectively, fused to 36 amino acids of GPI anchor in length. To produce the mutated form of the TIMP-3-GPI fusion construct (SEQ ID No. 4), the GAG binding domain of human TIMP-3, thought to be responsible for association of the protein with the cell surface, is mutated by the following six exchanges: R43A, K45A, K49A, K53A, K65A and K68A. The truncated TIMP-1-fractalkine-GPI fusion construct (SEQ ID NO: 5) contains the N-terminal portion of human TIMP-1 (amino acids 1-152) fused to amino acids 100-342 of the Human CX3CL1, also fused to the GPI anchor of 36 amino acids in length. Detailed examination of the truncated TIMP-1-GPI constructs (SEQ ID NO: 1), the truncated TIMP-1-mucin-GPI (SEQ ID NO: 2) and the truncated TIMP-1-fractalkine-GPI (SEQ ID NO: 5) . a) Incorporation of truncated TIMP-1-GPI, truncated TIMP-1-mucin-GPI, and truncated TIMP-1-fractalkine-GPI on the cell surface. Truncated TIMP-1-GPI, TIMP-1-mucin, is generated and isolated. Truncated GPI and TIMP-1-fractalkine-GPI truncated according to Djafarzadeh et al., 2004. The incorporation of purified fusion constructs in the cell surface membranes of the RCC, RCC-53, RCC-26 or A498 cell lines is demonstrated by incubation of the lines cells for one hour with 700 ng / ml of purified truncated TIMP-1-GPI, truncated TIMP-1-mucin-GPI and truncated TIMP-1-fractalkine-GPI, respectively, and compared with that of the truncated TIMP-1 protein respective TIMP-1 control that lacks the mucin, fractalkine and GPI domains. The surface-associated protein is then detected using the FACS analysis. It is expected that the addition of control TIMP-1 can not lead to any change in the FACS change, however, truncated TIMP-1-GPI, truncated TIMP-1-mucin-GPI and TIMP-1-fractalkine-GPI truncated, respectively, in fact result in a strong surface signal for TIMP-1. To demonstrate that the exogenously added protein is anchored to GPI, the RCC-53 cells are first incubated with truncated TIMP-1-GPI, truncated TIMP-1-mucin-GPI and truncated TIMP-1-fractalkine-GPI, respectively (200 or 700 ng / ml), and then treated with 60 ng / ml of phospholipase C (PLC, for its acronym in English). FACS analysis is expected to show a complete loss of the surface signal of the TIMP-1 cell after digestion with PLC. To measure the integration efficiency of the anchored TIMP constructs, the TIMP-1 constructs released from the membrane are collected in the wash buffer, and quantified using ELISA specific to TIMP-1. It is expected that most of the antigen Starting TIMP-1 can be recovered from the sample of 200 ng / ml. b) Truncated TIMP-1-GPI proteins, truncated TIMP-1-mucin-GPI, and truncated TIMP-1-fractalkine-GPI block the release of proMMP-2 and proMMP-9 from RCC-53 An increased expression of MMP-2 and MMP-9 are typically correlated with a poor RCC prognosis (Hemmerlein et al., 2004). The clear cell carcinoma cell line of stage IV, RCC-53 constitutively secretes both proMMP-2 and proMMP-9 (Figure 2). The effect of increasing the levels of surface T'IMP-1 in the constitutive release of the MMP-2 and MMP-9 proteins is tested using gelatinase zymography assays (Djafarzadeh et al., 2004; Klier et al., 2001) . The rhTIMP-1 protein at 600 or 1200 ng / ml has no effect on the secretion of proMMP-2 or proMMP-0. In contrast, starting at 10 ng / ml, truncated TIMP-1-GPI, truncated TIMP-1-mucin-GPI and truncated TIMP-1-fractalkine-GPI, respectively, is expected to show a concentration-dependent decrease in both proMMP release -2 and proMMP-9 in the growth medium, comparable to the TIMP-1-GPI treatment described in Example 2. c) Truncated TIMP-1-GPI protein, TIMP-1-mucin-truncated GPI, and TIMP Truncated-1-fractalkine-GPI impacts the proliferation of RCC To evaluate the effect of surface modulation of TIMP-1 on the proliferation of RCC, MTT assays are performed. Truncated TIMP-1-GPI proteins, truncated TIMP-1-mucin-GPI and truncated TIMP-1-fractalkine-GPI, respectively, exogenously aggregated, are expected to produce a dose-dependent decrease in proliferation of RCC-53 and A498 in 24, 48 and 74 hours comparable with Example 4, RCC-26 cells are expected to proliferate extremely slowly (48+ hours at speed duplivsfs); the general trend in effect suggests a suppression of proliferation. Example 18 Further evaluation of the additional fusion constructs of Example 17 The experiments of Examples 1-16 are repeated using the fusion constructs of Example 17; similar results are expected. In particular, TIMP-2 fusion constructs inhibit most MMPs (except MMP-9) and preferably inhibit MMP-2. TIMP-3 fusion constructs inhibit MMP-1, -2, -3, -9 and 12, as well as TACE. The mutant of the fusion construct of TIMP-3 (SEQ ID NO: 4) is examined with respect to its integration properties, which additionally shows an improved ability to integrate into the cell membrane (these experiments are performed according to the methods provided by Example 8). Truncated TIMP-1 fusion constructs (SEQ ID N0: 1, 2, 5) exhibit similar biological functions compared to the full-length TIMP-1 fusion constructs, thereby supporting the notion that the N-terminal portion of TIMP is essential for its inhibitory function. The fractalkine fusion construct (SEQ ID NO: 5) further shows a membrane integration capability comparable to that of the mucin fusion construct (SEQ ID NO: 4). Discussion of the Results Modification of the cell surface using GPI-anchored proteins according to the present invention offers several advantages over traditional gene transfer methods. 