PT94067A - METHOD OF OBTAINING POLYETHEPTIDO-ANTICORPH CONJUGATES FOR THE INHIBITION OF CELL ADHESION AND PHARMACEUTICAL COMPOSITIONS THAT CONTAIN THEM - Google Patents

Description

- 2 - 71 028 SCRF 146.0

DESCRIPTIVE MEMORY

Technical Field The present invention relates to the preparation of a polypeptide-antibody conjugate comprising an integrin-binding polypeptide operatively linked to an antibody molecule that immunoreacts with adherent cell surface antigens and thereby inhibits cell adhesion. Also contemplated is a therapeutical process in which a polypeptide-anorectic conjugate of the present invention is used to inhibit cell adhesion dependent processes such as tumor growth and platelet aggregation.

Background Cell adhesion is a critical process in several cell-cell and extracellular matrix-cell interactions, including tumor growth and platelet aggregation. Accordingly, agents that inhibit cell adhesion have therapeutic application at least to inhibit tumor growth or to inhibit platelet-mediated processes such as blood coagulation. Cell adhesion generally involves an interaction between cell surface receptors and matrix-specific proteins. Rouslahti et al., Science 238: 491-497 (1987). Of interest are receptors which specifically bind to matrix proteins at sites which have, as part of their structure, the Arg-Gly-Asp (KGD) amino acid residue sequence tripeptide. These receptors are known in the art as RGD-directed adhesion receptors and include the fibronectin receptor (rNF), the vitronectin receptor (rNV), the platelet receptor known as GPIIb / IIIa, and the collagen receptor. Recently, more RGD-directed adhesion receptors have been described in human melanoma M21 cells, Cheresh et al., J. Org. Biol. Chem, 262: 1434-1437 (1987), and in human endothelial cells, designated ECr, Cheresh et al., Proc. Natl. AcadY Sci. USA, 84: 6471-6475 (1987). 71 028 SCRF 146.0 - 3 - r

RGD-containing polypeptides derived from matrix protein sequences have been shown to inhibit the ability of RGD-directed adhesion receptors to interact and adhere to a substrate containing matrix proteins. Pytela et al., Froc. Natl. Acad. Know. USA, 82: 5766-70 (1985); and Pytela et al., Science 231: 1559-62 (1986).

RGD-containing polypeptides have also been used to inhibit cell adhesion to a substrate containing matrix proteins, in vitro. U.S. Patent Nos. 4,614,517, 4,517,686, 4,578,079, 4,683,291, 4,792,525; Pierschbacher et al., Nature, 309: 30-33 (1984); Pierschbacher et al., Proc. Natl. Acad. Know. USA, 81: 5985-88 (1984); Hayman et al., J. Cell. Biol., 100: 1948-54 (1985); Haverstick et al., Blood, 66: 946-52 (1985); Gins-berg et al., J. Biol. Chem., 260: 3931-36 (1985); Armant et al., Proc. Natl. Acad. Know. USA, 83: 6751-55 (1986); and Ouaissi et al., Science, 234: 603-7 (1986). It has been shown that a polypeptide containing RGD inhibited the adhesion of murine melanoma cells and also inhibited the in vivo colonization of mouse lung tissues. Hum-phries et al., Science, 233: 467-70 (1986).

Similar in vivo inhibition of colonization of lung tissues was described using a polypeptide derived from a laminin matrix protein sequence, Iwamoto et al., Science, 238: 1132-1111 (1987). Polypeptides containing RGD and other polypeptides obtained from laminin, which inhibited cell adhesion, were disclosed which contained the amino acid sequence of the pentapeptide Tyr-Ile-Gly-Ser-Arg (YIGSR), thus indicating that polypeptides containing RGD are not exclusive candidates for inhibition of cell adhesion. Inhibition of cell adhesion has also been demonstrated using monoclonal antibodies that immunoreact with a molecule present on the surface of the cells undergoing adhesion. For example, inhibitory antibodies to cell adhesion have been described, which immunoreact with RGD-directed adhesion receptors, such as GPIIb / IIIa / Pytela et al., Science, 231; 1559-62 (1986); Charo et 71 028

SCRF 146.0-4 al., J. Blol. Chem. 262: 9935-38, (1987), a 140-kilodalton / Brown et al., Science / 228: 1448-51 (1985) /, cell-substrate binding antigen (CSAT) / Horwitz et al., J. Cell. Blol. 101: 2134-44 (1985)) and the human melanoma cell receptor / Cheresh et al., J. Biol. Chem. 262: 17703-11 (1987) Cell adhesion inhibitory antibodies have been described that immunoreact with antigen from adherent cells when the antigens are not themselves RGD-driven adhesion receptors. For example, it has been shown that monoclonal antibodies which immunoreact with GD2 or GD3 disialogangliides inhibit tumor cell adhesion to various substrates containing matrix proteins, including collagen, vitronectin, la-reticulin and fibronectin. Cheresh et al., J. Cell. Biol., 102: 688-96 (1986).

In medicine, it has been concluded from the foregoing discussion that RGD-containing polypeptides and other polypeptides derived from primary sequences of matrix proteins are useful for inhibiting cell adhesion, in particular adhesion of tumor cells or platelets. The systemic use of adhesion inhibiting polypeptides is, however, limited because it produces the side effect of systemically and indiscriminately inhibiting normal cell adhesion phenomena.

BRIEF SUMMARY OF THE INVENTION

It has now been found that the cell-adhesion inhibiting effects of the polypeptides described herein can be directed (targeted) to particular tissues by their conjugation with specific antibodies so that the polypeptides can cease specific, undesirable phenomena, of cell adhesion without significantly affecting normal adhesion phenomena. The present invention provides an RGD antibody comprising an antibody molecule operably linked to an integrin-binding polypeptide. The antibody combining site of the RGD antibody immunoreacts with a cell surface antigen of the adherent cell and the polypeptide has an amino acid residue sequence of about 5 to about 50 residues in length, including a sequence of the formula: -RGD-; more preferably, with the formula: -GRGDSP- and most preferably of the formula -CGGAGACRGD SP-.

In addition, it is preferred that the antibody molecule of the RGD antibody is considered to immunoreact with an antigen on the surface of the adherent cell. The surface antigens of the cell adhere. Preferred are disialoganglioside GD2, disialoganglioside GD3, chondroitin sulfate proteoglycan, vitronectin receptor, endothelial cilalast receptor or collagen receptor. Preferably, the antibody molecule is produced by a hybridoma selected from the group consisting of 1418, 142A, 126, MB3.6, LM609 and LM142.

In preferred embodiments of the invention, a contemplated RGD antibody immunoreacts with an antigen present on the surface of platelets or tumor cells. The present invention also contemplates a method of inhibiting the binding of an RGD-containing matrix-adherent cell, which method comprises administering to a patient a therapeutically effective amount of an RGD antibody of the present invention. Preferably, the adherent cell is a tumor cell or a platelet, and thus the process contemplates inhibition of tumor cell binding or platelet aggregation, respectively.

In another embodiment, the present invention contemplates a YIGSR antibody comprising an antibody molecule that immunoreacts with an adherent cell surface antigen operably linked to a polypeptide that binds to a laminin receptor, having an amino acid residue sequence of from about 5 to about about 50 residues in length which includes a sequence of the formula: -YIGSR-. Preferably the polypeptide has the formula: YIGSR, CDPGYIGSR or RGDSGYIGSR. The invention further contemplates a chimeric antibody containing at least one hybrid protein molecule having a fragment which forms an antibody combining site fused to at least one integrin binding polypeptide in which the molecule of protein forms an antibody combining site that immunoreacts with a surface antigen of an adherent cell (adherent cell surface antigen). In a preferred embodiment, the integrin-binding polypeptide comprises an amino acid residue sequence of about 5 to about 50 residues in length, which includes a sequence of the formula: -RGD- and, more preferably, with the formula -GRGDSP -. The present invention also contemplates a recombinant DNA molecule encoding a hybrid protein molecule comprising: a) a first segment of DNA encoding a fragment that forms an antibody combining site, and b) a second DNA segment encoding a binding polypeptide integrin, which is operably linked in phase with the first segment, wherein the recombinant DNA molecule is capable, when present in a suitable expression vector, of expressing a hybrid protein molecule having an antibody combining site which immunorreage. with an adherent cell surface antigen.

Brief Description of the Drawings Figure 1 is a figure comprising two graphs (A and B) graphically illustrating the effects, after 20 minutes, of a polypeptide-antibody conjugate on. The. binding of M21 cells to microtiter wells, each well being coated either with von Willebrand fac- tor (Chart A) or with fibrinogen (Chart B). - 3

H-leucine-labeled M21 cells immunoreacted first with the indicated concentrations of either the Mab 142A (blank squares) conjugate or the Mab 142A (solid squares) activated control before the reaction with the polypeptide. The cells treated with anti-body were then placed in the coated microtiter wells and held for 20 minutes to allow adhesion of the cells to the wells as described in Example 4. The results are expressed as the total number of bound cells, that number being a function of the cell-associated marker, expressed as counts per minute (cpm x 10), bound to the maize protein, for each of the indicated Mab concentrations. Figure 2 is a graph comprising two graphs (A and B) graphically illustrating the effects after 90 minutes of a polypeptide-antibody conjugate on the binding of M21 cells to the microtiter wells, each well having been coated either with the factor of von Willebrand (Chart A) or with fibrinogen (Chart B). The reagents and methods are as described for Figure 1 and the results are shown as for Figure 1.

