JPH08504577A - Compositions and methods for modifying the regulatory activity of TGF-β - Google Patents

Abstract

  (57) [Summary] The present invention provides novel purified TGF-β binding glycoproteins, endoglins, and isolated nucleic acid molecules encoding amino acid sequences corresponding to TGF-β binding glycoproteins. Also provided are soluble endoglin-derived polypeptides and fragments thereof. A pharmaceutical composition comprising a purified or recombinantly produced endoglin-derived polypeptide and a pharmaceutically acceptable carrier, as well as a patient by administering the pharmaceutical composition of the invention to the patient. Further provided is a method of treating.

Description

Detailed Description of the Invention             Compositions and methods for modifying the regulatory activity of TGF-β                             Field of the invention    The present invention provides a method for modifying the biological activity of cell biology and cell regulators. Concerning the law. More specifically, the present invention relates to novel TGF-β binding glycoproteins .   Throughout this application various publications are shown in parentheses. Opening of these publications The drawings are for the purpose of more complete description of the technical contents related to the present invention. Are incorporated herein by reference.                             Background of the Invention   Post-translational processing involves the covalent attachment of one or more carbohydrate units to a protein. The combined glycoproteins are widely distributed. Several types including immunoglobulins Secreted proteins are glycoproteins and are large plasma membrane receptors such as cell membrane receptors. Being a component of the moiety, the glycoprotein may be involved in cell-cell attachment.   Transforming growth factor-β (TGF-β) is a variable form of many cell types. It is a family of multi-functional cell regulators that are produced. ,J. Cell Biol.105: 1039 (1987)). 5 different TGF-β variants The isoform was identified. TGF-β1 and TGF-β2 are well characterized It was TGF-β is incorporated herein by reference, It is the subject of U.S. Pat. Nos. 4,863,899, 4,816,561 and 4,742,003. T GF-β binds to cell surface receptors on various cell types and Dependently increases or suppresses the response of most cells to other growth factors Is known to do. TGF-β also regulates the differentiation of certain cell types, Promotes or suppresses cell proliferation. Another significant effect of TGF-β is the extracellular matrix. Promotion of cellular production of lix proteins and their receptors (see For Keski-Oja et al.J. Cell Biochem.33:95 (1987); Massague,Cell 49: 4 37 (1987); Roberts and Sporn, "Peptides Growth Factors and Their Recep. tors ", Springer-Verlag (1989)).   TGF-β regulatory activity despite beneficial and essential cell regulatory functions Can be harmful to the host organism. For example, mesenchymal cell growth and proliferation Is promoted by TGF-β, but using TGF-β as an autocrine growth factor Species tumor cells can also be promoted. In other cases, TGF-β induced cell proliferation Inhibition is likewise harmful to the host organism. An example is repairing tissue damage It is the prevention of new cell growth to help. Of extracellular matrix by TGF-β Stimulation of production is necessary for wound healing. However, in some cases the TGF-β response is regulated Instead, an excessive accumulation of extracellular matrix occurs. Excessive accumulation of extracellular matrix An example of a product is the “internal” scarring and cutaneous scar tissue formation that occurs in glomerulonephritis lesions. Ah It   Transforming growth factor-β receptor in most mesenchymal and epithelial cells The tar system consists of several components (Massague,J. Ann. Rev. Cell Biol.6: 597 (1 990); Lin, H .. Y. ,Cell 68: 775 (1992); Georgi, L .; L. ,Cell 61:63 5 (1990); Mathews. L.S. et al.Cell 65: 973 (1991); Attisano, L .; ,Cell  68 :: 97 (1992); Lopez-Casillas et al.Cell 67: 785 (1991) and Wang et al.,Ce ll  67: 796 (1991)), one of which is a β-glycan, a membrane anchor proteoglyca It is In addition to β-glycans, most mesenchymal and epithelial cell TGF-β receptors The pter system is a 53-kDa glycoprotein of type I, whose structure has not yet been determined. It consists of scepter and type II receptor, and is a protein serine / threonine It belongs to the Nase receptor family. More limited distribution of other cell surfaces TGF-β binding proteins have also been described.   Therefore, there is a need for the development of compounds that can modify the effects of cell regulators such as TGF-β. There is sex. The present invention meets these needs and provides related benefits as well.                             Summary of the invention   The present invention provides a novel purified TGF-β binding glycoprotein. With that protein An endoglin is expressed at high levels on human vascular endothelial cells .   Polypeptide derived from purified endoglin having the ability to bind TGF-β to TGF-β Code or any fragment thereof Contact with an effective amount of TGF-β regulates the pathological conditions mediated by Methods for treating are also provided by the present invention. Therefore, intact Natural endoglins and their soluble fragments are found in these methods. It is for. The present invention prepares full-length soluble endoglin-derived polypeptides. And a method of purification. Novel TGF-β binding glycoprotein and soluble ene An isolated nucleic acid encoding a dogrin-derived polypeptide also contains the nucleic acid. And a recombinant host cell transformed with such a vector It will be provided.                         Brief description of the drawings   FIG. 1 shows the domain structure of β-glycan and endoglin. 853 amino acids Linear sequences of β-glycans, which are transmembrane proteoglycans of A disulfide-bonded tolan composed of two identical subunits of 633 amino acids Look for similarities between the linear sequences of the membrane protein endoglin. FIG. 4 is a schematic diagram showing a region of a circle. End-Grin transmembrane area And short cytoplasmic regions (black squares) show a high degree of sequence similarity with the corresponding region of β-glycans. Have. The extracellular domains (ectodomains) of these proteins have a low degree of similarity. Two areas are identified (white squares). The numbers are β-glycan and endoglin Shows the percent amino acid sequence similarity between the indicated domains of. The ellipse is the system The position of the in residue is shown. Connection of glycosaminoglycan chains in β-glycans Two putative sites are shown.   FIG. 2 shows cell surface TGF-β1 binding protein expressed in HUVEC. HUV Cultures near EC confluence at 100 pM¹²⁵Incubation with I-TGF-β1 By subsequent chemical cross-linking with 0.16 mM disuccinimidyl suberate. And affinity labeled. A) Affinity labeled HUVEC Triton X-100 extract , SDS-PAGE gels under reducing (R) or non-reducing (NR) conditions. Leh C was extracted from affinity-labeled cells in the presence of excess unlabeled TGF-β1. It contains a good thing. The migration positions of TFG-β receptors I (RI) and II (RII) are shown. The 180-kDa arrowhead and higher molecular weight major affinity-labeled proteins are non-reducing Appeared on the original gel. The arrowhead indicates the 110-120 kDa affini found on the reducing gel. It is a T-labeled protein. B) Affinity labeled HUVEC detergent extract, First separated on gel under non-reducing conditions, then incorporated by reference herein. Cheifetz and Massague,J. Biol. Chem.266: 20767-20772 (1991) , And separation under reducing conditions in two dimensions. Move off the 110-120-kDa diagonal Marked species are shown (arrowheads).   FIG. 3 shows the specific immunoprecipitation of TGF-β1-Endoglin complex. HUVEC is a figure 100pM as described in 2.¹²⁵Affinity-labeled with I-TGF-β1. A) Incubate a detergent extract of affinity-labeled cells with mAb 44G4 for immunization. The complex was recovered with Protein G-Sepharose. After washing, sample Equivalent portion of SDS-PAGE under reducing (R) or non-reducing (NR) conditions (5-8% gradient polyacrylamide gel). B) Affinity labeled HUVEC lysis The material was incubated with 100 μl of 44G4-IgG-Sephaorse at 4 ° C for two consecutive 45 minute incubations. To maximally deplete endoglin. S) Second immunoprecipitation Supernatant after falling. I) First round 44G4 immunoprecipitation with 83% endoglin. T) Depletion experiment Amount corresponding to the total extract used for. All samples were run under reducing conditions I_ROther than Were analyzed under non-reducing conditions on SDS-PAGE. TGF-β receptor II (RII), and And the migration positions of endoglin monomer, dimer and oligomer.   