1) The method is applicable to cells that are difficult to transfect, for example, RCC cells resistant to FAS-mediated apoptosis, but also primary cultures, bone marrow progenitors, and cells of the immune system. 2) The method can be used when only a small number of cells are available or when the cells can not be easily propagated. 3) The cell surface can be modified independently of the cell type. 4) The amount of the protein last exhibited on the cell surface can be precisely controlled. 5) Proteins anchored to multiple GPIs can be incorporated sequentially or simultaneously into the same cells. The human RCC is a progressive tumor with limited therapeutic options due to the tumor resistance to current chemotherapeutic agents and to radiation. Immunotherapy has benefit for some patients suggesting that RCC can be objectified by immune effector mechanisms. Infiltrating tumor lymphocytes such as CD8 + CTL or NK cells are frequently seen in renal cancer tissues and often recognize autologous tumor cells when tested in vitro (summarized in Schendel et al., 1997). Despite these promising observations, tumors generally continue to grow, indicating that RCC may have acquired resistance to cytotoxic mechanisms. For a productive anti-tumor response to occur, the immune system must not only recognize the tumor, but the cancer cells must also be susceptible to the death mechanisms used by the CTL or NK cells. Cancer cells have involved several mechanisms to evade immune defenses including reduced sensitivity to apoptosis (reviewed in Dunn et al., 2004). CTL and NK cells kill their target cells by perforin / granzyme-dependent or FAS / FASL-dependent apoptosis (Kagi et al., 1994). The relative importance of activities lytic mediated by granular exocytosis against FAS / FASL for tumor control in vivo is controversial. While many studies indicate a dominance of the granular exocytosis mechanism, other studies using perforin agonist mice (Seki et al., 2002) suggest that FAS-dependent apoptosis may constitute a more prominent pathway in vivo. The majority of tumor cells, including RCC, are intrinsically resistant to death mediated by FAS (Frost et al., 2003). The use of TIMP anchored to GPI represents a promising therapeutic procedure to bring the tumor cells to be susceptible to FAS-mediated apoptosis. It has been shown in the present invention that the modification of the cells by exogenous addition of GPI-TIMP-1 can produce n increased and as well as novel biological activities of TIMP-1. TIMP-1-GPI administered exogenously becomes efficiently inserted into the RCC cell membranes and induces a variety of biological effects in the RCC lines with potential therapeutic relevance. Additionally, the TIMP-1 protein anchored to GPI dramatically alters the cell surface association of various MMPs expressed by RCC. This is reflected by a reduced secretion of MMP, including proMMP-2 and proMMP-9, from the RCC cells. While TIMP-1 will block the Enzymatic activity of MMP-2, is not intended to bind to the proforma of the enzyme. Even the data which demonstrate a secretion blockade of proMMP-2 after treatment with TIMP-1-GPI are suggestive of this action. It appears that the addition of an anchor of GPI to TIMP-1 leads to an altered surface stoichiometry that has increased the ability of the TIMP-1 protein to bind MMP. This apparent increased binding of TIMP-1-GPI is also demonstrated with membrane type MMPs. While not much is known about the association of TIMP-1 with MMP-15, it can not be predicted that the binding of TIMP-1-GPI protein to MMP-15 will occur based on the additional deficient avidity of TIMP-1. native to this protein (Lang et al., 2004). The mutational analysis of TIMP-1 critical circuits for MMP-15 binding shows that small, seemingly insignificant changes in TIMP-1 can dramatically change their inhibitory / binding characteristics. In this case, the altered stoichiometry of TIMP-1 on the cell surface seems to have been sufficient to change its binding to MMP-16. Sequestration of MMP on the cell surface is also associated with a reduced ability of the RCC-53 cell line to undergo ECM invasion. As demonstrated in the present invention, treatment with TIMP-1-GPI leads to a pronounced dose-dependent reduction in proliferation of the RCC lines.