DETAILED DESCRIPTION OF THE INVENTION A. Definitions

Amino acid: All amino acid residues identified herein are in the natural L configuration. Maintaining the classical nomenclature of the polypeptides, J. Biol, Chem. 243: 3557-59 (1969), the abbreviations of the amino acid residues are those set forth in the following Table of Correspondence: - 8 - 71 028 SCRF 146.0

CORROSPONDENCE SYMBOL Symbol Amino acid 1 letter 3 letters Y Tyr L-tyrosine G Gly glycine F Phe L-phenylalanine M Met L-methionine A Ala L-alanine S Ser L-serine I ile L-isoleucine L Leu L-leucine T Thr L -thyronine V Valine L Valine P Pro L-proline K Lys L-lysine H His L-histidine Q Gin L-glutamine E L-glutamic Glutate W Trp L-tryptophan R Arg L-arginine D L-aspartic acid N Asn L-asparagine C Cys L-cysteine

It will be appreciated that all amino acid residues sequences are represented herein by formulas whose left-to-right orientation is the conventional amino-terminal direction to the carboxy-terminus. It is further noted that a trace at the beginning or the end of an amino acid residue sequence indicates a bond to another sequence of one or more amino acid residues, up to a total of about 50 residues in the polypeptide chain.

Polypeptide and Peptide; Polypeptide and peptide are terms used interchangeably herein to designate a linear series of no more than 50 amino acid residues linked to each other by 71 028 SCRF 146.0

Peptide bonds between the alpha-amino and carboxyl groups of the adjacent residues.

Protein; Protein is a term herein used to denote a molecule with a primary structure consisting of a linear series greater than 50 amino acid residues linked to each other as in a polypeptide. Of course, both a protein and a polypeptide may further include intramolecular disulfide bonds and other bonds which cause the protein or polypeptide to assume a non-linear configuration when in solution. B. Polypeptide-Antibody Conjugates 1. Background The polypeptide-antibody conjugate of the present invention comprises, in its more generic embodiment, an antibody molecule operably linked to an in-tegyn-binding polypeptide. The bound antibody molecule immunoreacts with an antigen present on the surface of an adherent cell and the integrin binding polypeptide inhibits cell adhesion. The terms cell adhesion, integrin binding polypeptide and cell adhesion will be discussed and defined hereinbelow. The combination of an antibody molecule and a polypeptide as described herein produces a cell adhesion inhibitory molecule because, upon immunoreaction of the antibody with an adherent cell surface antigen, the antibody locates the binding polypeptide integrin, bound in sufficient proximity to the cell surface integrin molecules, so that the polypeptide can interact and be bound by the integrin, or integrin-bound, thus acting as an antagonist and inhibiting the ability of the bound integrin to participate in the process of accession.

In view of the utility of an integrin-binding polypeptide in inhibiting cell adhesion, the combination of the antibody molecule with the polypeptide described in the present invention provides the benefit of guiding the inhibitory effect to the cells that are present in the cell. SCRF 146.0 contain a certain antigen from the surface of the adherent cell. Without this guidance, the topical or systemic application of an integrin binding polypeptide can indiscriminately inhibit the adhesion of all exposed cells. to the polypeptide, including that of the adherent cells present in normal tissues. Those skilled in the art of cell adhesion can thus readily verify the benefit of localizing the inhibitory effects of cell adhesion on specific adherent cells. The " cell adhesion " refers to a process by which an adherent cell binds (adheres) to the surface of another cell or to a tissue substrate. The adhesion process occurs through a specific interaction between the adhesion cell surface protein receptors, designated integrins, and the ligands of the matrix protein which includes the tripeptide-RGD as part of its amino acid residue sequences. Thus, the adherent cells referred to herein are cells that adhere to biological surfaces by means of the specific interaction (formation of the receptor / ligand complex) between an integrin and RGD-containing matrix proteins. RGD-containing matrix proteins are also referred to as adhesion protein ligands and include fibronectin, vitronectin, fibrinogen, von Willebrand factor, laminin, thrombospondin, osteopontin, collagens and the like.

An adherent cell is also characterized as any cell that contains on its surface one or more cell surface receptors known as integrins. Integrins are also referred to as RGD or cytoadhesin-directed adhesion receptors, and include the vitronectin receptor, fibronectin receptor, collagen receptor, glycoprotein GPIIb /

Platelet inhibitor, human endothelial cell receptor and the like. Examples of adherent cells are platelets and numerous tumor cell types including the neuroblastoma, small cell carcinoma, lung, colon or breast adenocarcinomas, melanocyte, astrocytoma, glioma, squamous cell carcinoma, and the like. 71 028 SCRF 146.0-H-

An adherent cell surface antigen refers to any antigen present on the surface of an adherent cell. " Antigen " is a term herein used to denote any molecule that immunoreacts with an antibody molecule and is bound by the antibody to form an immunoreactant. Thus a surface antigen of an adherent cell may be any surface molecule of an adherent cell that immunoreacts with an antibody molecule while the antigen is present on the surface of the cell.

Examples of adherent cell surface antigens include GD2 and GD3 disialogangliosides, tumor cell surface antigens, platelet surface antigens, chondroitin sulfate proteoglycans, cell surface receptors such as integrins, and the like. 2. Antibody Molecules

Antibody molecules to be used as a component of the present invention include any antibody molecule that is capable of immunoreacting with a surface antigen from adherent cells to form an immunoreaction product (antigen-antibody complex). The term " antibody molecule " in its various grammatical forms as used herein encompasses both an intact immunoglobulin molecule and an immunologically active portion of an immunoglobulin molecule, i.e., molecules which contain a combination site of antibody or paratope. " An antibody combining site " is the structural portion of an antibody molecule consisting of variable and hypervariable regions of light and heavy chains, which binds specifically to the antigen.

Examples of antibody molecules include intact immunoglobulin molecules, substantially intact immunoglobulin molecules, and portions of an immunoglobulin molecule containing the paratopo, including the portions known in the art by Fab, Fab ', F (ab') 2 and F (V).

The Fab and F (ab ') 2 portions of the antibody molecules are prepared by the proteolytic reaction of papain and pepsin, respectively, with substantially intact antibody molecules, by well known methods. See, for example, U.S. Patent No. 4, 342, 566 to Theofilopolous and Dixon. Portions of the Fab 'antibody molecule are also well known and are produced from F (ab') 2 moieties, followed by reduction of disulfide bonds linking the two heavy chain moieties, e.g. and the alkylation of the mercaptan of the resulting protein with a reagent such as iodoacetamide. The term " antibody " in its various grammatical forms is used herein to refer to a composition containing several molecules, of the / antibody, e.g. e.g., an antiserum. An antibody that contains intact molecules / antibody is preferred. A monoclonal antibody is particularly preferred. The term " monoclonal antibody " or " Mab ", in its various grammatical forms, refers to an antibody containing only one antibody combining site species capable of immunoreacting with a particular antigen, and thus thus showing a single binding affinity for this antigen. A monoclonal antibody may therefore contain a bi-specific molecule / antibody, with two antibody combining sites, each immunospecific, for a different antigen.

A preferred monoclonal antibody is characterized in that it contains, within immunologically detectable limits, only a species of antibody combining sites capable of immunologically (immunoreacting with) a cell surface antigen of the adherent. Particularly preferred are monoclonal antibodies which immunoreact with the GD2 or GD3 disialogangliosides, with the vitronectin receptor, fibronectin receptor, cell receptor " 13-71 028 SCRF 146.0 human endothelial, chondroitin sulfate proteoglycan or GPIIb / IIIa platelet glycoprotein. Table 1 lists the immunospecific monoclonal antibodies useful in the present invention. Such Mabs are referred to by the name used in a publication, by the accession number, indicated by the American Type Tissue Collection (ATTC), of the hybridoma producing the antibody, and by the antigen with which each Mab reacts. In a footnote there is a quotation about the discussion of each Mab and its immunospecificity with respect to the indicated antigen. Hybridomas that produce monoclonal antibodies, functionally equivalent to those in Table 1, can be prepared by techniques well known in the art including those described herein. TABLE 1

Examples of Monoclonal Antibodies Immunoring with a Surface Cell Antigen of Adherent

Mab ATTC N9 Antigen 1418 HB 9118 disialoganglioside GD2 142A disialoganglioside GD2 126 HB 8568 disialoganglioside GD2 MB3.6 HB 8890 disialoganglioside GD3 11C64 disialoganglioside GD3 R24 disialoganglioside GD3 9227 chondroitin sulphate proteoglycan LM609 HB 9537 vitronectin receptor ^ LM142 - vitronectin receptor HT29 / 26 HB 8247 colon cancer glycoprotein gp29 ^ T16 HB 8279 human bladder tumor glycoprotein gp48 '*' * * 1116-NS-19-9 HB 8059 mono-sialoganglioside of colorectal carcinoma · 14-71 028 SCRF 146.0 CTHB-5 KB 135 " complement C3d receptor OKT 11 CRL 8027 14 Anti-HLA15 rosette receptors histocompatibility antigens (HLA) TM1 HB 169 leukocyte antigen (leu-5 ) 16 C7 CRL 1691 low density lipoprotein 17 receptor BBM HB 28 * | the microglobulin B-2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Cheresh et al., Cancer Res., 44: 5112-18 (1986).

Prepared as described in Example 1.

U.S. Patent No. 4,675,287.

Cheresh et al., Proc. Natl. Acad. Know. USA, 82: 5155-59 (1985). Cheresh et al., J. Clin. Blol., 102: 688-96 (1986).

U.S. Patent No. 4,507,391.

Bumol et al., Proc. Natl. Acad. Know. USA, 79: 1245-49 (1982). Cheresh et al., J. Biol. Chem., 262: 17703-11 (1987).

Cheresh et al., J. Biol. Chem., 262: 17703-11 (1987).