FIG. 4 shows that endoglin transiently expressed in COS-M6 cells binds to TGF-β1. Indicates that. COS-M6 cells have a full-length L-endoglin-encoding cDNA (Endogl in) or control vector (C). 150 cells pM¹²⁵I-TGF-β1 was affinity-labeled and the detergent extract was labeled with mAb 44G4 and then with And incubated with protein G-sepharose. Immunoprecipitated protein Quality was analyzed by SDS-PAGE under reducing (R) and non-reducing (NR) conditions and Visualized by geography.   Figure 5: COS cell transfectants and TGF-β isomers evaluated by HUVEC. Shows the specificity of endoglin for. A) Transfer with Endoglin vector The infected COS-M6 cells to 150 pM¹²⁵Affinity labeling with I-TGF-β1 only Or Alternatively, in the presence of 1 or 10 nM unlabeled TGF-β1, -β2 or -β3 Niti labeled. B) HUVEC, 100pM¹²⁵Affinity labeling with I-TGF-β1 only Or in the presence of 5 nM unlabeled TGF-β1 or TGF-β2 Labeled. Lysates of these cells were immunoprecipitated with MAb 44G4. Immunoprecipitate , SDS-PAGE gel was fractionated under reducing conditions. Contains Monomer Endoglin The area of the gel is shown with the migration position of the 100-kDa marker.   FIG. 6 is a vector pcNeoSolE having a 1.7 kb endoglin cDNA insert. The restriction enzyme map of ND is shown.   FIG. 7 shows partial nucleotide and deduced amino acid sequences of isolated S-Endoglin cDNA. Is shown. Nucleotides are numbered on the right. Amino acids are numbered on the left There is. Putative signal sequence (nucleotides 283-357), and transmembrane Region (nucleotides 2042-2116) is shown in bold. 135 bp insert fragment (bottom Lined) are donor / acceptor sites (GT, AG at positions 2134 and 2267). Splicing consensus sequence of, and lariat structural branch points ( It contains CTGAC at position 2234). Nucleotides 372-2021 and 2322-3073 are endothelium It was found to be identical to the corresponding cDNA sequence of ndugrin (Gougos, A. and A., et al. And Letarte, M. J.,J. Biol. Chem.265: 8361 (1990), and Table 1).   FIG. 8 shows a cytoplasmic region showing the presence of two different forms of endoglin. Show analysis. A. Of two selectable forms Schematic representation of the cDNA encoding the cytoplasmic region of endoglin. 3'with cytoplasmic domain Only the regions corresponding to the ends are shown. Isolated endoglin cDNA (S-endoglin ) Is a 135 bp insertion that is not present in the previously described sequence (L-Endoglin). Containing the input fragment (Gougos, A. and Letarte, M. J.,J. Biol. Chem . 265: 8361 (1990)). The sequence of the additional 135 bp insert fragment is shown in FIG. Is shown in. Position of stop codon and corresponding translated protein sequence (thick line) Is shown. B. L m. Biophys. Res. Commun.189: 356 (1992), and the S-endoglycan The cytoplasmic domain of transmembrane and transmembrane domain. Inside the enclosure, Contains the same array. Numbers indicate the position of the first amino acid in the entire sequence. Match 2 Two major areas were identified. The first region (73% concordance) is S- and L-endoglyme. Residues 587-617 of β-glycan and residues 780-810 of β-glycan. Second area (74% one Contains 634-660 residues of L-endoglin and 823-849 residues of β-glycan. The transmembrane region of S-Endoglin is shown in bold. Dash is for alignment purposes Inserted for. The asterisk indicates the last residue in the protein.   FIG. 9 shows the expression of L-endoglin and S-endoglin in transfected cells. Shows the present. Mouse fibroblasts with L-Endoglin or S-Endoglin cDNA Transfected and analyzed for expression of the endoglin molecule. A. Cell surface Exist in Analysis of existing endoglin by cytefluorometry . After trypsinization, cells were treated with monoclonal antibody 8E for indirect immunofluorescence. Stained with 11 (anti-endoglin). Endoglin-mock transfectant Control staining of the components is also shown. B. Immunoprecipitation analysis. Cell³⁵S] with methionine Metabolically labeled, dissolved, 44G4 (anti-endoglin) or HCl / l (anti-CDllc) mono Immunoprecipitated with Kroll antibody. Monoclonal antibody HCl / l, negative control Was included as. Load the sample onto a 6-12% gradient acrylamide gel under non-reducing conditions. Electrophoresed on a glass plate. C. Immunoblotting analysis. Pseudo (L cell) cDNA, L- Trans with Endoglin (L-Endo) cDNA and S-Endoglin (S-Endo) cDNA Lysis of transfected mouse fibroblasts and PMA-treated U937 cells with Triton X-100 Thawed and centrifuged to remove insoluble material. Protein contained in the supernatant Were electrophoresed on a 6% acrylamide gel under non-reducing conditions and subjected to nitrocellulose. It was transferred to a membrane. Endoglin immunodetection by chemiluminescence assay, 44G4 (anti- Endoglin) Monoclonal antibody. HCl / l (anti-CD11c) and X6 3 Monoclonal antibody was used as a negative control. D. Cell surface labeling Immunoprecipitation analysis of dogurin. Fake, L-Endoulin and S-Endoglin (S-Endo) transfected mouse fibroblasts,¹²⁵I labeled, lysed, 44G4 Immunoprecipitation with a monoclonal antibody (anti-endoglin). Reduce the sample (R) Alternatively, under non-reducing (NR) conditions, 10% acrylamide gel Electrophoresed on a glass plate.   Figure 10: Detection of L-Endoglin and S-Endoglin transcripts by PCR amplification. Is shown. Placenta, total RNA samples from PMA-treated HL-60 cells or PMA-treated U937 cells, Incubation was performed in the presence (+ RT) or absence (-RT) of reverse transcriptase. Got The prepared cDNA sample was used for S-Endoglin (S-Endo) or L-Endoglin (L-Endoglin) of pUC13. -Endo) clones and cDNA from an endothelial cell library along with S -In the presence of Endoglin-specific oligonucleotides # 14 and # 15 (panel A) , Or oligonucleotide # 12 common to L-Endoglin and S-Endoglin And was used for PCR amplification in the presence of # 11 (panel B). When the RT reaction was omitted No amplification was observed, eliminating the possibility of DNA contaminating the RNA sample. L-En Other bands below the specifically amplified fragment of dogrin and S-endoglin Indicates a primer-dimer artifact.   FIG. 11 shows binding of both S-Endoglin and L-Endoglin to TGF-β1. You S-Endoglin (L + -S) and L-Endoglin (L + -L) transfectants Confluent cultures of 100 pM¹²⁵Incubate with I-TGF-β1 only Or in the presence of 4 nM unlabeled TGF-β1 and then Affinity labeling was performed by chemically cross-linking with succinimidyl suberate. cooling( cold) by the 4-fold ratio between cpm binding in the absence and presence of competing ligands As demonstrated, all samples specifically bind to TGF-β1; parental L cells and mock Transfectants average 60,000 ratio cpm, while endoglin transfectants The components bound an average of 110,000 cpm. 44G4-IgG Sepharose L + -S trans Effectant immunoprecipitates were paired with control IgG-Sepharose at 700 cpm. And contained an average of 7700 cpm; L + -L transfectant immunoprecipitates It contained 4300 cpm for 630 cpm of Troll. These immunoprecipitates are reduced (R ) Or under non-reducing (NR) conditions on a 6-9% gradient acrylamide gel. Endoglin monomer, dimer and oligomer positions, and molecular weight markers Indicates the position of   FIG. 12 shows the multiple cloning site of pcDNAI / Neo.   FIG. 13 shows the amino acid sequence and nucleotide sequence of “L-Endoglin”. .                         Detailed Description of the Invention   Endoglin is a homozygous, composed of disulfide-linked subunits. It is a dimeric membrane glycoprotein. Endoglin derived from human live human cDNA At least two isomers expressed from rally, an estimated "90 kDa" L-isomer " It has been shown that there is a smaller or smaller "S-isomer" of approximately 85 kDa. From the organization Purified human endoglin was linked to two disulfide bonds of approximately 95 kDa each. It was shown to consist of subunits. This is a human pre-erythroblast, macro Expressed on phage and on leukocytes of lymphoid and myeloid lineage and on vascular endothelial cells It is expressed at higher levels. It is also a vegetative cell syncytial layer, with mother blood It plays an important role in configuring the interface and providing fetal nutrient exchange and immune defense. It is abundant in the multinucleated placental layer (Gougos et al.,Inter. Immunol.4: 83-92 (199 2)).   Endoglin was first identified in the pre-B leukemia cell line by virtue of its reactivity with mAb 44G4. Identified (Quackenbush and Letarte,J. Immunol.134: 1276-1285 (1985) ). It is due to childhood acute leukemia (ALL) with pre-B lymphoid and bone marrow phenotypes Present in low levels in native cells; not in T-ALL (Gougos and Letarte,J. Immunol. 