Perhaps more significantly, RCC lines normally resistant to FAS apoptosis are made susceptible to FAS / CD95 mediated death after treatment with the TIMP-GPI protein of the invention. However, the agent does not affect the perforin trajectory sensitivity. This suggests that TIMPI anchored to GPI mediates its anti-tumor effect by the path of apoptosis induced by FAS rather than by perforin / granzyme-mediated death by CTL / NK cells. The trajectory of apoptosis by FAS is regulated by caspase activation, whereas membrane damage by CTL / NK using granular exocytosis, as measured by the chromium release assay, occurs independently of caspases (Sayers et al. , 1998; Seki et al., 2002; Trapani et al., 2000). The upstream events that lead to the activation of caspase involve the balance between BCL-2-pro- and anti-apoptotic family proteins. As demonstrated in the present invention, treatment with GPI-TIMP-1 results in down-regulation of the anti-apoptotic BCL-2 protein and a corresponding increase in the pro-apoptotic BAX protein. This shift towards a higher concentration of pro-apoptotic proteins may be a reason for the increased sensitivity of FAS-mediated apoptosis of RCC cells surface modified by TIMP-1. These observations represent a novel action for TIMP-1. The actions similar are also shown for TIMP-3 and other TIMP (TIMP-2 and TIMP-4) (data not shown). It was found that overexpression of TIMP-1, -2 or -3 in vascular smooth muscle cells using adenoviral vectors inhibits their migration through the basement membrane model. Overexpression of TIMP-1 has no effect on cell proliferation, whereas TIMP-2 causes a dose-dependent inhibition of cell proliferation. The overexpression of TIMP-3 also causes a dose-dependent inhibition of proliferation and also leads to apoptosis through mitochondrial membrane depolarization and cytochrome-C leachate (Baker et al., 1999; Baker et al., 1998). Smith et al., 1997). TIMP-3 is only the TIMP protein that selectively binds to the surface of independent cells in association with other surface proteins (Majid et al., 2002; Smith et al., 1997). Focusing TIMP-1 on cell surfaces by anchoring to GPI leads to biological actions that seem to mimic the effects reported for TIMP-3. TIMP-3 has been shown to sensitize melanoma cells to apoptosis induced by anti-FAS antibody, TNF-alpha and TRAIL. The mechanism of action is linked to a general stabilization of FAS, TNF-RI and TRAIL-RI on the surface of melanoma cells treated with TIMP-3 (Ahonen et al., 2003). This increased surface expression of receptors are linked to the activation of caspase-8 and caspase-3 (Ahonen et al., 2003). In the experiments detailed in the present invention, RCC cells do not show a change in surface expression to FAS after treatment with TIMP-1-GPI (Figure 6A). In addition, RCC cells remain resistant to apoptosis induced by TNF-α (tested from 100 to 10,000 units per ml) independently of treatment with TIMP-GPI (data not shown). FACS analysis of the RCC cells subsequently show insignificantly perceptible levels of TNF-RI (p55) and TNF-RII (p75) in RCC-53 and no expression in the RCC-26 or A498 lines (data not shown). Surface expression does not change with treatment with TIMP-1-GPI. In this way, the increased sensitivity to FAS-mediated apoptosis after treatment with TIMP-1-GPI does not appear to be mediated through the general stabilization of cell surface death receptor proteins. What is clear is that treatment with TIMP-1-GPI does not alter the balance of Bcl-2 proteins to produce a more "pro-apoptotic" expression profile. The results of the present invention provide an additional link between tumor biology, MMP / TIMP function and apoptosis trajectories. Linking TIMP proteins to GPI directly or through domains of mucin represent a powerful anti-tumor agent to carry tumor cells, which are normally resistant against FAS-induced apoptosis, to be sensitive to FAS-induced apoptosis. By this mechanism the tumor cells will effectively die. Regardless of the use of the fusion constructs - in particular, the construction of TIMP-1-GPI, the construction of TIMP-1-muc-GPI and those as indicated in SEQ ID N0: 1, 2, 3, 4 and 5 - in the field of regenerative medicine, for example wound healing, the TIMP-GPI constructs of the present invention have been shown herein to efficiently inhibit the production and secretion of important enzymes and cytosines (fibronectin, collagen, IL-6, TGF-β) imply in the processes of tissue remodeling and fibrosis which leads to an increased production of