Field of the Invention This invention relates to methods and compositions for modulating cell adhesion and for inhibiting the interaction between integrins and their ligands. In particular, the invention relates to compounds that selectively inhibit a 1ß1 integrins.

Background of the Invention Integrins are a family of cell surface proteins that mediate adhesion between cells (cell-cell adhesion) and between cells and the extracellular matrix (cell-ECM adhesion). The known integrins are heterodimeric proteins composed of noncovalently bound a and P subunits. In humans there are at least 15 different a and eight different 6 subunits, which can combine to form unique integrins with diverse biological activities and ligand specificities.

The integrins mediate cell-cell and cell-extracellular matrix interactions by binding adhesive ligands carried by cells or found in the extracellular matrix.

Examples of adhesive ligands include Fibrinogen, fibronectin, collagen I, collagen IV,

and vascular cell adhesion molecule-1 (VCAM-1). The integrins play an important role in many diverse biological processes, including platelet-aggregation, tissue repair, angiogenesis, bone destruction, tumor invasion, inflammation, restenosis of the arteries following surgery or angioplasty, and immune reactions. Furthermore, integrin binding activity has been implicated in a number of disease states, including coronary thrombosis, atherosclerotic diseases, vascular disease, heart disease, diabetes, multiple sclerosis, rheumatoid arthritis, ulcerative colitis, arteriosclerosis, asthma, and autoimmune disorders. Consequently, integrins are important targets for therapeutic intervention in human disease and for regulation of normal biological processes.

The majority of integrins identified to date are RGD-dependent integrins which bind to a three amino acid RGD (arginine-glycine-aspartic acid) sequence found in their respective adhesive ligands. The aIIbß, avß3 and a5pl integrins are all examples of RGD-dependent integrins. Integrin aIlbp3 binds fibrinogen on the surface of platelets and mediates platelet-aggregation. Integrin avp3 is predominantly expressed on endothelial cells (where it is involved in angiogenesis) and osteoclasts (where it participates in bone destruction). Integrin a5pl is expressed by a variety of cell types and is involved in cell adhesion to the extracellular matrix as well as in the formation of tissues and organs during embryonic development.

Some integrins bind to sequences other than RGD. Such integrins are classified as RGD-independent integrins. An important RGD-independent integrin is the alp 1 integrin. This integrin is expressed by a variety of cell types and is involved in angiogenesis, vascularization of tissues, and lymphocyte migration. This integrin is also involved in the formation of basement membranes and in the interaction of cells with these membranes. Additionally, activation and up-regulation of cclpl integrin expressed by lymphocytes or macrophages is believed to play a significant role in the progression of many disease states and biological processes, including insulin dependent diabetes mellitus, multiple sclerosis, rheumatoid arthritis, ulcerative colitis, arteriosclerosis, asthma, allergy, organ rejection, restenosis of arteries after surgery or angioplasty, and angiogenesis.

Angiogenesis is the process in which new blood vessels grow into an area which lacks a sufficient blood supply. Angiogenesis commences with the erosion of the basement membrane surrounding endothelial cells and pericytes forming capillary blood vessels. Erosion of the basement membrane is triggered by enzymes released by endothelial cells and leukocytes. The endothelial cells then migrate through the eroded basement membrane when induced by angiogenic stimulants. The migrating cells form a"sprout"off the parent blood vessel. The migrating endothelial cells proliferate, and the sprouts merge to form capillary loops, thus forming a new blood vessel.

Angiogenesis can occur under certain normal conditions in mammals such as in wound healing, in fetal and embryonic development, and in the formation of the corpus luteum, endometrium and placenta. Angiogenesis also occurs in certain disease states such as in tumor formation and expansion, or in the retina of patients with certain ocular disorders. Angiogenesis can also occur in a rheumatoid joint, hastening joint destruction by allowing an influx of leukocytes with subsequent release of inflammatory mediators.

The evidence for the role of angiogenesis in tumor growth was extensively reviewed and present by O'Reilly and Folkman in U. S. Pat. 5,639,725, the entire disclosure of which is incorporated herein by reference. It is now generally accepted that the growth of tumors is critically dependent upon this process.

The adhesive ligand of alp 1 integrin is collagen IV. To date, the sequence on collagen IV to which the a 1 (31 integrin binds has not been identified.