141: 1925-1933 (1988b); Kreindler et al.,Leukemia and Lymphoma 3: 7-18 (1990)). Interestingly, the human endoglin gene is found on chromosome 9q34-qte. Region located at r and translocated to chromosome 22 of Philadelphia chromosome-positive leukemia Within (Fernandez-Ruiz et al.,Cytogen. Cell Gen.64: 204-207 (1993 )). In normal adult bone marrow, only 3-5% of mononuclear cells express endoglin, and They have a pro-erythroblast phenotype; pre-B and myeloid progenitors have detectable levels. Bell endglin not shown (Buhring et al.Leukemia 5: 841-847 (1991)). Normal B and T lymphocytes and unstimulated mononuclear cells do not express endoglin, , Upregulation is observed in activated macrophages (Lastres et al.,Eur. J. Immun ol. 22: 393-397 (1992)).   Endoglin expression is associated with various pathologies known to cause endothelial cell proliferation. Skin damage endothelium significantly increased (Westphal et al.,J. Invest. Dermatol.100: 2 7-34 (1993)). In tumors, capillary endothelial cells undergoing active angiogenesis also It shows higher endoglin levels than the resting endothelium of adjacent tissues.   There are various isomeric forms of purified human endoglin with Mr = 95,000, 90,000. It is composed of two 85,000 disulfide-bonded subunits, N- and And O-linked oligosaccharides. The primary sequence of human endoglin is a 25 amino acid sig Null sequence, extracellular domain of about 561 amino acids, one transduction of 25 amino acids It consists of a Bran region and a cytoplasmic tail of 14 to 47 amino acid residues (Goug os and Letarte,J. Biol. Chem.265: 8361-8364 (1990)).   The relationship between human endoglin and the TGF-β receptor system was demonstrated by the rat TGF-β binding process. Roteoglycan, β-glycan (Type III TGF- (Also known as the receptor) , Transmembrane and relatively short of this protein (43 amino acids ) The cytoplasmic tail is very similar to the corresponding region of endoglin (71% amino acid sequence). Column similarity and 63% amino acid identity) (see Figure 1). These two The extracellular domain of the protein exhibits homology limited to primary structure and It is not a proteoglycan, but contains N- and O-linked oligosaccharides.   Cloning of TGF-β receptor II (Lin et al.,Cell 68: 775-785 (1992))¹²⁵ When bound to I-TGF-β and chemically crosslinked, it previously had a polypeptide size of 80 kd. Functional transmembrane serine / threonine kinase identified as (Cheifetz et al.,J. Biol. Chem.265: 20533-20538 (19 Is associated with the type I receptor (53kd protein), and TGF-β signa Convert receptor; receptor I requires receptor II to bind TGF-β However, we believe that receptor II requires receptor I to mediate signals. (Laiho et al.,J. Biol. Chem.266: 9108-9112 (1991); Laiho et al.J. B iol. Chem. 265: 18518-18524 (1990); Wrana et al.,Cell 71: 1003-1014 (1992)) . Recently cloned receptor I, plus serine / threonine kinase The transfection experiment according to the above also further revealed that the receptor for binding TGF-β was used. Ta -Supports the notion that I must be linked to receptor II (Ebner et al.Sc ience  260: 1344-1348 (1993)).   The present invention provides human polypeptides derived from soluble endoglin that bind to TGF-β. To serve. A full-length soluble endoglin-derived polypeptide is produced during processing. Cleavage signal sequence, 561 amino acid cell of mature endoglin polypeptide It has an ectodomain, an inner membrane protein, which totals approximately 600 or 633 amino acids. Consisting of no acid. Nucleic acid sequences encoding human endoglin polypeptides are shown in Table 1 And identified in FIG. Encodes a polypeptide derived from soluble endoglin Nucleic acid sequences to It is included in the sequence shown in FIG. Further, according to the present invention, the inserted end piece is A vector having a genomic DNA molecule encoding phosphorus is provided. This vector Is based on the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the purposes of Patent Procedure. October 21, 1993, Coleccion Espanola de Cultivos Tipo (CECT), 461. 00 Burjasot (Valencia), Spain, deposited with CECT 4475. Therefore, the end Isolated genomic DNA encoding gulin is within the scope of the present invention.   The invention also provides a purified, encoding two isomers that differ from each other in the cytoplasmic domain. A human endoglin polypeptide is provided. "S-Endoglin" is a cytoplasmic region It has 14 amino acid residues in the L-endoglin and contains 47 amino acids in the cytoplasmic region. No acid residue. Extracellular regions in mature endoglin and 586 amino acids that align the transmembrane region are in agreement.   As used herein, the term "purified" refers to a molecule or compound That the material is substantially free of contaminants normally associated with native or natural environments. means. For example, mature human protein can be obtained from many methods. Membrane tan Methods available for protein purification include sedimentation, gel filtration, ion exchange, reverse phase And affinity chromatography. Another well-known method is Deut scher et al.Guide to Protein Purification: Methods in Enzymology Volume 182 (A Academic Press 1990), which is incorporated herein by reference. ing. Alternatively, the purified polypeptide of the present invention can also be isolated, for example, from Sambrook et al.Mo lecular Cloning: A Laboratory Manual Second Edition (Cold Spring Harbor Laborator y 1989) and can also be obtained by well known recombinant methods, which It is incorporated by reference in the detailed text. Preparation of soluble endoglin-derived polypeptide Examples of this method for producing are bacteria, yeast, by methods well known in the art. Or in a suitable host cell, such as a mammalian cell, encodes soluble endoglin The nucleic acid is expressed and expressed again using methods well known in the art. To recover the soluble protein that was recovered.   For purposes of illustration, soluble endoglin expression was completely expressed from pcEXV-L ENDO. A 2.3 kb long endoglin cDNA fragment was cut out and Hind III linker was added. And then Hind I After digestion with II, it was cloned into pBluescript vector (Stratagene) Subcloning into theth Digested with 111I and achieved. (This Enzyme cleaves the coding region approximately 80 bp upstream of the transmembrane region. . )   Synthetic complementary oligomer with in-frame stop codon and Bam HI overhang And ligated to the linearized plasmid. By Hind III and Bam HI Digestion and purification of the 1.7 kb fragment, pcDNAI / Neo (Invitrogen, San D iego, FIG. 12). After cloning, put the clone in the correct orientation It was then screened by restriction enzyme analysis and found to be in the correct orientation. Confirmation of cDNA inserts is performed using T7 and SP6 sequencing primers. , Confirmed the presence of start and stop codons at the 5'and 3'ends of the insert . 5'end primer 3'end primer   CHO-kl cells (ATCC) were transfected with pCDNA / Neo to express the construct. To produce stable transfectants that are resistant to G418. The appropriate conditions were determined. 5 (5) x 10⁶Cells were placed in a 960 uF condenser (Biorad Ge ne Puls e) with a voltage of 300 V and a time constant of 17.9 msec. Clotroporation to ensure maximum transfer of stable cells with minimal dead cells I got a cuntant.   The flow chart below illustrates the expression of nucleic acids encoding soluble endoglin. The method for   Soluble polypeptides and biologically active fragments thereof may also be chemically combined. Can be generated by Synthetic polypeptides are described in Applied Biosystems, Inc. Model  430A or 431A automated polypeptide synthesizer, and chemistry provided by the manufacturer It can be manufactured using chemicals. Soluble polypeptides are also described in detail below. Can be isolated directly from cells transformed with the expression vector.   As used herein, endoglin-derived polypeptides are shown in Table 1. Substantially identical to the 633 amino acid sequence or the 658 amino acid sequence shown in Figure 13. A human polypeptide having the amino acid sequence of means. As used herein, the term "polypeptide derived from soluble endoglin Is a soluble human endoglin expressed by the extracellular domain of this nucleic acid. A biologically active fragment of a polypeptide. For use herein The term "active fragment" or "biologically active fragment" refers to TGF-β Refers to any portion of the endoglin polypeptide that binds. Polypeptide is TGF-β A method for determining whether or not a protein can bind to, for example, as provided herein. It is well known to those skilled in the art.   