ECM. In this way, members of the TIMP family (TIMP-1, TIMP-2, TIMP-3, TIMP-4), if they are anchored, in the cell membrane by means of a GPI or mucin or fractalkine anchor and a GPI can be used to efficiently modulate wound healing processes by influencing the delicate balance between ECM production and ECM change. Materials and Methods Cell lines and cell culture The RCC, RCC-53 and RCC-25 lines are generated by D.J.S. (Munich, Germany) from the patient's samples. RCC-53, RCC-26 and A498 (American Type Culture Collection) (Girad et al., 1973) are cultured in RPMI1650 medium (GIBCO BRL, Life Technologies, GMBH, Eggenstein, Germany) supplemented with 2 mM L-glutamine (Biochrom KG, Berlin), 1 mM sodium pyruvate (GIBCO BRL, Life Technologies GmbH, Eggenstein, Germany), 12% heat-inactivated FCS (Biochrom KG, Berlin, No. SOI 15). The fresh medium is given every third day and the cultures are divided when the cells are confluent. Cytotoxic effector cells: JB4 is a cytotoxic T effector clone HLA-A2-alloreactive generated in the facilities of the present (E.N) and is expanded by stimulation biweekly as described (Milani et al., 2005). It is used in cytotoxicity assays on day 7 or 8 after stimulation. The human NK leukemic lines, NKL (Robertson et al., 1996) and NK-92 (Gong et al., 1994), are kindly provided by CS Falk (GSF-Institute of Molecular Immunology, Munich, Germany) and are grown in medium which contains 15% FCS inactivated with heat and 100 U / ml of recombinant IL-2. One day before use in cytotoxicity assays, the culture is adjusted to 0.3 x 106 cells / ml in fresh medium. Classification analysis of fluorescence activated cells (FACS) The cells are separated with 1.5 mM EDTA (Biochrom, A, Berlin, Germany No. L2113) in 1 x PBS and incubated for 60 minutes on ice with human-specific antibodies: TIMP-1 (IM32L), MMP-1 (IM35L-100), MMP-3 ( IM36L-100), MMP-8 (IM38L) (CALBIOCHEM, Merck Darmstadt, Germany); MMP-9 (IM 61-100), MMP-2 (IM 51L) (ONCOGENE, Bad, Soden, Germany); MMP-7 (MAB907), MMP-12 (MAB917), MMP-14 (MAB9181) (R &D Systems, Minneapolis, USA); MMP-13 (IM44L), MMP-15 (IM48L), MMP.16 (IM50L) (CALBIOCHEM, Merck Darmstadt, Germany) and IgGl? (SIGMA-ALDRICH, Taufkirchen, Germany no.

M9269). Antibodies ICAM-1 and HLA (W6 / 32 and HB82) are previously described (Johnson et al., 1998; Barnstable et al., 1978; Parham &Brodsky, 1981). Anti-FAS antibodies (H. Engelmann, unpublished data), anti-TNF-RI, anti-TNF-RII and isotype control are used as described (Bigda et al., 1994). The cells are washed three times with 1 x PBS, incubated with mAB anti mouse monkey conjugated to FTIC (DAKO A / S, Glostrup, Denmark No. F0313) for 45 minutes on ice, then washed three times with 1 x PBS and analyzed using a flow cytometer (FACSCalibur, Becton, Dickinson and Company, San Jose, CA, USA) and CellQuest software. Anti-BCL-2 (ALX-804-225) and anti-BAX (ANC-357-040) antibodies are obtained from ALEXIS (Grünberg, Germany). Purification of TIMP-1-GPI protein TIMP-1-GPI protein is produced and purified as previously described (Djafarzadeh, et al., 2004). Briefly, human TIMP-1 is cloned from cDNA using primers specific for hTIMP-1, fused without a translation stop codon to the GPI signal sequence cloned from LFA-3 (Kirby et al., 1995; et al., 1996) and subcloned in pEF-DHFR and stably introduced into Chinese hamster ovary (CHO) cells deficient in DHFR and selected as described (Mack et al., 1995) . The TIMP-1-GPI fusion protein is purified from CHO cells by Triton X-100 detergent extraction followed by column purification using DEAE, heparin sepharose and size exclusion (Djafarzadeh et al., 2004). TIMP-1 ELISA ELISA specific to human TIMP-1 is used which uses the protocol applied according to the manufacturing directions (MAB970, R &D Systems) to monitor the levels of TIMP-1 in solution. Anti-human coating TIMP-1 (MAB970), mAB detection of anti-human TIMP-1 (BAF970) and rhTIMP-1 protein (970-TM) are purchased from R &D Systems GmbH (Wiesbaden, Germany). Incorporation of TIMP-1-GPI into cell membranes RCC-53 cells (5-10 x 106 cells / ml) are incubated with 200 to 700 ng / ml of purified hTIMP-1-GPI at 37 ° C / 5% C02 . The cells are then washed three times with cold PBS and analyzed by FACS using monoclonal antibodies specific to human TIMP-1 (see above). GPI anchoring cleavage by phospholipase C Cells are incubated (5-10 x 10 6 cells / Mi) with 200 or 700 ng of the protein TIMP-1-GPI or rhTIMP-1 in serum-free medium for 1 hour at 37 ° C for 5% C02. The cells are washed three times with cold PBS and