U.S. Patent No. 4,579,827.

European Patent Application No. 84102517.4, Publication No. 0 118 891, published September 19, 1984.

U.S. Patent No. 4,471,057.

Weis et al., Proc. Natl. Acad. Know. USA, 81: 881-885 (1984). U.S. Patent No. 4,364,937.

ATTC Catalog of Cell Lines and Hybridomas, 6,. edition, 1988, pg. 283 lists more than 27 different Mabs that immunoreact with numerous HLA antigens.

Grumet et al., Human Immunol., 6: 63-73 (1983).

Goldstein et al., Cell, 30: 715-24 (1982).

Parham et al., Eur. J. Immunol, 9: 536-45 (1979). 18 71 028

SCRF 146.0 - 15 -

Antibodies or monoclonal antibodies that immunoreacted with antigens on the surface of adherent cells can be obtained from a number of commercial vendors.

Alternatively, the antibodies may be prepared by immunization techniques well known in the art. The preparation of monoclonal antibodies and their subsequent purification sufficient for the use of the present invention are well known. Typically, a monoclonal antibody is prepared from the supernatant of a hybridoma culture, separating the supernatant from the cultured hybridoma cells, to form a solution containing monoclonal antibody molecules free of the referred cells. Alternatively, the monoclonal antibody may be prepared from ascitic fluid by introducing a hybridoma cell, for example by injection, into the peritoneal cavity of a mammal, for example a mouse, and collecting the resulting peritoneal exudate (fluid of the ascitic tumor) of the mouse by well known techniques. See, for example, H. Zola, Monoclonal Antibodies: A Manual of Techniques CRC Press, Inc. (1987).

In addition, a monoclonal antibody may further be produced by recombinant DNA methodologies wherein I have provided a gene encoding a monoclonal molecule / antibody and cloned and manipulated in an expression medium suitable for obtaining an antibody molecule produced by recombinant process. The production of monoclonal antibodies by recombinant methods is described below in more detail in Section D entitled " Chimeric Antibodies ". " in

Particularly preferred are the monoclonal molecules / antibody 142A, 1418, 12.6, MB3.6, LM142 and LM609, described in more detail herein. Hybridomas producing the monoclonal antibody molecules 1418, 126, MB3.6 and LM609 were deposited, as required by the Budapest Treaty, at the American Type Tissue Collection (ATTC), Rockville, MD, on June 4, 1986 , 28 May 1984, 16 August 1985 and 15 September 1987, respectively, and assigned accession numbers HB 9118, HB 8568, HB 8890 and HB 9537, respectively. 71 028

SCRF 146.0 - 16 - 3. Polypeptides

An integrin-binding polypeptide useful in the present invention is a polypeptide having an amino acid residue sequence of about 5 to about 50 residues in length and which discloses the property of competitively inhibiting cell adhesion.

Useful polypeptide binds to an integrin expressed on the surface of an adherent cell. It also inhibits cell adhesion between an adherent cell and a surface containing a matrix protein.

Many polypeptides are known to exhibit these adhesion inhibition properties. See, for example, Rouslahti et al., Cell, 44: 517-518 (1986); Pierschbacher et al., Nature, 309: 30-33 (1984), Ouaissi et al., Science, 234: 603-607 (1986); Iwamoto et al., Science, 238: 1132-1134 (1987); Haverstick et al., Blood, 66: 946--952 (1985) and U.S. Pat. No. 4 578 079, 4 792 525, 4 614 517, 4 517 686 and 4 683 291. (The teachings of the art cited herein are incorporated by reference only).

Conditions for inhibition of adhesion to integrin binding polypeptides are well known in the art. Examples of such conditions are those described in Example 4 and those cited immediately above.

Usually, integrin-binding polypeptides contain a sequence of amino acid residues that corresponds to a portion of the sequence of an adhesion protein. This polypeptide sequence is present in that part of the adhesion protein which participates in contact with the integrin which binds upon adhesion. This contacting portion of the adhesion protein has been referred to as the cell recognition site. For many of the adhesion proteins identified, the recognition site includes the RGD amino acid triplet. See, for example, Rouslahti et al., Science, 238: 491-497 (1987). The ability of an integrin-binding polypeptide to bind to an integrin and to inhibit cell adhesion is quantified.

by means well known in the art. Examples of such media are referred to in detail in Examples 3 and 4. Further examples of such media can be found in the publications cited above which describe examples of integrin binding polypeptides.

A preferred integrin-binding polypeptide has an amino acid residue sequence which includes residues represented by the formula -RGD-. Examples of such polypeptides are those represented by the formulas: VTGRGD, GRGDS and RGDSPASSKP.

Preferably, the polypeptide sequence includes amino acid residues represented by the formula: - GRGDSP-. Examples of such polypeptides are those represented by the formulas: AVTGRGDSP GRGDSP and CGGAGAGRGDSP.

For a polypeptide with more residues than those included in the above-cited sequences, it is preferred that additional residues are chosen such that the entire resulting polypeptide has a sequence corresponding to a portion of the fibronectin amino acid residue sequence. The sequence of a 11.5 kilodalton fragment of fibronectin containing the RGD sequence is described by Pierschbacher et al., J. Biol. Chem., 257: 9593-97 (1982).

When a polypeptide-antibody conjugate contains triptemptide -RGD- as part of its integrin binding polypeptide, it is referred to herein as a -RGD antibody.

Another preferred integrin-binding polypeptide has an amino acid residue sequence which includes residues represented by the formula: -YIGSR-. For a polypeptide with more residues than this sequence, it is preferred that any residues

J 71 028 SCRF 146.0 18

are chosen so that the resulting whole polypeptide has a sequence corresponding to a part of the amino acid residue sequence of the Blin chain of the laminin. The Bl chain sequence of laminin is described by Sasaki et al., Proc. Natl. Acad. Know. USA, 84: 935-39 (1987).

When a polypeptide-antibody conjugate contains the pen-peptide -YIGSR- as part of its integrin-binding polypeptide. will be referred to herein as -YIGSR antibody. Examples of polypeptides suitable for use in a -YIGSR antibody are polypeptides represented by formulas

YIGSR CDPGYIGSR and RGDSGYIGSR.

An integrin-binding polypeptide to be used in a polypeptide-antibody conjugate of the present invention may be synthesized by any suitable method known to those skilled in the art of the polypeptides, including recombinant DNA techniques. Thus, the synthesis can be done exclusively by solid phase techniques, by partial solid phase techniques, by condensation or fragment splitting or by the addition of classical solutions.

The polypeptides are preferably prepared for reasons of purity, undesirable side product exemption, ease of production and the like, using Merrifield type solid phase synthesis. As, for example, described in Merrifield, J. Am. Chem. Soc., 85: 2149 (1964), although other equivalent chemical syntheses known in the art can also be used, for example the syntheses of Houghten, Proc. Natl. Acad. Know. USA, 82: 5132 (1985). It may also be seen from the many available techniques, in J.M. Steward and J.D. Young, " Solid Phase Peptide Synthesis ", W.H. Freeman Co., San Francisco, 1969; M. Bo-danszky et al., &Quot; Peptide Synthesis ", Jonh Wiley & Sons, Second Edition, 1976; and J. Meienhofer, " Hormonal Proteins and Peptides ", Vol. 2, p. 46, Academic Press (New York), 1983 for the syntheses of pêp-71 028

J

SCRF 146.0 - 19 - solid phase, and E. Schroder and K. Kubke, " The Peptides ",

Vol. I, Academic Press (New York), 1965, for classical syntheses in solutions 4. Cloning of the Polypeptide to the Antibody The polypeptide-antibody conjugate of the present invention contains an integrin binding polypeptide which is operably attached to a molecule / antibody, previously discussed, that imu-norreage with an antigen of the surface of an adherent cell. The term used herein " operatively " means that the polypeptide and the antibody molecule are physically associated by a binding medium that does not significantly interfere with the ability of any of the attached groups to function as described herein. This ability does not suffer significant interference when the number of operatively fixed polypeptides is limited, so that each antibody molecule is linked with from 1 to about 30 polypeptides. It is preferred that from about 8 to about 12 polypeptides per molecule / antibody, and still more preferably about 10 polypeptides per molecule.

Methods for controlling the number of polypeptides that are attached (conjugated) to an antibody molecule are well known in the art, and usually depend on the chemistry of the linkage used in particular. Typically, such processes are based on controlling the polypeptide: molecule / antibody ratio in the binding reaction mixture or the time left for the binding reaction to occur or both. An example of a binding process is described in Example 2.

Because the antibody molecules are themselves proteins, the protein conjugation or coupling techniques, through active functional groups, are particularly applicable to the operably attached polypeptide to the molecule / Importer. See, for example, Aurameas et al., Scand. J. Immunol., Vol. 8, Suppl. 7: 7--23 (1978) U.S. Patent Nos. 4,493,795 and 4,671,950.

One or more additional amino acid residues may be attached to the amino or carboxyl termini of the synthetic polypeptide to aid in binding of the polypeptide to a carrier. Cysteine residues added to the amino or carboxyl termini of the synthetic polypeptide were found to be particularly useful for forming linkages via disulfide bonds. Other methods well known in the art of preparing conjugates may, however, be used. Additional examples of binding processes include the Michael addition reaction products, dialdehydes such as glutaraldehyde, Klipstein et al., J. Infect. Pis., 147: 318-326 (1983), and the like, or the use of carbodiimide technology, for example the use of water soluble carbodiimide to form amide bonds to the antibody molecule.

It is preferred that the binding medium is a covalent coupling between a cysteine terminal in the polypeptide and the epsilon amino group of a lysine residue present in the monoclonal antibody molecule.