The present invention also produces the same phenotypic effects, albeit different from the nucleic acid molecules shown in Table 1. Nucleic acid molecules, such as the sequence nucleic acid molecules shown in FIG. These modifications A nucleic acid molecule that has, but is phenotypically equivalent, is referred to as an "equivalent nucleic acid." The present invention , The phenotype of the polypeptide produced therefrom when compared to the nucleic acid molecule above Includes nucleic acid molecules characterized by changes in non-coding regions that do not modify It The present invention further hybridizes to a nucleic acid molecule of the present invention or its complement. Nucleic acid molecule. The term "nucleic acid" as used herein refers to mRNA and cR NA, and single- and double-stranded genomic DNA, DNA, and cDNA. Further, the term "polypeptide" as used herein refers to S-endoglin and Any naturally occurring allelic variant, such as L-endoglin in Includes engineered forms.   The present invention provides an isolated nucleic acid molecule encoding a polypeptide derived from soluble endoglin. I will provide a. The term "isolated nucleic acid molecule" as used herein does not occur in nature By a form is meant a nucleic acid molecule. One means of isolating human endoglin nucleic acid is , Methods well known in the art (Gougos, A. et al.,J. Biol Chem.265: 8361 (1990 )) And the methods of the Examples provided below, Is an artificially engineered antibody to probe a human cDNA expression library That is. The DNA and cDNA molecules encoding the human endoglin polypeptide are Genomic DNA, cDNA, or R from a human, mammalian, or other animal source. It can be used to obtain NA. Isolated genomic DNA is also according to the invention as described above. Included. It is a nucleic acid sequence of the invention and is described in Sambrook et al. (Supra). Isolated by using methods well known to those of skill in the art.   The present invention further enables the transcription of RNA transcription promoters and other regulatory sequences. A linked isolated nucleic acid molecule is provided. As used herein, the term "operably linked" "Ligated" is placed so that the promoter can direct the transcription of RNA from the nucleic acid molecule. Means that Examples of such promoters are SP6, T4, and T7. Is. The promoter and the DNA fragment inserted into it A vector that has both a cloning site operably linked to the Well known in the art. Preferably, these vectors are in vitro or RNA can be transcribed in vivo. An example of such a vector is the pGEM series (Prome ga Biotec, Madison, WI).   The present invention provides DNA, cDNA encoding a polypeptide derived from soluble endoglin. , Or an expression vector or replication vector containing the isolated nucleic acid molecule of the invention, such as RNA. To provide the data. Examples of vectors are bacteriophage, baculovirus, And viruses such as retroviruses, cosmids, plasmids (such as pcEXV-2 ,) And other recombinant vectors. Nucleic acid molecules are well known in the art. It is inserted into the vector genome by the method. For example, both insert DNA and vector DNA Can be exposed to restriction enzymes, producing ends on both molecules whose base pairs are complementary to each other. Then they are ligated with a ligase. Alternatively, the synthetic nucleic acid linker is It can be ligated to the insert DNA corresponding to the restriction sites in the vector DNA and then to the specific It is digested with a restriction enzyme that recognizes the otide sequence. In addition, the termination codon And oligonucleotides containing the appropriate restriction sites include, for example, some of the following: Can be ligated for insertion into a vector containing: stable mammalian cells Or for the selection of transient transfectants, use the neomycin gene. Such selectable marker genes; flanking human CMV for high level transcription (i enhancer / promoter sequence derived from the early gene; SV40-derived transcription termination and RNA processing signals for SV40 polyoma origin of replication and ColE1 for somal replication; a versatile multiclot Cloning sites; and T7 for in vitro transcription of sense and antisense RNA. And SP6 RNA promoter. Other means are available.   Encodes a polypeptide derived from endoglin, or a soluble fragment thereof In bacterial cells, yeast cells, mammalian cells and other animal cells containing Vectors compatible with expression are also provided. The vector also includes soluble end Arranged to allow the expression of Glyin polypeptide-encoding DNA Required for expression of DNA in bacterial, yeast, mammalian, or animal cells. It may contain the necessary control elements. The regulatory elements required for expression bind to RNA polymerase Contains a promoter sequence for transcription and a transcription initiation sequence for ribosome binding . For example, bacterial expression vectors include promoters such as the 1ac promoter, and And the Shine-Dalgarno sequence for initiation of transcription and the initiation codon AUG (Sambrook et al., , 1989) Mu. Similarly, eukaryotic expression vectors may be heterologous or homologous for RNA polymerase II. Promoter, downstream polyadenylation signal, initiation codon AUG, and ribosomal Includes a stop codon for desorption. Such vectors are commercially available, or Methods well known in the art, such as those generally described above for constructing vectors. Can be assembled with the sequences described in. The expression vector is the above-mentioned polypeptide. It is useful for generating cells that express the gene.   The present invention provides endoglin-derived polypeptides or soluble fragments thereof. Provided is a mammalian cell containing a coding cDNA molecule. For example, in mammalian cells There are mammalian cells containing a plasmid compatible with the expression of This plasmid A cDNA molecule encoding a polypeptide derived from dogulin and the expression of this polypeptide. It contains the necessary control factors. Various mammalian cells can be used as the host, for example , Mouse fibroblasts NIH3T3, CHO cells, HeLa cells, Ltk-cells and the like. In the above Expression vectors as described include the calcium phosphate precipitation method, DEAE dextran. Like method, electroporation, or microinjection, Used to transfect mammalian cells by methods well known in the art Can be done.   The present invention provides a pharmaceutical carrier, a purified polypeptide and a purified soluble polypeptide. Tide, active fragments thereof, or purified mature protein and its activity Fragment Izu There is provided a pharmaceutical composition comprising them alone or in combination with each other. these The polypeptide or protein may be recombinantly derived, chemically synthesized, Or it can be purified from natural sources. As used herein, the term "pharmaceutically acceptable. An acceptable carrier is phosphate buffered saline, water, and oil / water or water / oil solution. Emulsions such as mulsion, as well as various types of wetting agents Standard pharmaceutical carriers of. Any of these pharmaceutical compositions are described below. In the method described above, or for the preparation of a medicament for treating the conditions described below. It is useful for   Specific reaction for the TGF-β binding polypeptide derived from endoglin of the present invention Anti-endoglin antibody 44G4 (Quackenbush, EJ, and Letarte, M ． J.,J. Immunol.134: 1276-1285 (1985)), or TGF-β binding Any antibody that has a specific reactivity to the active endoglin polypeptide is also Provided. Active fragments of antibodies are included within the definition of "antibody". The present invention Antibodies and Fragments of Rhodobacter can be produced by any method known in the art. obtain. For example, polyclonal and monoclonal antibodies are well known in the art Can be produced by, for example, these methods are incorporated herein by reference: Done by Harlow and Lane,Antibodies: A Laboratory Manual(Cold Spring H arbor Laboratory 1988). Polypeptides produce such antibodies. Can be used as an immunogen. Chimera, humanized, CDR- Modified antibodies, such as transplanted or bifunctional antibodies, can also be prepared by methods well known to those skilled in the art. Can be generated more. Such antibodies are also described, for example, in Sambrook et al. (Supra). Can be produced by hybridomas, chemical synthesis, or recombinant methods. This These antibodies are the endoglin-derived polypeptides of the present invention or their soluble fragments. Can be used to determine the presence or purification of pigments. Detection of such polypeptides For export, these antibodies are used in vitro diagnostic methods to determine the presence of endoglin. Method or in vivo imaging method.   