treated with 60 ng / ml phospholipase C specific for phosphatidylinositol (SIGMA-ALDRICH, Taufkirchen, Germany No. 661-9) in serum-free medium for 30 minutes at 37 ° C / C02 at 5%. The cells are washed three times, all the supernatants are harvested. Proliferation RCC-53, A498 or RCC-26 cells (30 x 103/100 μl medium) are cultured in 96-well microtiter plates for 24 hours under standard conditions to produce tightly bound cells that grow stably. After discharging the supernatants, 50 μl of medium containing TIMP-1-GPI, buffer, or rhTIMP-1 is added to the cells and incubated for 24 to 72 hours. Then 50μl of a 1mg / ml solution of (3,5-dimethylthiazol-2-yl) -2,5-diphenyl-tetrazolium bromide) MTt (SIGMA-ALDRICH, Taufkirchen, Germany No. M2128) is added. After 3 hours of incubation at 37 ° C, the formazan crystals are dissolved by the addition of 100 μl of isopropanol and 0.04 N HCl. absorbance at 550 nm using the GENios plus reader TEC AN ELISA. For each experiment at least 6 wells are analyzed for experimental condition and time point. Zimography RCC-53 cells are grown in a 24-well plate (5 x 10 4 cells / well). The medium is exchanged every 24 hours with serum-free medium which contains either rhTIMP-1 or increased amounts of TIMP-1-GPI and is incubated for 24 hours, 48 hours and 72 hours. The cellular supernatants are analyzed by gelatin zymography using gels of 10% SDS-polyacrylamide (Invitrogen, Groningen, The Netherlands, No. EC61755BOX) as described (Djafarzadeh et al., 2004). The recombinant MMP-9 enzyme (Amersham Biosciences, Uppsala, Sweden, No. RPN2634) is used as a positive control. Extracellular invasion assay The effect of TIMP-1-GPI treatment against rhTIMP-1 on the ability of cells to invade the ECM is evaluated using a commercial cell invasion assay (Chemicon International, Inc., Temecula, CA, No. ECM 555). The RCC-53 cells are first analyzed for their ability to invade ECM. The cells invaded at the bottom of the insert are separated, used and detected by the CyQuant dye as described in the attached protocol. Increased levels of vascular endothelial growth factor (VEGF for its acronym in English) 2 ng / ml to 8 ng / ml is used to increase the invasion. The optimal migration is seen in 4 ng / ml of VEGF (data not shown). The effect of the treatment with 350 ng / ml and 700 ng / ml of control rhTIMP-1 or TIMP-1-GPI in migration is then determined. To quantify the potential effects of TIMP agents, the baseline migration of RC-53 cells to 4 ng / ml of VEGF is set to 0. The value for 100% "inhibition" of VEGF-induced migration is set as migration / invasion of RCC-53 cells in the absence of VEGF. The effects resulting from treatment with rhTIMP or TIMP-1-GPI in invasion of RCC-53 are calculated as percentage of change (negative or positive) in relation to the "maximum" value. Detection of Annexin V of apoptosis The detection and quantification of apoptotic cells against necrotic cells at the single cell level is carried out using the annexin-V-FLUOS staining kit (Becton, Dickinson and Company, Heidelberg, Germany, No. 556547). The RCC-53 cells are seeded in 1 x 106 cells / well in 24-well plates and binding is allowed for 1 night. The wells are then rinsed 3 times with lxPBS and 1 ml of serum-free RPMI 1640 medium, followed by 700 ng / ml of TIMP-1-GPI or rhTIMP-1. The cells are incubated for 24 hours at 37 ° C / 5% C02. After 24 hours, 1 μg / ml of mAB L-957 Anti-FAS activator (H. Engellmann, unpublished data) or isotype control are added and the cells are further incubated for 16 hours at 37 ° C / 5% C02. The cells are washed with PBS, pelleted and resuspended in staining solution (Annexin-V-fluorescein labeling and propidium iodide (Pl) in Hepes buffer) for 15 minutes at room temperature. The cells are then analyzed by flow cytometry. A time course study shows that the annexin-V binding in the RCC cells precedes the reactivity of Pl. Measurement of apoptosis by chromatin-specific ELISA The apoptosis is measured using the ELISA Plus cell death detection equipment from Roche (Pensberg, Germany, No. 1774425). The RCC-53 cells are seeded in a 96-well disk at a concentration of 4 x 10 4 cells / well and allowed to bind overnight. The wells are rinsed 3 times with 1 x PBS and 200 μl of serum-free RPMI 1640 medium is added to each well, followed by 700 ng / ml of TIMP-1-GPI or rhTIMP-1. The cells are then incubated for 24 hours at 37 ° C / 5% C02. After incubation for 24 hours with TIMP-1-GPI or rhTIMP-1, 1 μg / ml of mAB L-957 anti-Gas activator or mAB isotype control