As is well known in the art, it is often beneficial to attach a synthetic polypeptide to its antibody molecule via an intermediate linking group. As noted above, glu-valaldehyde is one of such linking groups. However, when cysteine is used, the intermediate linking group is preferably m-maleimidobenzoyl-N-hydroxysuccinimide (MBS) or 4- (maleimidomethyl) - N -hydroxysuccinimide ester 1-cyclohexane-carboxylic acid (NHS), which was used herein.

In addition, MBS can first be coupled to the antibody molecule by an ester-amide exchange reaction, as described by Liu et al., Biochem, 80: 690 (1979). The addition may then be followed by the addition of a blocked mercapto group, such as thiolacetic acid (CH3), by the double bonding of the maleimide. After cleavage of the acyl blocking group, a disulfide bond is formed between the unlocked mercaptan linking group and the mercaptan of the cysteine residue of the synthetic polypeptide.

In addition, coupling reactions directed to the SCRF 146.0 site can be carried out such that the fixed polypeptide does not substantially interfere with the immunoreaction of the antibody molecule with the adherent cell surface antigen. See, for example, Rodwell et al., Biotech. 3: 889-894 (1984).

The polypeptides used in this invention may contain additional residues at either end, in order to provide a " spacer " by which the integrin-binding polypeptide can be operatively attached to the antibody. The use of a spacer will therefore elongate the polypeptide beyond the binding site on the antibody molecule, rather than if the polypeptide has been attached without a spacer.

Amino acid residue spacers usually have a length of from 1 to about 50 residues, more often from 2 to 10 residues and do not necessarily encompass sequences that correspond to the sequence of an adhesion protein.

A preferred spacer polypeptide contains the amino acid residue sequence CGGAGA operably linked by its amino terminal carbaxyl terminus, via a peptide bond, to the amino-terminal residue of the integrin-binding polypeptide.

More preferably, the spacer polypeptide, when attached to an integrin binding polypeptide, comprises the sequence CGGAGAGRGDSP.

A polypeptide used in the present invention may be linked together to form a polymer (synthetic multimer) comprising a multiplicity of repeating units of the polypeptide. This polymer usually has the advantage of increased exposure of the cell recognition site for potential binding to the integrins.

A polymer for use in this invention may be prepared by synthesizing a polypeptide as hereinbefore defined and including a cysteine residue at both the amino and carboxyl termini to give a " di-cys " polypeptide. After the synthesis of the polypeptide, 71 028 SCRF 146.0 is dissolved in a typical laboratory preparation.

10 mg of the polypeptide in di-cys (containing cysteine residues in non-oxidized form) in 250 ml of 0.1M ammonium bicarbonate buffer. The di-terminated polypeptide, once dissolved, is then oxidized to air by gentle stirring in the air of the resulting solution over a period of 18 hours or until there are no free mercaptans detectable by the Ellman test (see Ellman, Arch. Biochem. Biophys., 82: 70-77 (1959) J. The polymer thus prepared contains a multiplicity of repeating units of the synthetic polypeptide which are randomly bound by cysteine residues. These polymers typically contain their head-to-tail attached polypeptide repeat units as well as tail-to-tail head-to-head; that is, the amino termini of two repeating units of the polypeptide may be attached together by a single cysteine residue, for example two carboxyl termini, since the linking groups of both ends of the polypeptide are identical.

Alternatively, the polypeptides may be attached in various branched configurations of the polymer, offering a similar advantage over the polymer containing linearly disposed repeat units.

In another arrangement, it is contemplated that the different integrin-binding polypeptides may be operably fixed to a single molecule / antibody. For example, there may be two or more different polypeptides which are attached, independently of one another, as monomers, polymers or branched polymers, to the antibody molecule by the same binding means described above for the polymers. In this arrangement each polymer comprises different monomers of integrin-binding polypeptides. C. Therapeutic Methods and Compositions The present invention contemplates a polypeptide-antibody conjugate and compositions containing said conjugate, wherein said conjugate comprises an antibody molecule operably attached to an integrin binding polypeptide, said antibody being said antibody mm

JET. in . lecula / antibody capable of immunoreacting with an antigen on the surface of an adherent cell and said polypeptide being a polypeptide described herein with adhesion inhibiting properties.

Preferably, the polypeptide-antibody conjugate is an -RGD antibody, wherein said conjugate contains a polypeptide comprising an amino acid residue sequence of about 5 to about 50 residues in length, which includes a sequence having the formula: -RGD- and, more preferably, with the formula: -GRGDSP-.

In another arrangement, the polypeptide-antibody conjugate is an -YIGSR antibody wherein said conjugate contains a polypeptide having an amino acid residue sequence of 5 to about 50 residues in length, which includes a sequence of the formula: -YIGSR- . The present invention further contemplates that a polypeptide-antibody conjugate of the present invention and compositions containing said conjugate may be used in a method for inhibiting attachment of adherent cells to a surface containing adhesion proteins. The polypeptide-antibody conjugate of the present invention may be used in a pharmaceutically acceptable composition which, when administered to a patient in a therapeutically effective amount, is capable of inhibiting the ability of an integrin molecule present in the adherent cells to bind to a protein matrix containing RGD (adhesion protein). Such inhibition of integrin is believed to result from decreased (adhesion) binding ability, by cells containing an integrin molecule, to other tissue substrates in the body or to the surfaces of cells containing adhesion proteins. Thus, the in vivo administration of a composition with a polypeptide-antibody conjugate of the present invention, can be used to inhibit cell adhesion, in a specific way, which depends on the choice of the antibody molecule to be included in the conjugate, as hereinafter describe.

For example, the use of a polypeptide-antibody conjugate 71 028 SCRF 146.0

containing a monoclonal antibody molecule which immunoreacts with disialoganglioside GD2, a molecule that is present in several types of tumor cells, inhibits tumor cell adhesion and metastasis. tumor. This arrangement is efficient to inhibit the growth and metastasis of tumor cells because these phenomena depend on the fixation of tumor cells, which fixation is specifically inhibited by the adhesion inhibitor polypeptide present in the conjugate that is targeted to those tumor cells with GD2, as antigen on the cell surface, by means of the anti-GD2 antibody present in the conjugate. The use of a polypeptide-antibody conjugate containing a monoclonal antibody that immunoreacts with a platelet surface antigen, is also contemplated to inhibit platelet adhesion. It is preferred that the monoclonal antibody immunoreacts with platelet glycoprotein GPIIb / IIIa, a molecule related to platelet aggregation. Such use is effective in inhibiting blood clotting since coagulation depends in part on the ability of platelets to aggregate by adhesion of the cells.

Other uses of the adhesion inhibitors of the claimed polypeptide-antibody conjugate cells, which depend on the selection of the specific monoclonal antibodies, will be apparent to the person skilled in the art of adhesion.

It is thus seen that the polypeptide-antibody conjugate of the present invention is useful in various methods for inhibiting cell adhesion and for inhibiting processes that depend on a particular ability of the cell to adhere to the matrix proteins.

Thus, the present invention contemplates a method of inhibiting the attachment of a cell adherent to a matrix containing RGD, which comprises administering to a patient a therapeutically effective amount of a polypeptide-antibody conjugate described herein.

Preferably the conjugate is a -RGD or antibody -YIGSR antibody.

When the adhesion cell is a tumor cell, the invention contemplates a process for inhibiting tumor cell adhesion and tumor metastasis.

When the adhesion cell is a platelet, the invention contemplates a method for inhibiting platelet adhesion, platelet aggregation, and platelet dependent blood coagulation. The preparation of therapeutic compositions containing polypeptide-antibody conjugate molecules as active ingredients is well known in the art. Typically, these compositions are prepared as injectables either as liquid solutions or as suspensions, however, solid forms may also be prepared for solution or suspension in liquids prior to injection. The preparation can also be emulsified. The therapeutic active ingredient is often mixed with excipients which are pharmaceutically acceptable and compatible with the active ingredient. Suitable excipients are, for example, water, saline, dextrox, glycerol, ethanol or the like and combinations thereof. Further, if desired, the composition may contain minor amounts of auxiliary substances, such as wetting or emulsifying agents, or pH buffering agents which improve the efficiency of the active ingredient. The term used herein " pharmaceutically acceptable " refers to molecular entities and to compositions which do not produce allergic reactions or similar unpleasant reactions such as gastric discomfort, dizziness and the like when administered to a patient.

A polypeptide-antibody conjugate may be formulated into a therapeutic composition in pharmaceutically acceptable salt forms. Pharmaceutically acceptable salts include the acid addition salts (formed with the free amino groups of the polypeptide-antibody conjugate molecule) and which are prepared with inorganic acids such as, for example, hydrochloric or phosphoric acids, or with organic acids such as such as acetic, oxalic, tartaric, mandelic and the like. Salts formed with free carboxylic groups can also be obtained from inorganic bases, such as, for example, sodium, potassium,

Ammonium, calcium or ferric, and organic bases such as isopropylamine, trimethylamine, 2-ethylaminoethanol, histidine, procaine, and the like.

Therapeutic compositions containing polypeptide-antibody conjugated molecules are conveniently administered intravenously, for example, by injection of a unit dose. The term " unit dose " when used in reference to a therapeutic composition of the present invention, relates to physically discrete units suitable as unitary dosages, each unit containing a predetermined amount of active principle, calculated to produce the desired therapeutic effect in association with the appropriate diluent, is such as carrier or carrier.

Therapeutic compositions containing polypeptide-antibody conjugates may also be administered topically, for example by means of an ointment or ointment, or by application of a package or dressing containing the composition in a formulation suitable for diffusion of the active principle in the tissue in contact , as is already well known.