The disintegrins are a family of low molecular weight, cysteine-rich peptides that interfere with the binding of integrins to their adhesive ligands. Disintegrins carry sequences identical or analogous to the binding sites in the adhesive ligands of integrins, and bind integrins with an affinity comparable to that of monoclonal antibodies. Many disintegrins have been isolated from the venom of various snakes, as well as other sources, and several disintegrin subfamilies have been identified.

These disintegrin subfamilies differ from each other on the basis of peptide chain length, number of conserved cysteines, dimerization state and type of integrin binding

site (reviewed in McLane et al., P. S. E. B. M. 219: 109-119 (1998)). Disintegrins containing the RGD sequence prevent the binding of RGD-dependent integrins to their adhesive ligands, presumably through a competition-type mechanism. For example, the RGD motif is present in a biologically active fragment of the short disintegrin echistatin. This fragment prevents the RGD-dependent allbp3 integrin from associating with the RGD sequence in its adhesive ligand fibrinogen.

Because disintegrins interfere with the binding of integrins to their adhesive ligands, they may potentially be used in the prevention and treatment of diseases involving integrin binding. Additionally, disintegrins may be used to affect normal integrin-mediated biological processes. However, the vast majority of disintegrins described to date contain an RGD sequence, or a conservative substitution of one of the residues present in the RGD sequence (such as the KGD, or lysine-glycine- aspartic acid, sequence of the disintegrin barbourin). Thus, only diseases or biological processes involving RGD-dependent integrins can be affected with these disintegrins.

In contrast, the treatment of diseases or the modulation of biological processes involving RGD-independent integrins has heretofore been difficult. As discussed above, many disease states and biological processes are associated with RGD- independent integrins such as the alp 1 integrin. Consequently, there is a need for potent and specific inhibitors of alp 1 integrin. Such inhibitors can be used in the treatment of disease or the modulation of biological processes involving the alpl integrin.

Summary of the Invention The invention provides compounds which are potent and specific inhibitors of alpl integrin. The alpl integrin-inhibiting compounds of the invention include the peptide obtustatin as well as fragments, derivatives, homologs and analogs thereof.

Thus, in one aspect, the invention provides a substantially purified peptide of SEQ ID NO : 1, or a biologically active fragment, derivative, homolog or analog thereof. The compounds of the invention may also be modified, for example with a label or a targeting group.

According to one embodiment, the compounds of the invention are biologically active peptides comprising the sequence of Xl-Ser-Leu-X2 wherein XI is from zero to twenty-five amino acids, and X2 is from zero to twenty amino acids, and wherein said compound optionally comprises an amino-terminal and/or carboxy-terminal protecting group.

In one embodiment, Xi is (i) zero amino acids, or (ii) the segment Cys-Thr-Thr-Gly-Pro-Cys-Cys-Arg-Gln- Cys-Lys-Leu-Lys-Pro-Ala-Gly-Thr-Thr-Cys-Trp-Lys-Thr, or an amino-terminal truncation fragment thereof containing at least one amino acid, and

X2is (i) zero amino acids, or (ii) the segment Thr-Ser-His-Tyr-Cys-Thr-Gly-Lys-Ser-Cys- Asp-Cys-Pro-Leu-Tyr-Pro-Gly, or a carboxy-terminal truncation fragment thereof containing at least one amino acid.

Preferably, Xi is Lys-Thr. More preferably, Xi is Thr.

In another embodiment, the compounds of the invention are biologically active peptides comprising the sequence Cys-Xaa-Xaa-Xaa-Xaa-Cys-Cys-Xaa-Xaa- Cys-Xaa-Xaa-Xaa-Xaa-Xaa-Xaa-Xaa-Xaa-Cys-Xaa-Xaa-Xaa-Ser-Leu- Xaa-Xaa-Xaa- Xaa-Cys-Xaa-Xaa-Xaa-Xaa-X3-Xaa-Cys-X4, where X3 is any amino acid; and each Xaa is independently any amino acid; and X4 is zero, 1,2,3,4, or 5 amino acids and wherein said compound optionally comprises an amino-terminal and/or carboxy-terminal protecting group.

In a preferred embodiment, X3 is Cys.