is added and incubated for 16 hours at 37 ° C / C02 at 5%. Then the plate is centrifuged and the supernatant The cell pellet is placed in 200 ml of lysis buffer provided by the manufacturer for 30 minutes and centrifuged. The aliquots of the supernatant (20 μl) are used in ELISA with anti-DNA and anti-histone antibodies to detect the presence of cytoplasmic nucleosomes. Western blot Western blot is used for the detection of MMP in serum free growth medium. The anti-MMP antibodies used are described above (FACS analysis). Western blotting standards of recombinant human MMP are purchased from R &D Systems (Minneapolis, USA) and included; MMP-1 (WBC024), MMP-2 (WBC025), MMP-3 (WBC015), MMP-8 (WBC017), MMP-12 (WBC019) and MMP-13 (WBC020). Western blot is also used for the detection of BCL-2, BAX (see above for mAbs) and β-actin (Acris Hiddenhausen, Germany, No. ab8227). All proteins are detected using a commercial Western blot analysis kit, Chemiluminiscent Immunodetection System (Invitrogen, Groningen, The Netherlands). Cell mediated cytotoxicity Target cells are labeled with Cr51 for 1-2 hours, washed and co-incubated with effector cells at a constant cell number of 2000 cells per well in 96-V bottom plates. Measurements in duplicate of the four Stage titers of effector cells are used in all experiments. Spontaneous and maximum releases are determined by incubating the target cells alone and by labeled cells counted directly, respectively. After 4 hours of incubation at 37 ° C, in a humidified 5% C02 atmosphere, the supernatants are harvested, transferred to Lumaplate solid scintillation microplates, dried overnight and counted in a microplate scintillation counter. TopCount (Packard, Meriden, CT). For each E: T ratio, the percentage of lysis is calculated as follows:% specific lysis = (experimental cpm-spontaneous cpm / maximum cpm, spontaneous cpm) x 100. Spontaneous release of objective cells is always < 15% of the total maximum release. References: Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) J. Mol. Biol 215: 403-410. Ahonen, M., Poukkula, M., Baker, A.H., Kashiwagi, M., Nagase, H., Eriksson, J.E. & Kahari, V.M. (2003). Oncogene, 22.2121-34. Baker, A.H., Gorge, S.J., Zaltsman, A.B., Murphy, G. & Newby, A. C. (1999). BrJCancer, 79,1347-55. Baker, A.H., Zaltsman, A.B., George, S.J. & Newby, A. C. (1998). JClin Invest, 101.1478-87. Barnstable, C.J., Jones, E.A. & Crumpton, M.J. (1978). BrMedBull, 34.241-6. Bigda, J., Beletsky, I., Brakebusch, C, Varfolomeev, Y., Engelmann, H., Holtmann, H, & Wallach, D. (1994). JExpMed, 180.445-60. Bloomston, M., Shafii, A., Zervos, E.E. & Rosemurgy, A. S. (2002). JSurgRes, 102.39-44. Bode, W. & Maskos, K. (2003). BiolChem, 384,863-72. Bond, M., Murphy, G., Bennett, M.R., Newby, A.C. & Baker, A. H. (2002). JBiol Chem, 277, 13787-95. Brand, K. (2002). Curr Gene Ther, 2,255-71. Brew, K., Dinakarpandian, D. & Nagase, H. (2000). Biochim Biophys Acta, 1477, 267-83. Brown, O., Cowen, R.L., Preston, C.M., Castro, M.G .: & Lowenstein, P.R. (2000). Gene Ther, 7.1197-53. Djafarzadeh, R., Mojaat, A., Vicente, A.B., von Luttichau, I. & Nelson, P.J. (2004). Bio Chem, 385, 655-63. Dunn, G.p., Oíd, L.J. & Schreiber, R.D © 2004). Annu Rev Immunol, 22,329-60. Egeblad, M. & Werb, Z. (2002). Na t Rev Cancer, 2,161-74. Frost, P., Caliliw, R., Belldegrun, A. & Bonavida, B. (2003). IntJ Oncol, 22,431-7. Giard, D.J., Aaronson, A.A., Todaro, G.J., Arnstein, P., Kersey, J.H., Dosik, H. & Parks, W.P. (1973). JNa tl Cancerlnst, 51, 1417-23.

Gong, J.H., Maki, G. & Klingemann. H.G. (1994). Leukemia, 8,652-8. Hemmerlein, B., Johanns, U., Halbfass, J., Bottcher, T., Heuser, M., Radzun, H.J. & Thelen, P. (204), In tJ Oncol, 24, 1069-76. Igney, F.H. & Krammer, P.H. (2002). Na t Rev Cancer, 2,277-88. Itoh, Y. & Nagase, H. (2002). Essays Biochem, 38.21-36. Johnson, J.P., Stade, B.G., Hupke, U., Holzmann, B. & Riethmuller, G. (1988). Immunobiology, 78,275-84. Kagi, D., Vignaux, F., Ledermann, B., Burki, K., Depraetere, V., Nagata, S., Hengartner, H. & Golstein, P. (1994). Science, 265,528-30. Kirby, A.C., Hill, V., Olsen, I. & Porter, S.R. (1995). Biochem Biophys Res Commun, 214,200-5. Klier, C.M., Nelson, E.L., Cohen, C.D., Horuk, R., Schlondorff, D. & Nelson, P.J. (2001). Biol Chem, 382, 1405-10. Lang, R., Braun, M., Sounni, N.E. Noel, aA. , Frankenne, F., Foidart, J.M., Bode, W. & Maskos, K. (2004). JMolBiol, 336.213-25. Maskos, K. And Bode W. (2003). Mol Biotechnol, 25: 241-266. Mack, M., Riethmuller, G. & Kufer, P. (1995). Proc Na tlAcadSci USA, 92,7021-5.