The compositions of polypeptide-antibody conjugates administered usually contain 0.1-95% of the active ingredient, preferably 25-70%.

The compositions are administered in a manner compatible with the dosage formulation and in a therapeutically effective amount. The amount to be administered depends on the patient being treated, the ability of the patient's system to utilize the active ingredient without adverse side effects and the degree of inhibition of the desired interaction between an integrin and an adhesive protein. The term " therapeutically effective amount " when used herein refers to an amount of polypeptide-antibody conjugate sufficient to measurably inhibit the ability of a given cell to adhere to a substrate containing an adhesion protein. Typically, this ability is measured in vitro by an adhesion assay, such as. which is described in Example 4. 71 028

SCRF 146.0 - 27 -

Measuring a therapeutically effective amount by means of an in vitro adhesion assay depends on the type of cell to be treated which, in turn, depends on the particular antibody molecule contained in the polypeptide-antibody conjugate. Thus, the cells to be used in the adhesion assay are of a type that contains an antigen on the surface of the adherent cell with which the molecule / antibody immunoregates, preferably the same cell type to be treated with the use of the composition and most preferably the cells used will be a population of the cells obtained from the patient to be treated.

A therapeutically effective amount is an amount sufficient to cause a decrease in the ability of the cell to adhere in the in vitro adhesion adhesion test to at least 10%, and preferably to about 50%. This amount can be expressed as the effective concentration determined by the in vitro assay. Thus, a therapeutically effective amount may be expressed as the amount sufficient to provide an effective concentration in the patient's blood. .

The precise amounts of active ingredient that are to be administered depend on the judgment of the physician and are peculiar to each individual. However, suitable dosage ranges are in the range of 0.1 to 10 mg, preferably one to several milligrams, of active principle per kilogram of patient body weight per day, and depend on the route of administration. Suitable regimens for initial administration and booster doses are also variables but are typified by an initial administration followed by repeated doses at one or more hours apart by injection or other route of administration. Alternatively, continuous intravenous infusion is considered sufficient to produce effective intravascular concentrations, preferably intravascular concentrations of about 10 nanomolar to about 10 micromolar. D. Chimeric Antibodies The present invention also contemplates a chimeric antibody capable of immunoreacting with a surface antigen of adherent cells and inhibiting integrin-mediated adhesion of cells. The term " chimeric antibody " used herein refers to a type of antibody molecule that contains the amino acid residue sequence of an integrin-binding polypeptide of the present invention, as part of the primary structure of the antibody molecule, e.g. ex. the sequence of amino acid residues. As an antibody molecule, a chimeric antibody is comprised of light and heavy chains, of which at least one chain is a hybrid protein molecule, with a fragment that forms an antibody combining site, which fuses with a integrin binding polypeptide. The term " antibody combining site fragment " used herein refers to a linear series of amino acid residues corresponding to the variable region portion of either heavy chain or light chain, one antibody molecule which contributes to the structure known as the antibody combining site. The term " hybrid protein molecule " refers herein to a molecule which contains a continuous length of amino acid residues comprising a first length corresponding to the fragment which forms an antibody combining site fused to a second length corresponding to an integrin binding polypeptide. The term " fused " refers to the manner in which the fragment and polypeptide are associated and is used to distinguish a chimeric antibody from a polypeptide-antibody conjugate, in which the polypeptide is associated with the antibody molecule by operative attachment. For a hybrid protein molecule, the fusion refers to the peptide bond between adjacent amino acid residues, one residue being adjacent the carboxyl terminal residue of the first length, and the other residue being adjacent the amino terminal residue of the second length.

Although a chimeric antibody normally contains only one integrin-binding polypeptide component fused to each fragment forming an antibody combining site, this is not to be considered a limit. For example, a hybrid protein may contain a fragment that forms an antibody combining site, which is fused to one or more integrin-binding polypeptides. In addition, the included polypeptides may be polymers consisting of either different or different integrin-binding polypeptides, or both.

In preferred embodiments, the integrin-binding polypeptide portion of a chimeric antibody is the same polypeptide described herein as a component of the polypeptide-antibody conjugate.

A hybrid protein may additionally contain more than two lengths, for example when the first and second lengths are followed by a third length of amino acid residues wherein the amino-terminus of the third length is fused to the carboxyl-terminal residue of the second length. The third length may have an amino acid sequence corresponding to a second intergyn-binding polypeptide to a portion of an antibody molecule or to a portion of a third protein molecule unrelated to the first two.

In a preferred arrangement, the third length has an amino acid sequence that corresponds to a portion of the Fc region of an antibody molecule. Usually the hybrid protein resulting in this arrangement has a structure in which the integrin-binding polypeptide comprising the second length is inserted into an intact (natural) heavy chain of an antibody molecule so as to disrupt the natural heavy chain molecule and to produce a first length, which includes a fragment which forms an antibody combining site, and a third length which includes a part of the Fc region.

From the foregoing it can be seen that the location of the second switch length within the amino acid residue sequence of the antibody heavy chain may vary along the length of the Fc part of the natural heavy chain. In preferred arrangements, the second length interrupt site will be closer

J 71 028

SCRF 146.0 from the carboxyl terminus than from the hinge region which forms the amino terminus of the Fc region. More preferably, however, the second length will be fused to the carboxyl terminus of the heavy chain of an antibody molecule. The general structure of the antibody molecules and the terminology used herein to identify the heavy and light chains, the hinge region and the Fc region of the antibody molecules (imu-globulins), are well known. See, for example, Paul et al., &Quot; Fundamental Immunology ", Fourth Printing, Raven Press, NY (1984), pages 7 and 150. The combination of an antibody combining site and a binding polypeptide integrin in the chimeric antibody, exhibits the same uses and benefits described above for a polypeptide-antibody conjugate. The present invention therefore contemplates a chimeric antibody containing at least one hybrid protein molecule with a fragment which forms an antibody combining site fused to at least one integrin binding polypeptide such that the hybrid protein forms an antibody combining site capable of immunoreacting with an antigen on the surface of an adherent cell and thereby inhibiting cell adhesion.

In preferred embodiments, a chimeric antibody contains a polypeptide having an amino acid residue sequence of about 5 to about 50 residues in length, which includes a sequence of the formula: -RGD-, and more preferably includes residues represented by formula: -GRGDSP-.

A preferred chimeric antibody further contains a part of a fragment which forms an antibody combining site of an antibody molecule whose primary sequence of amino acid residues corresponds to the sequence of amino acid residues of a monoclonal antibody described herein. Examples of monoclonal antibodies are shown in Table 1. A preferred chimeric antibody contains a fragment which forms an antibody combining site whose amino acid residue sequence corresponds to 71,028

SCRF 146.0 sequence of amino acid residues present in the monoclonal antibody Mab 142A. The chimeric antibody preparation of the present invention can be carried out using well known processes involving the cloning of immunoglobulin genes, the subsequent manipulation of the cloned genes to incorporate sequences encoding integrin-binding polypeptides, introduction of the genes engineered into a expression vector and in a suitable expression medium, and the production of the resulting hybrid protein molecules, to form chimeric antibodies. The above-mentioned preparation steps are the routine of recombinant DNA cloning and the methods of manipulation and expression, and are described in more detail below. The inventive aspect of a chimeric antibody is not in the chimeric antibody preparation techniques but in the resulting manipulated gene and its expressed antibody molecule having a combination of integrin binding polypeptide fused to an antibody combining site which immunoreacts with an antigen from the surface of an adherent cell. The cloning of genes encoding the immunoglobulin protein and their manipulation to form recombinant DNA molecules encoding antibody combining sites fused to other heterologous protein fragments generally constitute processes known in the art. Once prepared, these recombinant DNA molecules are introduced into an expression medium that produces constructed antibody molecules capable of immunoreacting with the same antibody specificity produced by the antibody producing cells from which the genes encoding the immunoglobulin protein have been isolated. See, for example, Roberts et al., Pro-tein Engineering, 1: 59-65 (1986); Morrison, Science, 229: 1202-07 (1985), U.S. Patent No. 4,474,893 and published patent applications EP 0125023, EP 0239400 and WO 89/00999.

In order to isolate a gene encoding the immunoglobulin protein in the preparation of a chimeric antibody, the gene of any cell producing an antibody molecule can be obtained which immunoreacts with an antigen on the surface of an adherent cell. A preferred cell is a hybridoma cell. Hybridomas may be prepared as described herein, or obtained from commercial sources. Examples of hybridomas are shown in Table 1.

As examples of general methods of recombinant DNA cloning, see Molecular Cloning, Maniatis et al., Cold Spring Harbor Lab. N.Y., 1982; DNA Cloning, Glover ed .. IRL Press, McLean, VA (1985). For genomic cloning and expression of immunoglobulin genes in lymphoid cells, see Neuberger et al., Nature, 312: 604-8 (1984); Ochi et al., Proc. Natl. Acad. Know. USA, 80: 6351-55 (1987); and Oi et al., Proc. Natl. Acad. Know. USA, 80: 825-29 (1983). For the cloning of immunoglobulin genes from hybridoma cells and their expression in Xenopus oocytes, see Roberts et al. Protein Engineering, 1: 59-65 (1986) and Wood et al., Nature, 314, 446-9 (1985) for expression in yeast.

A DNA segment is included as part of the engineered recombinant DNA molecule and that segment has a nucleotide sequence encoding at least one polypeptide whose amino acid residue sequence corresponds to the amino acid residue sequence of an integrin binding polypeptide of the present invention.