In another aspect, the invention is directed to a composition comprising an al (31 integrin-inhibiting compound and a pharmaceutically acceptable carrier. In preferred embodiments, the composition comprises the peptide of SEQ ID NO : 1, or a biologically active fragment, derivative, homolog or analog thereof and a pharmaceutically acceptable carrier.

In another aspect of the invention, there is provided a method of inhibiting the binding of alp 1 integrin to their adhesive ligands, comprising contacting a sample with an effective amount of al (31 integrin-inhibiting compound so that the binding of

al (31 integrin to their adhesive ligands is inhibited. The sample may comprise α1ß1 integrin not bound to a cell membrane or cells expressing an alp 1 integrin.

In a further aspect of the invention, there is provided a method of treating diseases or biological conditions associated with the binding ofalpl integrin to their adhesive ligands, comprising administering to a subject an amount of alp 1 integrin- inhibiting compound, or a pharmaceutically acceptable salt thereof, sufficient to inhibit the binding of α1ß1 integrin with their ligands.

In another aspect of the invention, there is provided a method of detecting α1ß1 integrin in a sample, comprising: (a) contacting the sample with an al (31 integrin-inhibiting compound modified with a label for a time sufficient to allow binding of the labeled al (31 integrin-inhibiting compound to any al (31 integrin present in the sample; and (b) detecting the labeled al (31 integrin-inhibiting compound bound to the α1ß1 integrin.

The sample may comprise a plurality of unknown peptides, or cells expressing or suspected of expressing al (31 integrin. In one embodiment, the unknown peptides or cells are immobilized on a solid support.

In a still further aspect, the invention provides methods of isolating al (31 integrin from a sample, comprising: (a) contacting the sample with an al (31 integrin-inhibiting compound modified with a selectable label for a time sufficient for the modified α1ß1 integrin-inhibiting compound to bind to any α1ß1 integrin present in the sample; and

(b) separating the selectable label modified algol integrin-inhibiting compound bound to alpl integrin from the sample.

The sample may comprise a plurality of unknown peptides, or cells expressing or suspected of expressing alpl integrin. In a preferred embodiment, the selectable label which modifies the algol integrin-inhibiting compound is fluorescein isothiocyanate and the al (31 integrin expressing cells are isolated by flow cytometry.

The present invention also provides antibodies against the compounds of the invention. The antibodies may be a monoclonal or polyclonal antibody or an antibody fragment that is capable of binding antigen. One aspect of the invention is a hybridoma that produces a monoclonal antibody which specifically binds the compounds of the invention.

Amino Acid Abbreviations The nomenclature used to describe the peptide compounds of the present invention follows the conventional practice wherein the amino group is presented to the left and the carboxy group to the right of each amino acid residue. In the formulae representing selected specific embodiments of the present invention, the amino-and carboxy-terminal groups, although not specifically shown, will be understood to be in the form they would assume at physiologic pH values, unless otherwise specified. In the amino acid structure formulae, each residue is generally represented by a one- letter or three-letter designation, corresponding to the trivial name of the amino acid, in accordance with the following schedule:

A Alanine Ala C Cysteine Cys D Aspartic Acid Asp E Glutamic Acid Glu F Phenylalanine Phe G Glycine Gly H Histidine His I Isoleucine Ile K Lysine Lys L Leucine Leu M Methionine Met N Asparagine Asn P Proline Pro Q Glutamine Gln R Arginine Arg S Serine Ser T Threonine Thr V Valine Val W Tryptophan Trp Y Tyrosine Tyr Definitions The following definitions, of terms used throughout the specification, are intended as an aid to understanding the scope and practice of the present invention.

By"alpl integrin"is meant any integrin comprising an al subunit and a pi subunit.

By"integrin"is meant a family of heterodimeric cell surface proteins which mediate adhesion between cells as well as between cells and extracellular matrix proteins.

The expression"amino acid"as used herein is meant to include both natural and synthetic amino acids, and both D and L amino acids."Standard amino acid" means any of the twenty standard L-amino acids commonly found in naturally occurring peptides."Nonstandard amino acid residues"means any amino acid, other than the standard amino acids, regardless of whether it is prepared synthetically or derived from a natural source. As used herein,"synthetic amino acid"also

encompasses chemically modified amino acids, including but not limited to salts, amino acid derivatives (such as amides), and substitutions. Amino acids contained within the peptides of the present invention, and particularly at the carboxy-or amino- terminus, can be modified by methylation, amidation, acetylation or substitution with other chemical groups which can change the peptide's circulating half life without adversely affecting their activity. Additionally, a disulfide linkage may be present or absent in the peptides of the invention.