Majid, M.A :, Smith, V.A., Easty, D.L., Baker, A.H. & Newby, A. C. (2002). BrJOphtha lmol, 86,97-101. Mannello F., Gazzanelli, G. (2001), Apoptosis, 6 (6): 479-82. Medof, M.E., Nagarajan, S. & Tykocinski, M.L. (nineteen ninety six) . FASeb J, 10,574-86. Milani, V., Frankenberger, B., Heinz, O., Brandl, A., Ruhland, S., Issels, R.d .: & Noessner, E. (2005). Int Immunol. Moniaux N., Adr ianifahanana M., Brand R.E., Batra S.K. (2004), British Journal of Cancer, 91.1633-1638. Nagase, H. & Woessner, J.F., Jr. (1999). JBiolChem, 274.21491-4. Nagata, S. ( 1999 ) . Annu Rev Genet, 33.29-55.

Parham, P. & Brodsky, F.M. (1981) . Hum Immunol, 3,277-99. Rigg, A.S. & Lemoine, N.R. (2001). Cancer Gene Ther, 8,869-78. Pearson WR & Lipman DJ. (1988). Proc Nati Acad Sci USA 85: 2444-2448. Robertson, M.J., Cochran, K.J., Cameron, C., Le, J.M., Tantravahi, R. & Ritz, J. (1996), Exp Hematol, 24,406-15. Sayers, T.J., Brooks, A.D., Lee, J.K., Fenton, R.G., Komschlies, K.L., Wigginton, J.M., Winkler-Pickett, R & Wiltrout, R.H. (1998). J Immunol, 161, 3957-65. Schendel, D.J., Frankenberger, B., Jantzer, P., Cayeux, S., Noessner, E., Willimsky, G., Maget, B., Pohla, H. & Blankens t ein, T. (2000). Gene Ther, 1, 2007-14. Schendel, D.J., Gansbacher, B., Oberneder, R., Kriegmair, M., Hofstetter, A., Riethmuller, G. & Segurado, O.G. (1993). J Immunol, 151.4209-20. Schendel, D.J., Oberneder, R., Falk, C.S :, Jantzer, P., Kres sens t ein, S., Maget, B., Hofstetter, A., Riethmuller, G. & Noessner, E. (1997). JMolMed, 75, 400-13. Seki, N., Brooks, A.D., Carte, C.R., Back, T.C., Parsoneault, E.M., Smith, M.J., Wiltrout, R.H. & Sayers, T.J. (2002). J Immunol, 168.3484-92. Singer A. and Clark R. (1999). N. Engl. J. Med., 341, 738-746. Smith, M.R., Kung, H., Durum, S.K., Colburn, N.H. & Sun, Y. (1997). Cytokine, 9,770-80. Span, P.N., Lindberg, R.L., Manders, P., T jan-Hei jen, V.C., Heuvel, J.J., Beex, L.V. & Sweep, C.G. (2004). JPathol, 202,395-402. Trapani, J.A., Davis, J., Sutton, V.R. & Smyth, M.J. (2000). Curr Opin Immunol, 12,323-9.

Vogelzang, N.J. & Stadler, W.M. (1998).

Lancet, 352, 1691-6. Zacchigna, S., Zentilin, L., Morini, M., Dell'Eva, R., Noonan, D.M., Albini, A. & Giacca, M. (2004). Cancer Gene Ther, 11.73-80. It is noted that in relation to this date, the best method known to the applicant to carry out the aforementioned invention, is that which is clear from the present description of the invention.

Claims (30)

1. CLAIMS Having described the invention as above, the content of the following claims is claimed as property: 1. A fusion construct (TIMP-mucin-GPI), characterized in that it comprises an amino acid sequence of a metalloproteinase (TIMP) tissue inhibitor, or a biologically active fragment thereof, wherein the TIMP or biologically active fragment thereof is linked to a mucin domain followed by a glycosylphosphatidylinositol (GPI) anchor.
2. 2. The fusion construct according to claim 1, characterized in that the mucin domain is a membrane-bound mucin domain.
3. 3. The fusion construct according to claim 2, characterized in that the mucin domain comprises an amino acid sequence encoded by a gene selected from the group which consists of MUC1, MUC3A, MUC3B, MUC4, MUC11, MUC12, MUC16 and MUC17 , or a portion thereof.