A DNA segment may further encode a fragment which forms an antibody combining site. The nucleotide sequence of the DNA segment may correspond to the sequence of a fragment which forms an antibody combining site of an immunoglobulin protein, cloned as described above, or the sequence may correspond, for example, to sequences known per se . Examples of nucleotide sequences encoding a fragment which forms an antibody combining site are described by Ka-bat et al. In " Sequences of Proteins of Immunological Interest ", 4.9 Edition, National Institutes of Healht, Bethesda, MD 1987). The DNA segment encoding an integrin binding polypeptide, a fragment which forms an antibody combining site, or both, may be synthesized by chemical techniques, such as, for example, the phosphotriester method of Mattencci et al., J. Am. Chem. Soc., 103: 3185 (1981). Once prepared, the segment is included in the recombinant DNA molecule being manipulated

71 028 SCRF 146.0 to encode and express the entire hybrid protein molecule. The present invention therefore contemplates a recombinant DNA molecule encoding the hybrid protein described herein and is useful in the preparation of a chimeric antibody of the present invention.

Thus, one of the arrays of the present invention contemplates a recombinant DNA molecule encoding a hybrid protein molecule comprising a first segment of DNA encoding a fragment that forms an antibody combining site operably linked in phase to a second DNA segment which encodes an integrin binding polypeptide such that the recombinant DNA molecule is capable, when present in an appropriate expression vector, of expressing a hybrid protein molecule having an antibody combining site that immunoreacts with a surface antigen adherent cell.

A DNA segment, as generally understood and used herein, refers to a molecule comprising a linear extent of nucleotides where the nucleotides are present in a sequence encoding, through the genetic code, a molecule comprising a linear sequence of residues of amino acids referred to as protein, protein fragment or polypeptide.

A particularly useful DNA segment is characterized in that it comprises a DNA sequence encoding (1) a fragment which forms an antibody combining site, or (2) a DNA sequence encoding an integrin binding polypeptide. The term " recombinant DNA molecule " as used herein, refers to a molecule which contains two or more segments of DNA operably linked to produce a third segment of DNA. Thus, a recombinant DNA molecule (rDNA) is a hybrid DNA molecule comprising at least two nucleotide sequences (DNA segments) which are not normally found in nature. The " operably linked " indicates that the two joined segments of DNA are linked by a normal phosphoryl bond

fodiyester normally found between residues of adjacent nucleotides in a DNA sequence. When each of the two joined segments of DNA encodes a sequence of amino acid residues of interest, the expression will further indicate that the linkage maintains the reading frame of the joined segments of DNA so that the two coded sequences of amino acid residues can be translated from the hybrid DNA molecule in phase as a single major hybrid protein.

A recombinant DNA molecule of the present invention contains at least one DNA segment encoding a fragment that forms an antibody combining site operably linked in phase to a DNA segment encoding an integrin binding polypeptide to form a DNA segment which encodes the hybrid protein. Typical recombinant DNA molecules also contain a vector operably linked to the DNA segment encoding the hybrid protein. The term " vector " as used herein refers to a DNA molecule capable of autonomous replication in a cell and to which a further segment of DNA may be operatively linked so as to replicate the linked segment. Vectors capable of directing the expression of a DNA segment encoding a hybrid protein are referred to herein as " expression vectors ". The recombinant DNA molecule having a DNA segment encoding a hybrid protein operably linked to a DNA segment containing an expression vector provides a system for expressing a hybrid protein product in a medium compatible with the included expression vector. The choice of the vector to which the recombinant DNA molecule of the present invention is operably linked depends directly, as is well known in the art, on the desired functional properties, for example the expression of the protein, and the host cell to be transformed, these being , limitations inherent in the art of constructing and expressing recombinant DNA molecules. However, the vector contemplated by the present invention is at least capable of directing replication, and preferably also expression, of the DNA segment encoding the hybrid protein included in the recombinant DNA molecule SCRF 146.0 to which it is operatively connected.

In preferred embodiments, the vector contemplated by the present invention includes a prokaryotic replicon, i.e., a DNA sequence capable of directing the autonomous replication and maintaining the recombinant DNA molecule extrachromosomally in a prokaryotic host cell such as a host cell of bacterial nature, transformed with it. Such replicons are well known in the art. In addition, such arrays that include a prokaryotic replicon also include a gene whose expression confers resistance to a bacterial host transformed with it. Typical bacterial drug resistance genes are those that confer resistance to ampicillin or tetracycline.

Vectors including a prokaryotic replicon may also include a prokaryotic promoter capable of directing expression (transcription and translation) of the DNA segment encoding a hybrid protein in a bacterial host cell, such as E. coli, transformed simultaneously. A promoter is an expression control element formed by a DNA sequence that allows polymerase binding and RNA transcription to occur. Promoter sequences compatible with bacterial hosts are usually provided on plasmid vectors containing restriction sites suitable for insertion of a DNA segment of the present invention. Plasmid vectors are pUC8, pUC9, pBR322 and pBR329, available from Biorad Laboratories (Richmond, CA) and pPL and pKK223 available from Pharmacia (Piscataway, Nj).

Expression vectors compatible with eukaryotic cells, preferably those compatible with vertebrate cells may also be used to form recombinant DNA molecules of the present invention. Eukaryotic cell expression vectors are well known in the art and are available from a number of commercial sources. Typically these vectors are provided containing restriction sites suitable for insertion of the desired DNA segment. Typical vectors are pSVL and pKSV-10 (Pharmacia), pBPV-1 / pML2d (International Biotechnologies, Inc., New Haven, CT) and pTDT1 (ATCC, # 31255). 71 028

In preferred embodiments, the eukaryotic cell expression vectors used to construct the recombinant DNA molecules of the present invention include a selection marker that is effective in a eukaryotic cell, preferably a drug resistance selection marker. A preferred drug resistance marker is the gene whose expression results in resistance to neomycin, i.e. the neomycin phosphotransferase (neo) gene. Southern et al., J. Mol. Appl. Genet., 1: 327-341 (1982). The use of retroviral expression vectors to form the rDNAs of the present invention is also contemplated. The term " retroviral expression vector " used herein refers to a DNA molecule which includes a promoter sequence derived from the long terminal repeat region (LTR) of a retrovirus genome.

In preferred embodiments, the expression vector is usually a retroviral expression vector which is preferably incapable of replication in eukaryotic cells. The construction and use of retroviral vectors has been described by Sorge et al., Mol. Cell. Biol., 4: 1730-37 (1984). The present invention also contemplates RNA molecules equivalent to the above-described recombinant DNA molecules.

EXAMPLES

The following examples are intended to illustrate but not limit the present invention. 1. Preparation of Monoclonal Antibodies The hybridoma cell line 1418, deposited with the ATCC as HB 9118, produces a monoclonal antibody (Mab) which immunoreacts with the disialoganglioside GD2. The hybridoma cell line 142A was isolated as one of the isotype variance variants of a 1418 hybridoma culture using fluorescence activated cell (FACS) aorting techniques described by Kipps et al., Et al., 71,028

SCRF 146.0-37. J. Exp. Med., 161: 1-17 (1985). Hybridoma 142A produces a monoclonal antibody which also immunoreacts with GD2. Hybridoma 1418 and FACS-isolated hybridomas were then characterized to determine the isotype of their produced antibodies using the following isotyping ELISA assay.

In each well of a polyvinyl chloride flat bottom microtiter plate (Dynatech, Alexandria, VA), 50 microliter (1: 1000 dilutions in PBS) volumes of IgG1, IgG2a, IgG2b, IgG3 or IgM (Southern Biotechs Associates, Birmingham, AL), rabbit anti-mouse. The plates were then held overnight at 37 ° C in a drying oven. The dried plates were stored at 4Â ° C until used. Before the ELISA, the dried plates were rehydrated with two washings of two minutes each with wash buffer (PBS 10 millimole / ώΜ /, pH 7.4) containing 0.1% polyoxyethylene / 207-sorbitan monolaurate / Tween 20 / and 0.02% Thimerosal / sodium ethylmercurothysalicylate; Sigma, St. Louis, MO. FACS-isolated hybridoma culture supernatant was diluted 1: 2 in wash buffer containing 0.1% BSA (bovine albumin serum), added to fifty microliters per well and maintained at 4 ° C for 1 hour. After two rinses with wash buffer, 50 microliters (Âμl) of horseradish peroxidase-labeled goat anti-mouse immunoglobulin (Biorad Laboratories, Richmond, CA), diluted at 1: 1000 and held at 4 ° C for 1 hour. The substrate used to evaluate bound peroxidase activity was prepared immediately before the test and consisted of 400 micrograms / ml of o-phenylenediamine (Sigma, St. Louis, MO) in 80 mM citrate-phosphate buffer, pH 6.0, containing 0.23% K2 CO3.

After two final rinses with wash buffer, 50 μl of the substrate solution was added to each well and the color was allowed to develop for 15 minutes in the dark. Color development was stopped by the addition of 25 μl Normal (N) 1 SO 4 to each well, and the optical density at 492 nanometers (nm) was read with the Multiskan ELISA Plate Reader (Bio-Tek Instruments Inc., Burlington, VA). 71 028 SCRF 146.0

38

Isotopic designations were assigned to the culture supernatants produced by the hybridomas, which generated optical densities at 492 nm of at least 10 times the value obtained for a control, non-reactive antibody. By this assay it was found that Mab 1418 was an IgG3 isotype, and Mab 142A was identified as an IgG2a isotype.