Amino acids have the following general structure: Amino acids are classified into seven groups on the basis of the side chain R: (1) aliphatic side chains, (2) side chains containing a hydroxylic (OH) group, (3) side chains containing sulfur atoms, (4) side chains containing an acidic or amide group, (5) side chains containing a basic group, (6) side chains containing an aromatic ring, and (7) proline, an imino acid in which the side chain is fused to the amino group.

"Amino-terminal truncation fragment"with respect to an amino acid sequence means a fragment obtained from a parent sequence by removing one or more amino acids from the amino-terminus thereof.

"Analogs"are small molecule compounds which exhibit one or more biological activity of obtustatin.

The terms"antibodies"or"antibody,"as used herein, refer to intact immunoglobin molecules, as well as fragments of immunoglobulin molecules, such as Fab, Fab', (Fab') 2, Fv, and SCA fragments, which specifically bind to an epitope of those compounds of the invention which are peptides, fragments, derivatives or homologs.

As used herein,"protecting group"with respect to a terminal amino group refers to a terminal amino group of a peptide, which terminal amino group is coupled

with any of various amino-terminal protecting groups traditionally employed in peptide synthesis. Such protecting groups include, for example, acyl protecting groups such as formyl, acetyl, benzoyl, trifluoroacetyl, succinyl, and methoxysuccinyl; aromatic urethane protecting groups such as benzyloxycarbonyl ; and aliphatic urethane protecting groups, for example, tert-butoxycarbonyl or adamantyloxycarbonyl. See Gross and Mienhofer, eds., The Peptides, vol. 3, pp. 3-88 (Academic Press, New York, 1981) for suitable protecting groups.

As used herein,"protecting group"with respect to a terminal carboxy group refers to a terminal carboxyl group of a peptide, which terminal carboxyl group is coupled with any of various carboxyl-terminal protecting groups. Such protecting groups include, for example, tert-butyl, benzyl or other acceptable groups linked to the terminal carboxyl group through an ester or ether bond.

"Carboxy-terminal truncation fragment"with respect to an amino acid sequence means a fragment obtained from a parent sequence by removing one or more amino acids from the carboxy-terminus thereof.

"Derivative"includes any purposefully generated peptide which in its entirety, or in part, has a substantially similar amino acid sequence to obtustatin, and which retains at least one biological property of obtustatin. Preferably, the derivative retains the ability to inhibit the binding of cd integrin to its adhesive ligand.

Derivatives of obtustatin may be characterized by single or multiple amino acid substitutions, deletions, additions, or replacements. These derivatives may include (a) derivatives in which one or more amino acid residues of obtustatin are substituted with conservative or non-conservative amino acids, (b) derivatives in which one or more amino acids are added to obtustatin, (c) derivatives in which one or more of the amino acids of obtustatin includes a substituent group, (d) derivatives in which obtustatin or a portion thereof is fused to another peptide (e. g., serum albumin), (e) derivatives in which one or more nonstandard amino acid residues (i. e., those other than the 20 standard L-amino acids commonly found in naturally occurring proteins) are incorporated or substituted into the obtustatin sequence, and (f) derivatives in

which one or more nonamino acid linking groups are incorporated into or replace a portion of obtustatin.

"Fragment"refers to a portion of the obtustatin sequence comprising at least two amino acid residues. Fragments may be generated by amino-terminal truncation, carboxy-terminal truncation or both of these. Fragments may also be generated by chemical or enzymatic digestion.

"Homolog"includes any nonpurposely generated peptide which in its entirety, or in part, has a substantially similar amino acid sequence to obtustatin and exhibits at least one biological activity of obtustatin. Homologs may include paralogs, orthologs, or naturally occurring alleles of obtustatin. Preferably, the biological activity exhibited by obtustatin homologs is the inhibition of al 1 integrin- mediated cellular adhesion to collagen IV.

By"label"is meant any substance which may be incorporated into or conjugated to a compound, by chemical bonds or any other means, and which may be detected.

By"libraries"is meant pools and subpools of pro-analogs.