Useful for in vitro detection of target soluble endoglin-derived polypeptides in samples Immunological techniques include immunoassays using detectable antibodies. like that Immunoassays may be performed, for example, by ELISA, Pandex microfluorescence assay, well known in the art. Assay, agglutination assay, radioimmunoassay, flow cytometry, serodiagnosis assay Includes sei and immunohistochemical staining techniques. Antibodies can be any of a variety known in the art. Can be detected by the means of. For example, the detectable marker can be direct or It may be bound to the antibody indirectly. Useful markers include, for example, radionuclides, enzymes, firefly. Includes photochromophores, chromophores, and chemiluminescent labels.   The present invention provides a method of modifying a biological function mediated by the regulatory activity of TGF-β. Therefore, an appropriate sample containing TGF-β was added to an effective amount of biologically active endoglin. Contact with a native polypeptide, such as soluble endoglin, or a pharmaceutical composition as described above. A method is provided that includes the step of touching.   As used herein, the term "effective amount" refers to a polypeptide that is sufficient to bind TGF-β. And thus interferes with or inhibits its regulatory activity. This way Are particularly useful for altering the regulatory activity of TGF-β1 or TGF-β3. control Examples of activity include stimulation of cell proliferation, cell growth inhibition, extracellular matrix protein Includes, but is not limited to, promotion and regulation of immune function. TGF-β is a monocyte Is known to be a potential chemoattractant for Can induce IL-1, TNF-α, TGF-β, and surface FcγRIII, all of which are flame Included in illness response. In contrast, TGF-β has antimicrobial activity and superoxide activity. Macrophages can be inactivated by inhibiting the development of nions, and macrophages Induces suppression of class II-restricted antigen presentation by phage.   The method of the present invention may be performed in vitro or in vivo. The method of the present invention is When performed in vitro, the sample and the above-mentioned polypeptide, protein, or The contacting step is carried out by incubating with the pharmaceutical composition.   In vitro, the novel nucleic acid molecules and antibodies of the invention can be administered to subjects such as human patients. Useful for detecting and quantifying the amount of TGF-β in samples isolated from test specimens . Detection of TGF-β is associated with diseases associated with overexpression of TGF-β, such as glomerulonephritis. Useful for monitoring progress.   However, in a preferred embodiment, the contacting step is as described above, A subject, eg, a human patient, is administered a polypeptide, protein or pharmaceutical composition. It is administered by in vivo administration.   Methods of administration are well known to those of skill in the art, and are given orally, intravascularly, or It includes, but is not limited to, parenteral administration. The regulatory activity is effectively modified Can be administered at such doses. Administration is such that this amount is effective for its intended purpose. As such, it can be carried out continuously or intermittently.   The present invention also treats pathological conditions caused by TGF-β regulated activity. Which provides TGF-β from purified soluble endoglin. A polypeptide, an active fragment thereof, a polypeptide derived from endoglin, or Contacting with any of the active fragments. TGF-β is Pathological conditions associated with lipopeptides and thereby caused by TGF-β regulatory activity Treat the condition. The term "pathological condition" as used herein refers to TGF-β-induced Resulting from regulatory activity, eg inflammation, rheumatoid internoditis, inflammatory skin damage, scarring For tissue formation, lung fibrosis, liver fibrosis, atherosclerosis, and glomerulonephritis. Say the pathology of the meaning. Mesenchymal cell growth and proliferation is stimulated by TGF-β, but only However, some tumor cells also used TGF-β as an autocrine growth factor Can be stimulated by. Examples of inhibition states include tissue damage for ulcers and immunosuppression. The disruption of new cell growth that aids repair. Extracellular mat with TGF-β Stimulation of lix production is essential for wound repair. But yes In some cases, TGF-β responses were unregulated and excess extracellular matrix Accumulation occurs. An example of excessive accumulation of extracellular matrix is glomerulonephritis. Another An example of pathology is cancer.   In one embodiment, the method of the invention is performed on a subject, eg, a human patient or a mammal. To the effective amount of purified endoglin protein or endoglin-derived soluble po Lipids, or biologically active fragments thereof or pharmaceuticals as described above It is carried out by administering the composition. The method of administration is outlined above.   The present invention also relates to endoglin and a polypepper capable of binding an effective amount of endoglin. Contacting with tide causes the endoglin to bind and thereby endoglin. Disclosed are methods of inhibiting the activity of endoglin, which inhibits the activity of phosphorus. This specification The term "polypeptide capable of binding endoglin" as used in the text Any substance capable of forming a complex with the protein, such as TGF-β1 or TGF-β3, or Means active fragments thereof. The active fragment is the endoglin The amino acid corresponding to the fragment of TGF-β1 or TGF-β3 that retains the ability to bind. It is the sequence of the no acids. How to create such a fragment is Methods of measuring binding activity are well known to those of skill in the art. The present invention also has an end Retains their activity of binding to gulin but of functional ligands to receptors Includes polypeptides with broken bonds that no longer mediate the corresponding biological response It Like this These "mutated" polypeptides are the end-groups of ligands that function normally on cells. By blocking binding to phosphorus, ligand-mediated endogliosis It may act as an antagonist to the biological function of the protein.   The present invention also provides agents for modifying biological functions regulated by TGF-β. Includes the use of a composition as defined above for its preparation. These biological functions are Will be described in detail below.   Modifications that do not substantially affect the activity of the various molecules of the present invention also include the definition of said molecules. It is understood to be included in.   The following examples are intended to be illustrative and not limiting of the invention. A. Human endoglin protein and encodes human endoglin protein Nucleic acid isolation                                Example I                   Cell culture and transfection   Human umbilical vascular endothelial cells (HUVEC, CRL 1730 ATCC) were supplemented according to the supplier's instructions Maintained in α-minimum essential medium as described above or as previously described (Gougos, A. et al. ,J. Immunol.141: 1925 (1988)) prepared from umbilical vessels. From any source Similar results were obtained with cells. Dulbecco's supplemented with 10% bovine serum COS-M6 cells maintained in modified Eagle medium were transformed with the mammalian expression vector pcEXV (Mil ler, J. ,J. Exp. Med． 164: 1478 (1986)) linked to the EcoRI site. Coat End Grin Control vector without cDNA or cDNA insert (pcMV5; Lopez-Cas illas, F. ,Cell 67: 785 (1991)) at the DEAE-dextran-chloroquine method. (Seed, B. et al.PNAS. USA84: 3365 (1987)). . Twenty-four hours after transfection, cells were trypsinized and quantified. Re-inoculate into a raster dish, and as described below¹²⁵Afini with I-TGF-β1 It was grown for an additional 48 hours before being labeled with tee.                                 Example II                 Receptor affinity labeling and immunoprecipitation   TGF-β1 and TGF-β2 were purchased from R & D Systems (Minneapolis.MN). . And TGF-β3 was obtained from Oncogene Science (Manhassett. NY). The present invention Used in the study¹²⁵I-TGF-β1 was prepared as previously described (Cheifetz, S. et al.,J . Biol. Chem. 265: 20533 (1990), prepared by the chloramine-T method, or Or Amersham Corp. Purchased from: both preparations gave identical results.¹²⁵I-TG Cells were treated with F-β1 and disuccinimidyl suberate (Pierce Chemical Co.). Conditions for affinity labeling monolayers have been previously described (Massague, J.,Met hods Enzymol. 