Monoclonal antibodies Mab 142A were prepared by introducing a culture of 142A hybridoma cells into the peritoneal cavity of a Balbc / byj mouse (Scripps Clinic Vivarium, La Jolla, CA) and collecting the resulting peritoneal exudate (ascitic tumor fluid) of the mouse by techniques well known in the art. The Mab 142A antibody / molar antibodies present in the recovered ascites fluid were then isolated by affinity chromatography using Protease-A Sepharose (Bio-Rad Laboratories, Richmond, CA) at a ratio of 1.0 milliliter (ml) ml packaged Sepharose beads, according to the methods recommended by the manufacturer, except that the elution buffer was adjusted to pH 5.0 before use. The Sepharose eluant was collected in fractions and the protein content of the fractions was determined using BCA Protein Assay Reagent available from Pierce Chemical Co. (Rockford, IL). Fractions containing protein peaks were pooled to form purified Mab 142A.

In addition to Mab 142A, the following Mabs were used to demonstrate the present invention, and are listed herein with their respective isotypes and sources indicated in parentheses: Mab 9227 (IgG2a, Dr.D. Cheresh, Scripps Clinic and Research Foundation, La Jolla, CA, hereinafter, Scripps), directed against the proteoglycan con-droitin-sulfate of human melanoma cells. Cheresh et al., J. Cell. Biol., 102: 688-696 (1986); Harper et al., J. Immunol, 132: 2096-2104 (1984); and Bumol et al., Proc. Natl. Acad. Know. USA, 79: 1245-1249 (1982) and Mab KS14, a control antibody that does not immunoreact with M21 melanoma cells but immunoreacts with UCLA-P3 cells.

Further purified monoclonal antibodies are prepared by the same method as described above for Mab 142A except that hybridomas 9227 and KS14 were used to form purified Mab 9227 and purified Mab KS14, respectively. 2. Coupling of the Polypeptide to the Monoclonal Antibody

A 3 ml solution containing 14.4 mg of purified Mab 142A in phosphate buffered saline (PBS, pH 7.8), 100 millimolar (mM), was prepared as described in Example 1. 25 microliters (ÂμL) of a 75 mM NH 4 -hydroxysuccinimide N-hydroxysuccinimide ester was added to N-methylpyrrolidone 7 with the solution containing Mab 142A and the mixture was allowed to stir for 3 hours at room temperature to obtain a solution containing activated Mab 142A. The solution containing activated Mab 142A was then loaded onto a Sepharose G-25 column with a bed volume of 1.0 ml and pre-equilibrated with 10 ml, and the protein content of the resulting eluant collected in fractions. The protein-containing peak fractions were pooled to give activated Mab 142A. Dr. MD Pierschbachen (La Jolla Cancer Research Foundation, La Jolla, CA) provided a synthetic polypeptide having the amino acid sequence Cys-Gly-Gly-Ala-Gly-Ala-Gly-Arg-Gly-Asp-Ser-Pro ( CGGAGAGRGDSP) after having been synthesized by Peninsula Laboratories (San Carlos, CA). Ten mg of the polypeptide was mixed as solid substance to the above prepared activated Mab 142A and the mixture was kept under stirring for 3 hours at room temperature to form the reaction mixture. The reaction mixture was then applied to a Sephadex G-25 column with a bed volume of 1.0 ml, primed with 10 ml, and the protein content of the resulting eluant was monitored as above, and the peak fractions containing protein were collected to give a Mab 142A polypeptide conjugate (conjugate-Mab 142A).

Mab-polypeptide conjugates were also prepared by the above procedure using purified Mab 9227 or control Mab KS14 prepared as in Example 1 in place of the purified Mab 142A to give the Mab-9227 conjugate and the control Mab-conjugate, respectively. 71 028 SCRF 146.0 40

The Mab-polypeptide conjugates may also be prepared by the said processes of Examples 1 and 2 using monoclonal antibodies produced by hybridomas 1418, 126, MB3.6, 9227, LM142 or LM609. 3. Binding of the Conjugate from Monoclonal Antibody to Culture Cells M21 human melanoma cells were kindly provided by Dr. DL Morton (University of California, Los Angeles, hereinafter designated UCLA) and grown as a suspension culture in a culture medium. growth / RPMI 1640 containing 2 mM L-glutamine and 50 mg / ml gentamicin sulfate; all from GIBCO Laboratories (Grand Island, NY) containing 10% fetal calf serum (FCS) at 37øC in 7.5% CO 2 / 92.5% air. UCLA-P3 human lung adenocarcinoma cells were supplied by Dr. Martin (UCLA) and developed as culture suspension, as above, for M21 cells.

M21 culture cells were washed twice in normal saline and placed in polyvinyl lysate funnel (Dynatech, Chantilly, VA) microtiter plate at 5 x 10 cells per well in 50 μΐ growth medium containing 10% FCS. The plates were then held at 37 ° C for 12-18 hours to dry the dishes and to obtain dried M21 plates. UCLA-P3 plates were prepared in the same manner using UCLA-P3 culture cells.

The dried cells were rehydrated with two washes, each comprising, first, addition of wash buffer / 10 mM PBS, pH 7.4, 0.1% Tween 20 (polyoxyethylene sorbitan monolaurate), 0.02% Thimerosal (sodium ethylmercurotiosalicylate, then maintaining the plates at room temperature for 2 minutes, and finally removing the liquid by inverting the plates.

The purified Mabs prepared in Example 1 and the Mab conjugates prepared in Example 2 were then each serially diluted to the final concentration shown in Table 2 using 1.10% (v / v) in wash buffer containing 0.1% BSA (bovine albumin serum) and 50æl of each was mixed in each well 71 028 SCRF 146.0

41 of the rehydrated plates to form the immunoreactor mixtures. Plates were held for 1 hour at 4Â ° C in a shaker, emptied of their liquid contents by inversion and shaking, and washed twice as described above to isolate the phase-linked immunoreaction products solid. Then, 50 of a peroxidase-linked goat anti-mouse Ig (Tago, Burlingame, CA) diluted 1: 1000 (v / v) solution in diluent were mixed in each well of the plates, and held for 1 hour at 4 ° C to allow the formation of labeled immunoreaction products. The plates were again emptied by inversion and shaking, washed twice as described above, and 50 μl of freshly prepared chromogenic substrate solution / 400 μg / ml O-phenylene diamine ( Sigma Chemical Co., St. Louis, MO), 80 mM citrate-phosphate buffer, pH 5.0, 0.012% (v / v) H202 (Eastman Kodak, Rochester, NY) The plates were kept in the dark for 30 minutes after which, in each well, 15 .mu.l of H2 SO4. 4N were mixed. The optical absorbance at 492 nm (%) of the resulting solutions in each well was then measured with a Multiscan ELISA plate reader (Flow Laboratories, McLean, VA).

Results of the purified (" Free ") and Mab-conjugated (" Conjugate ") Mab binding studies by immunoreaction with surface anabancens from adherent cells present in M21 and UCLA-P3 cells are shown in Table 2 , below. TABLE 2

Binding of Monoclonal Polypeptide / Antibody Conjugates to Cells A. Cell M21

Free to

Conjugations b

/ Mab? 0 9227 142A

Control 9227 142A Control 5.01.0 0.5 1,171 1,995 1,146 1,602 1,142 1,590 0,147 1,088 1,687 0,994 0,114 0,985 1,758 0, -951 0,071 1,023 1,786 0,963 42 71 028 SCRF 146.0 0.1 1,101 1,402 0,032 0,929 1,471 0,554 0,05 0,510 0,105 0 0.408 0.078 0 0.005 0.261 0.015 0 0.121 0.017 0 0 0 0 0 0 0 0 B. Conjugated UCLA-P3 Cells * 3 / Mab / C 9227 142A Control 9227 142A Control 5.0 0.090 0.256 1.754 0.234 0.345 1.858 1.0 0.093 0.116 2.192 0.122 0.174 2.076 0.5 0.077 0.149 2.127 0.126 0.155 1.996 0.1 0.022 0.081 1.815 0.070 0.085 2.047 0.05 0.030 0.085 0.241 0.001 0.114 0.335 0.005 0.056 0.042 0.103 0.050 0.070 0.148 0 0.050 0.037 0.021 0.013 0.034 0.017 a " Free " designates the use of the indicated purified Mabs (Mab 9227,

Mab 142A or Control Mab) prepared as described in Example 1 and free of the polypeptide conjugates. " Conjugate " designates the use of the same Mabs indicated for " Free " except that the Mabs were operably linked (conjugated) to the CGGAGAGRDSP polypeptide as described in Example 2. YMab / 1 'indicates the final protein concentration of the diluted serial preparation of Mab (free) or Mab- Conjugate (conjugate) Table 2 indicates that Mab 9227, both free and conjugated, were able to immunoreact with the chondroitin sulfate proteoglycan present in M21 cells, such as Mab 142A preparations capable of immunoreacting with GD2 ganglioside in same cells. For both these Mabs, the immunoreaction was only slightly decreased when the Mab-conjugate was used, as compared to the free Mab. Similarly, both the conjugated and the free control Mab KS14 were able to immunoreact with the target tumor present in UCLA-P3 cells. These results indicate that the operative attachment of a binding polypeptide to an antibody molecule by the described binding processes does not significantly interfere with the ability of the Mabs-conjugates to immunoreact with their target antigen when present on target surfaces. adherent cells.

Control Mab KS14 that has not been immunoreacted with M21 cells has been found to bind to M21 cells following polypeptide conjugation. Because this increased binding occurred only after formation of a Mab-conjugate, binding is believed to result from an interaction between the RGD-containing polypeptide present in the Mab-conjugated and the RGD-directed adhesion receptors present in the M21 cells. It is not known why the Mab 9227 or Mab 142A conjugates do not also show an increase in binding to UCLA-P3 by the above mechanism proposed for Mab KS14 on M21 cells. However, UCLA-P3 cells may have significantly fewer RGD-directed adhesion receptors than the M21 cells measured by the assay. 4. Monoclonal Antibody Conjugates Inhibit Cell Adhesion In Vitro The following cell adhesion assay was performed to characterize the ability of monoclonal antibody (Mab) conjugates to inhibit cell adhesion.