"Peptide"and"protein"are used interchangeably, and refer to a compound comprised of amino acid residues covalently linked by peptide bonds. A protein or peptide must contain at least two amino acids and no limitation is placed on the maximum number of amino acids which may comprise a protein's or peptide's sequence. The amino acids of the peptides described herein and in the appended claims are understood to be either D or L amino acids with L amino acids being preferred.

"Pharmaceutically acceptable"means physiologically tolerable, for either human or veterinary application.

As used herein,"pharmaceutical compositions"include formulations for human and veterinary use.

"Pro-analogs"are compounds which may potentially be obtustatin analogs.

Pro-analogs are regarded as analogs once it is determined that they exhibit one or more biological activities of obtustatin.

"RGD-independent integrins"are integrins which bind to amino acid sequences other than RGD.

"RGD-dependent integrins"are integrins which bind to the RGD amino acid sequence.

By"selectable label"is meant any substance which may be used to label the compounds of the invention and which may be selectively removed from a sample.

By"specifically bind,"as used to describe the interaction between an antibody and another molecule, is meant that the two bind to each other with greater affinity than to other, non-specific molecules.

As used herein, a peptide or a portion of a peptide which has a"substantially similar amino acid sequence"to obtustatin means the peptide, or a portion thereof, has an amino acid sequence identity or similarity to obtustatin of greater than about 30%, preferably greater than about 60%, more preferably greater than about 80%, and most preferably greater than about 90%. Amino acid sequence similarity or identity may be computed by using the BLASTP and TBLASTN programs which employ the BLAST (basic local alignment search tool) 2.0.14 algorithm; BLASTP and TBLASTN settings to be used in such computations are indicated in Table 1 below.

Amino acid sequence identity is reported under"Identities"by the BLASTP and TBLASTN programs. Amino acid sequence similarity is reported under"Positives" by the BLASTP and TBLASTN programs. Techniques for computing amino acid sequence similarity or identity are well known to those skilled in the art, and the use of the BLAST algorithm is described in Altschul et al., J Mol. Biol. 215: 403-10 (1990) and Altschul et al., Nucleic Acids Res. 25: 3389-3402 (1997), the disclosures of which are herein incorporated by reference in their entirety. BLASTP and TBLASTN programs utilizing the BLAST 2.0.14 algorithm and may be accessed at http ://www. ncbi. nlm. nih. gov/.

Table 1-Settings to be used for the computation of amino acid sequence similarity or identity with BLASTP and TBLASTN programs utilizing the BLAST 2.0.14 algorithm. Expect Value 10 Filter Low complexity filtering using SEG program* Substitution Matrix BLOSUM62 Gap existence cost 11 Per residue gap cost 1 Lambda ratio 0. 85 Word size 3

*The SEG program is described by Wootton and Federhen, Comput. Chem. 17: 149- 163 (1993).

"Substantially purified"refers to a population of peptides or cells which is substantially homogenous in character due to the removal of other compounds (e. g., other peptides, nucleic acids, carbohydrates, lipids) or other cells originally present.

"Substantially purified"is not meant to exclude artificial or synthetic mixtures with other compounds, or the presence of impurities which do not interfere with biological activity, and which may be present, for example, due to incomplete purification, addition of stabilizers, or formulation into a pharmaceutically acceptable preparation.

"Synthetic mutant"includes any purposefully generated mutant derived from obtustatin. Such mutants may be purposefully generated by, for example, chemical mutagenesis, polymerase chain reaction (PCR) based approaches, or primer based mutagenesis strategies well known to those skilled in the art.

Brief Description of the Figures FIGURE 1 is the elution profile for Viper lebetina obtusa venom proteins applied to a C18 reverse phase HPLC chromatography column running a linear 0-80% acetonitrile in H20 gradient for 45 minutes. The fraction containing obtustatin elutes at approximately 21 minutes. After elution the obtustatin containing fraction was lyophylized. Protein elution was followed using A206nm

FIGURE 2 is the elution profile for Viper lebetina obtusa venom proteins present in the lyophylized, obtustatin containing fraction collected as described in FIGURE 1. The lyophylized proteins were resuspended in trifluoroacetic acid and applied to a C18 reverse phase HPLC chromatography column running a linear 20- 70% acetonitrile in H20 gradient for 70 minutes. Obtustatin elutes at approximately 23 minutes. Protein elution was followed using A2o6"m.