146: 174 (1987)). Used in each experiment¹²⁵I-TGF-β1 and competition Concentrations of unlabeled ligands that are combined are indicated in the figure legends. Affinity label Triton X-100 extract of lysed cells was treated with sodium dodecyl sulfate-polyacrylic acid. Analyzed directly on a mid gel (SDS-PAGE) or first paired with human endoglin. Monoclonal antibody (mAb) 44G4 (Quackenbush, EJ, et al.J. Immunol.13 4: 1276 (1985)) or control antibody (see below). For immunoprecipitation, the detergent extract was diluted with phosphate buffer containing an equal volume of 1% Triton X-100. Diluted with saline and prior to overnight incubation at 4 ° C with mAb 44G4 , Protein G-Sepharose (Pharmacia LKB Biotechnology Inc.) and 2 at 4 ℃ Pre-cleared by incubating for 0 minutes. Immune complex Harvested by incubating with G-sepharose for 1 hour at 4 ° C. How many In that experiment, mAb 44G4 was used to bind to Sepharose. Immunoprecipitate Wash 3 times (1% Triton X-100 in saline) and then dithiotre Analyzed by SDS-PAGE in the presence or absence of itol (DTT) and Visualization was carried out by radiography. To monitor the specificity of immunoprecipitation The irrelevant mAb (44D7) used in the control experiment was labeled with any affinity Bands were not immunoprecipitated.                                 Example III                         SDS-PAGE and 2D gel analysis   Analysis of the affinity labeling profile of HUVEC was performed on vascular endothelial cells from other sources. Like cells, show that these cells have little or no β-glycans. However, this is characteristically as a diffusion band between 200 KDa and 400 KDa on reducing SDS-PAGE. Move (Fig. 2A). Instead, HUVEC is a disulfide-bonded cell surface protein. Which expresses TGF-β receptors I and II Together¹²⁵Affinity labeled by cross-linking with I-TGF-β1. Reception Tar I and II as labeled complexes of approximately 65 KDa and 100 KDa in HUVEC, respectively. The size of these labeled receptors detected and reported for other human cell lines Is similar. Relative migration of affinity labeled proteins fractionated on SDS-PAGE The degree of comparison is that HUVEC's major affinity labeling protein is 95-120 KDa on reducing gel. , But on the other hand, on the non-reducing gel the major affinity label was shown. The protein is between 100-110 KDa (presumed to be receptor II) and 180 KDa. It was shown to have migrated to and above (Endoglin) (Fig. 2). This pattern is , Showed the presence of disulfide-bonded TGF-β binding protein.   Analysis of these disulfide-bonded TGF-β1 binding proteins on a two-dimensional gel ( Figure 2B) shows disulfide-bonded complexes (probably dimers and higher oligos). About 95 KDa (reduced 110 KDa affinity labeled complex from cross-linked TGF- Subunit of the value estimated by subtracting 12.5 KDa of β1 monomer amount) It is confirmed to include. Disulfide-bonded TGF-β1 binding with type II receptor Compatible proteins are the major affinity labeled species expressed by HUVEC.                                 Example IV                    Immunoprecipitation with anti-endoglin mAb   The disulfide-bonded TGF-β binding protein on endothelial cells is endoglin. Affinite to determine whether or not -HUVEC extract labeled with a monoclonal antibody specific for human endoglin Antibody (mAb) 44G4 (Georgi, LL et al.,Cel1 61: 635 (1990); MacKay, K .; ,J . Biol. Chem. 266: 9907 (1992); Merwin, J. et al. R. ,Am. J. Pathol.138: 37 (1991)). Electrophoretic analysis of these immunoprecipitates was The subunit structure of the protein complex is similar to that of endoglin (FIG. 3A). Thus, 180K when analyzed under non-reducing conditions Da, and a major of approximately 110 KDa under reducing conditions, migrating as a complex larger than 200 KDa. Affinity labeled band required. Higher-order oligomers themselves It contains multiple endoglin molecules cross-linked by TGF-β1 which is a sulfide bond dimer. I got it. Repeated immunoprecipitation with 44G4-IgG-Sepharose was performed from cell extracts. These labeled species were completely depleted (Figure 3B). Lacking end grin, and this Three other human cell lines (A549, used as negative controls in these experiments) No affinity labeled band was immunoprecipitated from Hep G2, MCF-7). Human Thus, the monoclonal antibody specific to endoglin is It is shown to be the major TGF-β binding protein in human vascular endothelial cells.                                 Example V                   Ectopic expression of endoglin in cells   This dimeric TGF-β binding protein of HUVEC is identical to endoglin That is, full-length enz It was confirmed by ectopically expressing the dogrin cDNA.¹²⁵Afini with I-TGF-β1 After labeling, the labeled species with the characteristics of endoglin are endoglin trans. Only the detergent extract of the fectant could be specifically precipitated with mAb 44G4 (Fig. 4). Differences in glycosylation differ in COS cells against the endogenous endoglin of HUVEC This may explain the smaller size of the expressed endoglin.                     B. Isolation of S- and L-endoglin                                Example VI                 Cloning and sequencing of endoglin cDNA   Approximately 2.5 x 10 from λgt10 cDNA library^FiveClones J.6: 4023 (1987)). (50,000 pfu / 150 mm dish), Screened with a 700 bp PstI fragment from the cDNA (Gougos, A. and And Letarte, M. J.,J. Biol. Chem.265: 8361 (1990)). In the range 2.4-3.1 kb After 3 rounds of plaque purification with size, 13 hybridizing clones Was isolated. The longest clone (clone 3.3; 3073 bp) was subcloned into pUC13. And dideoxy chain termination method (Sanger, F. et al.,PNA S USA  74: 5463 (1977)).   Clone 3.3 in pUC13 was digested with BbrPI and BamHI. Acorn Blunt ends and sucks Expression vector pcEXV (Miller. J. and Germain, R.,J. Exp. Med.164: 1478 (1986)) to generate pcEXV-EndoS. Leader distribution in cDNA (Table I) Lack of columns (Gougos, A. and Letarte, M.J.,J. Biol. Chem.265: 8361 (1990 )) Was overcome by the construction of pcEXV-EndoL. Digest pcEXV-EndoS with Mlul / BamHI And a 563 bp Mlul / BamHI fragment specific for endoglin cDNA (Table I). To obtain pcEXV-EndoL. Transfer transfectants to pcEXV-EndoS or pc Co-transfection of mouse L cells with either EXV-EndoL and psV2neo It was generated by Endoglin positive clones were isolated after G418 selection. same Endoglin negative clones were selected as mock transfectants in the same experiment .                                Example VII           Antibodies, immunofluorescence, immunoprecipitation, and immunoblotting   The endoglin-specific monoclonal antibody used was 8Ell (Lastres , P. ,Eur. J. Immunol.22: 393 (1992)), and IgGl subclass 44G4 ( Gougos, A. And Letarte, M.,J. Immunol． 141: 1925 (1988)). Ko Control monoclonal antibodies are IgGl subclasses HCl / l (anti-CDllc) and X6 Was 3. FCM and immunoprecipitation analysis are described in Lastres, P .; ,Eur. J. Immunol.twenty two: 393 (1992), the disclosure of which is incorporated herein by reference. ing. 5x10 for immunoblotting studies⁶250 μl of cells for 30 minutes at 4 ° C 1% Triton X-100 in PBS, 1 mM PMSF. Insoluble material, 1 Removed by centrifugation at 00,000 xg for 1 hour in a Beckmam TL100 centrifuge. In the supernatant The contained proteins were analyzed by SDS-PAGE under non-reducing conditions using a mini gel system. Separated and transferred to a nitrocellulose membrane (Hybond, Amersham). This membrane First, the blocking solution (10% fetal bovine serum, 0.5% Tween-20, 1 M group in PBS) was used. Course, and 10% glycerol), then monoclonal antibody 44G4 and ink Was added. The presence of endoglin was determined by chemiluminescence assay (ECL detection kit, Ame rsham).                                Example VIII                   Reverse transcription polymerase chain reaction (RT-PCR)   Total cellular RNA was guanidinium isothiocyanate and cesium chloride ultracentrifuged Isolated by a separation step (Chirgwin, JM et al.,Biochemistry 18: 5294 (1979) ). Purify poly (A) + RNA by oligo (dT) affinity chromatography (Maniatis, T. et al., “Molecular Cloning: A Laboratory Manual” Cold S pring Harbor Laboratory, Cold Spring Harbor (1982)). Single-stranded cDNA It was synthesized from (A) + RNA (0.5 μg / 40 μl reaction) by AMV reverse transcriptase. 