M21 cells were grown in suspension culture in RPMI 1640 growth medium containing 10% fetal calf serum (FCS) at 37øC with 7.5% CO 2 / 92.5% air. These cells were metabolically labeled in leucine-free growth media containing 50 microcurie (ÂμCi) / ml3H-leucine (ICN, Irvine, CA) for 72 hours at 37Â ° C. The labeled cells were then washed by centrifugation in 1% FCS growth medium to remove unincorporated H-leucine-labeled 3 M21 cells. 3 3

Approximately 5x10 H-leucine-labeled M21 cells were resuspended in 100 μl of the growth medium containing the various concentrations of a Mab-conjugate prepared as described in Example 2 to form an immunoreaction mixture. As controls, labeled cells were resuspended alone in growth medium, or a control antibody was used. Control mAb i known as activated Mab was prepared as described in Example 2.

The mixture was then held for 1 hour at 4 ° C to allow the formation of Mab-M21 cell immunoreaction products.

The cells were then washed in two cycles, first centrifugation at low speed at 400 x g, followed by resuspension of the resulting pelletised using a growth medium containing 1% FCS. After washing, the cells were resuspended in growth medium to form antibody treated M21 cells.

Microtiter plate wells of polystyrene (96 wells, Flow Laboratories, McLean, VA) were coated with matrix proteins (adhesion substrates) by mixing PBS solutions (pH 7.2) containing 5æg / ml of a substrate of adhesion, to each well. The wells were maintained overnight at 25 ° C to allow the adsorption of proteins in the wells. The matrix proteins used included fibrinogen provided by Dr. E. Plow (Research Institute of Scripps Clinic, La Jolla, CA, hereinafter referred to as RISC); von Willebrand factor, supplied by Drs Z. Ruggeri and T. Zimmerman (RISC); and human fibronectin, provided by Dr. M. Pierschbacher (La Jolla Cancer Research Foundation, La Jolla, CA, hereinafter referred to as LCRF).

Fifty æl of growth medium containing 3x10 10 antibody treated M21 cells was mixed in a well coated substrate to form a reaction-adhesion mixture. The blends were kept at 37øC in a humidified incubator for 20 minutes to allow cell adhesion to occur, then the plates were inverted to remove the growth medium and unbound cells. A set of two plates were prepared analogously, which were maintained during a 90 minute adhesion period. All wells were washed twice with 150 ml PBS (pH 7.2) to ensure removal of loose cells. The remaining fixed cells were collected by coupling 100 μg Trypsin / EDTA (1 X; Gibco Laboratories, Grand Island, NY) to each well, incubating the cells at 37 ° C for 30 minutes, removing , then the cells and collecting them in glass fiber filters using a Skatron automated re-harvester (Skatron Instruments, Sterling, VA) according to the instructions issued by the manufacturer. The fiber filter discs were placed into small vials containing 3 ml of " cocktail " of liquid scintillation and the amount of radioactive label present, in counts per minute, was determined in a liquid scintillation counter. The results were expressed as the total number of cells (count per minute) that adhered (bound) to the matrix protein at the indicated concentrations of added Mab-con-jugate or control Mab.

As seen in Figures 1 and 2, adhesion of M21 cells was inhibited by Mab 142A conjugated either for 20 minutes (Figure 1) or for 90 minutes (Figure 2) of the adhesion period, when either the adhesion substrate was used von Willebrand factor (vWF) or fibrinogen (Fb). Inhibition at concentrations below 2æg conjugated Mab per ml was observed. The adhesion inhibition potential (IC50) can be expressed as the concentration of conjugated Mab sufficient to cause a 50% decrease in adhesion, as compared to the adhesion obtained without addition of conjugated Mab. Thus the ICgQ for the conjugated Mab 142A, during a 20 minute adhesion time, was 4 μg / ml for vWF adhesion and 0.5 μg / ml for Fb adhesion. Similarly, the IC, -q for the conjugate, during a 90 minute adhesion time, was 15 æg / ml for vWF adhesion and 0.5 æg / ml for Fb adhesion.

Using the in vitro cell adhesion assay essentially described above, the soluble polypeptide CGGAGAGRGDSP from Example 2, at various concentrations, was coupled to the cell suspensions in place of the conjugated Mab 142A. The results of these adhesion measurements are expressed as potential inhibition of adhesion (IC 50) in Table 3 below. TABLE 3

Effect of CGGAGAGRGDSP, Soluble or Conjugated with Mab 142A, on binding of the M21 Cell to the Fibrinogen, von Willebrand Factor or Fibronectinab Mab 14 2 A TEMPOa VWF SOLUBLE (μΜ), CONJUGATE (μΜ) 20 min. 10 0.24 90 min. 55 0.91 Fibrinogen 20 min. N / T ° N / T 90 min. 2.7 0.03 Fibronectin 20 min. 106 > 18 90 min. > 500 > 18

Period of time in which cells were maintained for adhesion to occur. The effect on cell attachment is expressed in micromolar (ÂμM) concentration of polypeptide present in the solution or on the conjugate, sufficient for a 50% decrease in adhesion as compared to adhesion without polypeptide. c N / T indicates " not tested ". Table 3 shows that for each adhesion time, 20 or 90 minutes, and for all adhesion substrates tested, there was a significant reduction in the concentration of polypeptide required to inhibit the adhesion of guando cells the polypeptide was conjugated.

Thus, although RGD-containing polypeptides and other polypeptides derived from adhesion proteins are known to have the ability to inhibit cell adhesion, the results in Table 3 show that the potential for inhibition of adhesion of such polypeptides has been enhanced by the conjugation to an antibody which immunoreacts with an anion from the surface of an adherent cell. The foregoing specification, including the specific arrangements and examples, is intended to illustrate the present invention and not to limit it. Many other variations and modifications can be made without departing from the true spirit and scope of the present invention.

Claims (13)

71 028 Ψ A method of obtaining an RGD antibody comprising an antibody molecule which is capable of immunoreacting with an adherent cell surface antigen operably linked to an integrin binding polypeptide comprising a sequence of residues of amino acid having a length of about 5 to about 50 residues, including a sequence having the formula: -RGB-, characterized in that it comprises covalently linking said antibody molecule with said polypeptide. 2. A process according to claim 1, wherein said polypeptide comprises a sequence having the formula: -GRGDSP- or -CGGAGAGRGDSP-. A method according to claim 1, wherein said antibody molecule is capable of immunoreacting with an adherent cell surface antigen selected from the group consisting of GD2 disialoganglioside, GD3 disialoganglioside, chondroitin sulfate proteoglycan, vitronectin receptor and endothelial cell receptor. A method according to claim 1, wherein said antibody molecule is selected from the group of monoclonal antibodies consisting of 1418, 142A, 126, MB3.6, ΙΜ609 and LM142. A method according to claim 1, wherein said antibody molecule contains 2 to about 30 polypeptides operably linked by said covalent bond to said antibody molecule. A method according to claim 1, characterized in that said antibody molecule is operably linked by said covalent bond, through a polypeptide spacer, having a length of from 2 to about 50 amino acid residues. 71 028 SCRF 146.0 - 49 - 7. A method of preparing an RGD antibody comprising an antibody molecule that is capable of immunoreacting with a cell surface antigen which is selected from the group of monoclonal antibodies consisting of 1418, 142A, 126, MB3.6, 11C64, R24, LM609 and LM142 and which is operably linked to an integrin binding polypeptide having the amino acid residue sequence CGGAGAGRGDSP, which comprises covalently linking said polypeptide to said antibody molecule in a ratio of about 10 polypeptides per antibody molecule, by means of an NHS linker molecule. A method of obtaining a YIGSR antibody comprising an antibody molecule which is capable of immunoreacting with an adherent cell surface antigen operably linked to a laminin receptor binding polypeptide comprising a sequence of amino acid residues with a length of about 5 to about 50 residues, including a sequence having the formula: - YIGSR-, characterized in that it comprises covalently attaching said antibody molecule to said polypeptide. A method according to claim 8, wherein said polypeptide has the formula: YIGSR, CDPGYIGSR, or RGDSGYIGSR. A method of preparing a chimeric antibody containing at least one hybrid protein molecule having an antibody combining site-forming fragment fused to at least one integrin binding polypeptide, said protein molecule hybridizes to an antibody combining site which immunoreacts with an adherent cell surface antigen which comprises fusing said hybrid protein molecule with said polypeptide by means of a peptide bond. - 50 - 71 028 SCRF 146.0 11. A process according to claim 10, wherein said integrin-binding polypeptide comprises an amino acid residue sequence having a length of about 5 to about 50 residues, including a sequence having the formula -RGD - or -GRGDSP-. A method of preparing a recombinant DNA molecule encoding a hybrid protein molecule, comprising: a) a first DNA segment encoding a antibody-combining site-forming fragment, and b) a second DNA fragment encoding a integrin polypeptide which is operably linked in phase to the first segment, said recombinant DNA molecule being capable, when present in an appropriate expression vector, of expressing a hybrid protein molecule having a combination site with which antibody immunoreacts with an adherent cell surface antigen, characterized in that it comprises the binding of the first and second DNA segments in phase. A process for the preparation of a pharmaceutical composition comprising mixing the antibodies obtained by the process of claims 1 to 11 with excipients, pharmaceutically acceptable and compatible with the active ingredient. Lisbon, ', 6. Htó 1990 By SCRIPPS CLINIC AND RESEARCH FOUNDATION - 0 OFFICIAL AGENT -