4. 4. The fusion construct according to claim 1, characterized in that the mucin domain comprises the mucin stalk of the CXCL16 chemokine or fractalkine (CX3CL1) associated superficially.
5. 5. The fusion construction in accordance with any of claims 1-4, characterized in that the TIMP is selected from the group which consists of TIMP-1, TIMP-2, TIMP-3 or TIMP-4.
6. 6. The fusion construct according to any of claims 1-5, characterized in that the TIMP is human TIMP-1.
7. The fusion construct according to any of claims 1-6, characterized in that the fusion construct contains one or more GPI signal sequences to direct the GPI anchor.
8. 8. The fusion construct according to any of claims 1-7, characterized in that the GPI anchor is derived from antigen associated with lymphocyte function (LFA-3) or a portion thereof.
9. 9. The fusion construct according to any of claims 1-8, characterized in that the TIMP-mucin-GPI construct is inserted into the cell membranes of tumor cells.
10. 10. A nucleic acid molecule, characterized in that it comprises a nucleic acid sequence encoding the fusion construct according to any of claims 1-9.
11. 11. An expression plasmid, characterized in that it comprises the nucleic acid molecule according to claim 10 and additional expression elements.
12. 12. A host cell, characterized in that it comprises the expression plasmid according to claim 11.
13. 13. A vector, characterized in that it comprises the nucleic acid molecule according to claim 10.
14. 14. A pharmaceutical composition, characterized in that it comprises the construction of fusion according to any of claims 1-9 or the nucleic acid molecule according to claim 10, and a pharmaceutically acceptable carrier.
15. 15. The use of a fusion construct according to any of claims 1-9 for the preparation of a medicament for the treatment of cancer.
16. 16. The use according to claim 15, wherein the cancer is selected from the group which consists of breast cancer, renal cancer, prostate cancer, leukemias, seminomas, melanomas, teratomas, lymphomas, neuroblastomas, gliomas, rectal cancer , endometrial cancer, kidney cancer, adrenal cancer, thyroid cancer, blood cancer, skin cancer, brain cancer, cervical cancer, intestinal cancer, liver cancer, colon cancer, stomach cancer, intestinal cancer, gastrointestinal cancer, cancer of the lymph node, cancer of the esophagus, colorectal cancer, cancer of the pancreas, cancer of the ear, nose and throat (ENT), cancer of the uterus, ovarian cancer and lung cancer and their metastasis.
17. 17. Use according to claim 15 or 16, where cancer is residual cancer after surgical removal.
18. 18. The use according to claim 17, wherein the fusion construct is administered locally as an anti-tumor adjuvant for the treatment of residual cancer in patients with breast cancer and patients with glioblastoma (astroctoma VI).
19. 19. The use according to claim 17 or 18, wherein the fusion construct is administered in a concentration of 0.5 to 5 μg / ml, preferably 1 μg / ml.
20. 20. The use according to any of claims 15-19, wherein the fusion construct is administered by spray in the wound and / or injection in regions that are not available for surgery.
21. The use of any of claims 15-20, wherein the fusion construct is a fusion construct (TIMP-GPI) that does not include the mucin domain.
22. 22. An in vitro method for the inhibition of cancer cell proliferation, characterized in that it comprises the step of subjecting a cancer cell line to an effective amount of fusion construct of TIMP- mucin-GPI or TIMP-GPI.
23. 23. The method according to claim 22, characterized in that the cell line is a renal cell carcinoma cell line (RCC).
24. 24. The use of TIMP-mucin-GPI or TIMP-mucin-GPI to bring tumor cell lines resistant to apoptosis by FAS to be sensitive to FAS-induced apoptosis.
25. 25. A method for treating a skin lesion to prevent or inhibit the formation of a scarring, characterized in that it comprises administering the pharmaceutical composition according to claim 14 or a pharmaceutical composition which comprises the fusion construct in accordance with any of claims 1-9, wherein the fusion construct is a fusion construct (TIMP-GPI) that does not include the mucin domain, and a pharmaceutically acceptable carrier to the site of the skin lesion, or scarring.
26. 26. The method according to claim 25, characterized in that the healing is a hypertrophic scarring or a keloid formation.
27. 27. The method according to claims 25 and 26, characterized in that the pharmaceutical composition is used together with a detergent, sealant or carrier substance.
28. 28. The use of a fusion construct according to any of claims 1-9 or according to claims 1-9, wherein the fusion construct is a fusion construct (TIMP-GPI) that does not include the mucin domain for the preparation of a medicament for the treatment of a lesion of the skin of a subject to prevent or inhibit the formation of a scar.
29. 29. The use according to claim 28, wherein the healing is a hypertrophic scarring or a keloid formation.
30. 30. The use according to claims 28 and 29, wherein the pharmaceutical composition is used together with detergent, sealant or carrier substance.