5 μl c The DNA was used in a 50 μl PCR reaction (Gilliland, G. et al., Innis, MA, Gelfand, D. et al. H., Sninsky, J. J. And White, T .; J. Hen) ("PCR Protocols" Academic P ress, San Diego 1990, p. 60). The oligonucleotide primers used for amplification are , E # 12 (nucleotides 1945-1964), E # 11R (nucleotides 2475-2494 reverse complement ), E # 14 (nucleotides 2136-2158), And E # 15 (reverse complement of nucleotides 2247-2273). Amplification is for each 0.25 μM concentration 0.2 mM each with primers and 0.25 U / 50 μl Taq DNA Polymerase (Promega) Of dATP, dCTP, dGTP, dTTP of 1 × Taq buffer (Promega). amplification Was performed in a thermocycler as follows: 95 ° C for 5 minutes; E # 12-E # 11R oligonuc For leotide pairs, 94 ° C for 45 seconds, 54 ° C for 45 seconds, and 72 ° C for 1/35 size. Cour, or E # 14-E # 15 pair, 94 ° C for 45 seconds, 68 ° C for 45 seconds, and 7 35 cycles of 1 minute at 2 ° C, and then 10 minutes at 72 ° C. Omitting reverse transcriptase The same control reaction was carried out at the same time except for. Amplified product on agarose gel It was analyzed by electrophoresis and ethidium bromide staining.                                 Example IX                        Receptor affinity label   TGF-β1 was purchased from R & D Systems (Minneapolis, MN), and¹²⁵I-TGF-β 1 was obtained from Amersham (Oakville, Canada) with a specific activity of 2000 Ci / mmol. Parent mau L cells, S-endoglin or transfectants expressing L-endoglin Tomato L cells and mock transfectant L cells at confluence ( 5 × 10 per plate⁶Cells) grown and 4 nM competing unlabeled TGF-β1 100 pM for 4 hours in the presence or absence of¹²⁵Incubated with I-TGF-β1 It was Wash cells and wash with 128 mM NaCl, 5 mM KCl, 5 mM MgSO 4._Four, 1.2 mM CaCl₂ Final concentration of 0.16 mM in buffer containing 50 mM Hepes, pH 7.5. Disuccinimidyl suberate (Pierce Chemical Co.) (DSS) at 4 ℃ for 15 minutes Crosslinked The cells were washed 4 times and 1% Triton X-100 and as described above. Una (Massague, J.,Methods Enzymol.146: 174 (1987)) proteolytic in Solubilization was performed directly on Petri dishes containing minimal solubilization buffer with inhibitor. Extraction The material was immunoprecipitated and then SDS-PAGE analyzed.                                   result                 Isolation and characterization of endoglin cDNA clones                                           Identification of cytoplasmic variants   A full-length cDNA clone was prepared from λgt10 library prepared from PMA-treated HL60 cells. It came from Lee. Due to the large size of the insert, clone number 3.3 Selected for characterization. Partial sequence of 3073 bp cDNA insert from clone 3.3 Is shown in FIG. This clone contains 281 bp of 5'untranslated region followed by 1875 bp. Including open reading frame. Deduced protein sequence is expected (Gougos, A. and Letarte, M.J.,J. Biol. Chem.265: 8361 ( 1990)), an endoglin leader peptide containing 25 amino acids (aa), and then cells The transmembrane region spans 561 residues and 25 amino acids in the external portion. Shi However, clone 3.3 has nucleotide (nt) 213 inserted within the known cDNA sequence. It contains a 135 bp segment starting at 4 (Figure 8). The first 21 nucleoti of this insert De is in frame with the preceding sequence and has a new 7-amino acid sequence To code. This sequence is the C-terminal 40 of endoglin. Amino acids, thus replacing the 47 previously reported residues (Gougos, A. and Let. arte, M. J.,J. Biol. Chem.265: 8361 (1990)). No acid has a cytoplasmic tail (Fig. 8), and there are two selectable forms of endoglin. Suggest presence. The predominant form of endoglin in the myelomonocytic cell line HL60 is cytoplasmic It seems to be a form containing 47 residues in the main. Because this sequence was analyzed This is because it was present in 12 out of 13 clones. 14 amino acids and 47 amino acids Cytosolic tail isomers of non-acids and the corresponding cDNAs, Called Phosphorus and L-Endoglin.                   Expression of two different forms of endoglin   To further characterize the two selectable forms, the short form and the endoglin form And an independent construct corresponding to the long form were inserted into the expression vector pcEXV and the Transfected into fibroblasts (Fig. 9). S-End Green and L-End Both of the grins were highly expressed on the cell surface as measured by FCM (Fig. 9). . In addition, immunoprecipitation secondary to metabolic labeling of transfectants is associated with anti-endoglin. 170-kDa (L-endoglin) specifically recognized by monoclonal antibody Showed a protein of 160-kDa (S-endoglin) (Fig. 9). Transfection The distinct size of S- and L-endoglin in the tant also indicates immunity under non-reducing conditions. It was detected by blotting analysis (Fig. 9). In these experiments, L-Endog Phosphorus is the same M as endoglin detected on U937 cells._rAnd this is the front monocyte cell It suggests that it is the predominant form in the system.   Immunoprecipitation after cell surface radiolabeling of transfectants revealed that both isomers Expressed as a disulfide-bonded homodimer (Figure 9) and present in the extracellular portion It is shown that the cysteine residue that acts as an intermediary interchain disulfide bond.             Differential expression of S-Endoglin and L-Endoglin mRNA   Individual expression of S- and L-endoglin was analyzed by RT-PCR. S-Acorn Amplification with two primers from the unique 3'region of the cDNA On transfectants and on PMA-treated monocyte cell lines, endothelial cells, and embryos. It produced the expected 137 bp product on board (Fig. 10), and this isomerism in several cell types. Showed the presence of the body. Use common primers for L-Endoglin and S-Endoglin When used, the 411 bp L-endoglin-specific fragment produced PMA-treated monocyte cells. Of the 546 bp S-endoglobin, which could be amplified on the cell lineage, placental cells, and endothelial cells. Phosphorus Fragment on control S-Endoglin transfectants Only amplified (Figure 10). When PCR is used to amplify a competing template, Templates can suppress amplification of fewer templates (Gilliland, G. et al. ,In:[Innis, M. A., Gelfand, D. H., Sninsky, J. J. And White, T .; J ． (Eds.) "PCR Protocols", Academic Press, San Diego 1990, p. 60). Therefore , These results show that L-Endoglin is the predominant form, and S-Endoglin is It is shown to be expressed at lower levels in these cell types. This is Similar to the endoglin molecule observed on transfectant L-endoglin M_rIs consistent with the presence of the endoglin molecule in PMA-treated U937 cells (Fig. 9), It is estimated that glycosylation levels are similar on both cell types.           Binding of L-Endoglin and S-Endoglin to TGF-β1   Endoglin found to be a component of the receptor system for TGF-β (Cheifetz, S. et al.,J. Biol. Chem.267: 19027 (1992)), It was important to analyze the ability of each isomer to bind this ligand. Short form And transfectants expressing long forms of endoglin and parental L cells and And pseudotransfectants were compared for their ability to bind TGF-β1 . Immunoprecipitation analysis allowed the identification of the TGF-β1-Endoglin complex (FIG. 11). . Under non-reducing conditions, the endoglin dimer is M_rIs 170 kDa (short form) and 1 It was observed as a radiolabeled band of 75 kDa (long form). Endoglin oligomer Is M in both cases_rWas observed to migrate above 270 kDa. Similar Was also previously observed in a cross-linking experiment with human endothelial cells, and TGF-β It can be shown that the endoglin molecule cross-linked by 1 is its own dimer (Roberts, A.B. and Sporns, M.B., Sporn, M.B. and Roberts, A.B. ． Ed., "Peptide growth factors and their receptors" Springer-Verlag, Hei delberg 1990, p. 419). Under reducing conditions, the short form of endoglin is 97 kDa. And the 107 kDa major band, and the long form of endoglin. Bands of 102 and 112 kDa were observed. These dimers are It has been observed on cells, and the TGF-β1 monomer (12.5 kDa) or Could represent endoglin bound to the immer (25 kDa) (Roberts, AB and S. porns, M. B., Sporn, M. B. And Roberts, A. B. Chapter, `` Peptide growth fac tors and their receptors '' Springer-Verlag, Heidelb l. Chem.266: 20767 (1991)). Subtracting the contribution of TGF-β1, S-endoglin For 85 kDa and for L-endoglin 90 kDa. obtain.                                   Consideration   Sequence analysis revealed two distinct sequences named L-Endoglin and S-Endoglin. Indicating the presence of a cDNA variant. These two isomers are the majority of transcripts synthesized. Number clearly corresponds to the L-endoglin isomer, but bone marrow cells, endothelium, and placenta Are co-expressed by The mechanism by which the two isomers occur should be determined. Both The isomers are most likely produced by alternative splicing. fact, Consensus sequence of donor / acceptor sites at the 5'and 3'ends (GT, AG) , And lasso-type bifurcation consensus sequence (CTGAC), S-Endoglin cDN It was found on the novel 135 bp insert of A (Figure 7). "Retained intron" A similar example of a cytoplasmic variant produced by the splicing mechanism of TGF-β receptor Recently reported on the activin receptor of the Ter family (Attisan o, L. ,Cell 68:97 (1992)).   