FIGURE 3 is a comparison of the obtustatin amino acid sequence with the short disintegrins echistatin and eristostatin. Conserved cysteine residues are boxed.

FIGURE 4 shows the effect of obtustatin, EP-obtustatin (an ethylpyridylated derivative of obtustatin) and fragments of obtustatin on the al integrin-mediated adhesion of K562 cells to collagen IV. Filled circles represent native obtustatin; peptide sequence: CTTGPCCRQCKLKPAGTTCWKTSLTSHYCTGKSCDCPLYPG (SEQ ID NO : 1). Open circles represent EP-obtustatin. Filled triangles represent peptide sequence CWKTSLTSHYC (SEQ ID NO : 2). Open triangles represent peptide sequence TSLTS (SEQ ID NO : 3). Filled squares represent peptide sequence CKLKPAGTTC (SEQ ID NO : 4).

FIGURE 5 shows the effect of obtustatin on adhesion of MV3 cells to immobilized collagen IV. Filled circles show the effect of obtustatin on alpl and cc2ol integrin-mediated binding to collagen IV ; no blocking antibodies are present.

Open circles show the effects of obtustatin on a2pl binding when the α1ß1 integrin is blocked by the al integrin subunit specific AJH10 antibody. Filled triangles show the effect of obtustatin binding on alpl Integrm binding when the a2 (31 integrin is blocked by the a2 integrin subunit specific P1E6 antibody.

FIGURE 6 shows the effect of obtustatin and eristostatin on angiogenesis in a Japanese quail chorioallantoic membrane (CAM) assay. Dr represents fractal dimension as measure of space-filling branching pattern for skeletonized images. Bar A: PBS (control); bar B: eristostatin (20 p. g) ; bar C: eristostatin (50 pg) ; bar D: obtustatin (20 µg) ; bar E: obtustatin (50 pLg).

FIGURE 7 shows the effect of single amino acid changes within the peptide sequence CWKTSLTSHYC (SEQ ID NO : 2) on adhesion of alK562 cells to collagen IV.

Detailed Description of the Invention The present invention provides compounds which inhibit the activity of al (31 integrin. The al ß 1 integrin-inhibiting compounds of the invention are based on the novel short disintegrin obtustatin. Obtustatin is a peptide of 41 amino acid residues, which contains eight conserved cysteine residues and disrupts the biological activity of a1 (31 integrin, for example by specifically inhibiting cellular adhesion to collagen IV. Obtustatin does not contain an RGD sequence. The active sequence of obtustatin appears to be contained within the amino acid sequence KTSLT, and may be SL, SLT or KTS. The primary amino acid sequence of obtustatin as determined by automated Edman degradation is: CTTGPCCRQCKLKPAGTTCWKTSLTSHYCTGKSCDCPLYPG (SEQ ID NO : 1).

Obtustatin was substantially purified from the venom of the viper Vipera lebetina obtusa, and has an apparent molecular weight of 4395 daltons as determined by mass spectroscopy. Obtustatin elutes at approximately 21 minutes from a Cis HPLC reverse phase chromatography column running a linear 0-80% acetonitrile in H20 gradient for 45 minutes (Figure 1) and at approximately 23 minutes from the same column running a linear 20-70% acetonitrile in H20 gradient for 70 minutes (Figure 2). The carboxy-terminal portion of obtustatin is unconserved relative to eristostatin and echistatin (Figure 3), in which the carboxy-terminal amino acid sequences appear to be involved in the selective inhibition of integrins.

According to a preferred embodiment of the invention, the biological activity exhibited by the compounds of the invention is the inhibition of alpl integrin- mediated cellular adhesion to collagen IV. This activity may be measured by cell

adhesion based assays or epitope exposure assays capable of detecting integrin binding activity, which are well known in the art. See, for example, Marcinkiewicz et al., Biochem. J. 317: 817-825 (1996), the entire disclosure of which is herein incorporated by reference, and the cell adhesion assays presented in Example 3. The activity of the compounds of the invention may be expressed as an ICso value, that is, the concentration of an a 1 (31 integrin-inhibiting compound which inhibits 50% of the activity level of an al (31 integrin as measured in the absence of the cclpl integrin- inhibiting compound. It is further preferred that, in cell adhesion assays, the compounds of the invention have an ICso value between 1 pM and 1 M, more preferably between 0.01 nM and 10 nM, and most preferably between'1 nM and 3 nM. The in vitro ICso of native obtustatin is 2 nM (Figure 4 and Table 5). EP- obtustatin a reduced form of full-length obtustatin in which the cysteine thiol groups have been alkylated with 4-vinylpyridine, has an ICso value of 30 uM.