The behavior of the two endoglin isomers was analyzed by transfection studies . Both forms are expressed on the cell surface as disulfide-linked homodimers, It is shown that cysteine residues present in the extracellular region are involved in dimerization. Different Despite their cytoplasmic domains, both forms are TGF-β1 binding proteins. Behave.   S-Endoglin, L-Endoglin, and β-glycans are transmembrane It contains 22 identical residues in the domain and the adjacent cytoplasmic domain (Figure 9). This This is the outline of the first conserved motif between the β-glycan and the two endoglin forms. , Two tyrosine residues and two Ser / Th whose phosphorylation status should be analyzed Contains r residues. The second region of high identity in the cytoplasmic tail is the long form of the enzyme. It is shared only with ndoglin and β-glycans. Ser / Thr residues in this motif The high content (40%) of Tyr and the absence of Tyr caused this region to be secreted by Ser / Thr kinase. Suggest that it may undergo phosphorylation. Interestingly, TGF-β receptor II cells Cytoplasmic domain exhibits Ser / Thr kinase activity (Lin, HY et al., Cell 68: 775 (199 2)).   The present invention will be described with reference to the disclosed embodiments, but various modifications may be made. It should be understood that it can be made without departing from. Therefore, the present invention Is limited only by the appended claims.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁶ Identification code Internal reference number FI A61K 38/00 ABG ABX ACJ ACS ADS C07H 21/02 8615-4C 21/04 B 8615-4C C07K 14/47 8318 -4H C12N 5/10 C12P 21/02 C 9282-4B 9455-4C A61K 37/02 ACJ ABB ABE ABG ADS ABX ACS (81) Designated country EP (AT, BE, CH, DE, DK, ES, FR, GB , GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM, GA, GN, ML, MR, NE, SN, TD, TG), AU, BB, BG, BR, CA, CZ, FI, HU, JP, KP, KR, LK, LV, MG, MN, MW, NO, NZ, PL, RO, RU, SD, SK, UA, US, VN (71) Applicant Consejo Superior de Investigasiones Scientifica Spain 28006 Madrid, Velázquez 144 (72) Inventor Letart, Mitchell Canada M4K2Z7, Ontario, Toronto, Howland Road 35 (72) Inventor Masserg, Joan United States New York 10021, New York, York Avenue 1233 (72) Inventor Berna View, Carmelo Spain 28006 Madrid, Velazquez 144, Centro Investigation Ones Biolohicas (72) Inventor Chefez, Sera Canada L4K2R8, Ontario, Concord, New Seaberry Drive 88

Claims (1)

[Claims]   1. The endoglin polypeptide or biologically active fragment thereof Isolated nucleic acid molecule.   2. A soluble endoglin-derived polypeptide or biologically active polypeptide thereof A nucleic acid molecule that encodes a fragment.   3. The nucleic acid encodes a human endoglin polypeptide, or The nucleic acid according to 2.   4. The nucleic acid is genomic DNA, cDNA, mRNA, or cRNA. Is a nucleic acid molecule according to 2.   5. The nucleic acid molecule according to claim 4, wherein the genomic DNA is CECT 4475.   6. The method according to claim 1 or 2, operably linked to a promoter for RNA transcription. Listed nucleic acid molecules.   7. A vector comprising the nucleic acid molecule according to claim 5.   8. A host cell containing the vector of claim 7.   9. The host cell according to claim 8, wherein the cell is a prokaryotic cell or a eukaryotic cell.   10. Preparation of endoglin-derived polypeptides or active fragments thereof A method comprising the following steps:   a. Encodes an endoglin-derived polypeptide or active fragment thereof Inserting the nucleic acid molecule into a suitable expression vector;   b. A process for inserting the obtained vector into an appropriate host cell. Degree;   c. The host cell thus obtained is induced to induce the polypeptide or activity derived from endoglin. Expressing the active fragment thereof; and   d. Purification of the resulting endoglin-derived polypeptide so produced The process of doing.   11. Soluble endoglin-derived polypeptide.   12. An endoglin-derived polype produced by the method of claim 10. Peptide or a biologically active fragment thereof.   13. Polypeptide according to claim 11 or 12 and pharmaceutically acceptable A pharmaceutical composition comprising a carrier.   14. Pharmaceutically acceptable carrier and substantially other host cell proteins Quality-free, purified and isolated human endoglin-derived polypeptide Or a biologically active fragment thereof.   15. It is a method of modifying the biological function mediated by the regulatory activity of TGF-β. And TGF-β in an effective amount of endoglin-derived polypeptide or biologically active Contacting with the fragment, thereby improving the biological function. How to change.   16. 16. The method of claim 15, wherein TGF-β is TGF-β1.   17． 16. The method of claim 15, wherein TGF-β is TGF-β3.   18. The step of contacting is performed in vitro. Item 15. The method according to Item 15.   19. 16. The method of claim 15, wherein the contacting step is performed in vivo.   20. 16. The method according to claim 15, wherein the regulatory activity is cell growth stimulation or cell growth inhibition. The method described.   21. 16. The method according to claim 15, wherein the regulatory activity is promotion of extracellular matrix production. How to list.   22. 16. The method of claim 15, wherein the regulatory activity is regulation of immune function.   23. The polypeptide is a disulfide linked sub-unit of about 80 kDa to about 95 kDa. Purified human endoglin-derived polypeptide containing a knit homodimer 16. The method of claim 15, which is also a biologically active fragment thereof.   24. The purified polypeptide is substantially free of other human proteins. , A purified and isolated human endoglin-derived protein, The method according to claim 23.   25. Said polypeptide is a soluble endoglin-derived polypeptide The method according to claim 15.   26. Treatment of pathological conditions caused by cell growth stimulation controlled by TGF-β The method of claim 13, wherein the amount is an amount effective to bind to TGF-β. Administering the pharmaceutical composition of claim 1 to a subject, thereby resulting in cell growth. A method of treating a pathological condition caused by irritation.   27. 27. The method of claim 26, wherein TGF-β is TGF-β1.   28. 27. The method of claim 26, wherein TGF-β is TGF-β3.   29. 27. The method of claim 26, wherein the subject is a human patient.   30. Cures pathological conditions caused by inhibition of TGF-β-controlled cell growth A method of treatment, wherein the amount is effective for binding to the TGF-β. Administering the pharmaceutical composition according to claim 4 to a subject, whereby cell formation is increased. A method of treating a pathological condition caused by inhibition of length.   31. 31. The method of claim 30, wherein TGF-β is TGF-β1.   32. 31. The method of claim 30, wherein TGF-β is TGF-β3.   33. 31. The method of claim 30, wherein the pathological condition is ulcer or immunosuppression.   34. 31. The method of claim 30, wherein the subject is a human patient.   35. Caused by promoting TGF-β-regulated extracellular matrix accumulation A method of treating a pathological condition in which an effective amount of TGF-β is bound is claimed. Item 12. A method comprising the step of administering the pharmaceutical composition according to Item 12 or 13 to a subject, To treat pathological conditions caused by promotion of extracellular matrix by Law.   36. 36. The method of claim 35, wherein the subject is a human patient.   37. The pathological condition is inflammation, rheumatoid arthritis, inflammatory Skin damage, scar tissue formation, lung fibrosis, liver fibrosis, atherosclerosis, or glomerulus 36. The method of claim 35, which is nephritis.   38. Endoglin and an amount of endoglin effective for binding endoglin Contacting with a polypeptide capable of binding to the A method of inhibiting the activity of gulin.   39. 39. The method of claim 38, wherein the polypeptide is TGF-β.   40. Fragment of TGF-β having the ability of said polypeptide to bind to endoglin. 39. The method of claim 38, having an amino acid sequence corresponding to a segment.   41. 40. The method of claim 39, wherein the TGF-β is TGF-β1 or TGF-β3. .   42. 41. The method of claim 40, wherein the TGF-β is TGF-β1 or TGF-β3. .   43. 39. The method of claim 38, wherein the polypeptide is an anti-endoglin antibody. Law.   44. 44. The method of claim 43, wherein the antibody is a monoclonal antibody.   45. 39. The method of claim 38, wherein the contacting step is performed in vivo.   46. 39. The method of claim 38, wherein the contacting step is performed in vitro.