The al (31 integrin-inhibiting compounds of the present invention include, in one embodiment, substantially purified obtustatin and biologically active fragments thereof which are capable of inhibiting al (31 integrin binding to its adhesive ligand.

Preferably, the fragments of the invention are 5-30 amino acids although smaller fragments comprising from 5-25 or 5-20 amino acids, for example, are also contemplated.

In one embodiment, the present invention provides biologically active peptides comprising the sequence of Xl-SL-X2, wherein Xi is the segment CTTGPCCRQCKLKPAGTTCWKT (SEQ ID NO : 5) or an N-terminal truncation fragment thereof containing at least one amino acid, and X2 is the segment TSHYCTGKSCDCPLYPG (SEQ ID NO : 6), or a carboxy-terminal truncation fragment thereof containing at least one amino acid.

The N-terminal truncation fragments comprising Xi may be formed by the sequential removal of amino acids from the N-terminus of SEQ ID NO : 5 ; these truncation fragments are given in Table 2 below.

Table 1-N-Terminal Truncation Fragments Of SEQ ID NO : 5 Fragment Sequence SEQ ID NO. A CTTGPCCRQCKLKPAGTTCWKT 7 B TTGPCCRQCKLKPAGTTCWKT 8 C TGPCCRQCKLKPAGTTCWKT 9 D GPCCRQCKLKPAGTTCWKT 10 E PCCRQCKLKPAGTTCWKT 11 F CCRQCKLKPAGTTCWKT 12 G CRQCKLKPAGTTCWKT 13 H RQCKLKPAGTTCWKT 14 I QCKLKPAGTTCWKT 15 J CKLKPAGTTCWKT 16 K KLKPAGTTCWKT 17 L LKPAGTTCWKT 18 M KPAGTTCWKT 19 N PAGTTCWKT 20 O AGTTCWKT 21 P GTTCWKT 22 Q TTCWKT 23 R TCWKT 24 S CWKT 25 T WKT 26 U KT 27 V T 28

The C-terminal truncation fragments comprising X2 may be formed by the sequential removal of amino acids from the C-terminus of SEQ ID NO : 6 ; these truncation fragments are given in Table 3 below.

Table 3-C-Terminal Truncation Fragments of SEQ ID NO : 6 Fragment Sequence SEQ ID NO. 1 TSHYCTGKSCDCPLYP 29 2 TSHYCTGKSCDCPLY 30 3 TSHYCTGKSCDCPL 31 4 TSHYCTGKSCDCP 32 5 TSHYCTGKSCDC 33 6 TSHYCTGKSCD 34 7 TSHYCTGKSC 35 8 TSHYCTGKS 36 9 TSHYCTGK 37 10 TSHYCTG 38 11 TSHYCT 39 12 TSHYC 40 13 TSHY 41 14 TSH 42 15 TS 43 16 T 44

Biologically active fragments of the formula Xl-SL-X2 are made by combining the fragments of SEQ ID NO : 5 from Table 2 (representing Xl) and the fragments of SEQ ID NO : 6 from Table 3 (representing X2) with the core sequence SL. It is understood that all possible combinations of N-and C-terminal truncation fragments with the core sequence are contemplated as being part of the present invention.

These combinations are shown in Table 4 as follows: The N-terminal truncation fragments from Table 2, represented by letters A through V, are listed in the leftmost column of Table 4. The C-terminal truncation fragments from Table 3, represented by the numerals 1 through 16, are listed in the topmost row of Table 4.

Biologically active fragments are identified by creating a matrix of N-terminal fragments and C-terminal fragments as seen in Table 4. It is understood that the N- terminal fragments are attached to the S residue of the core sequence SL, and the C- terminal fragments are attached to the L residue of the core sequence SL. For ease of illustration, the core sequence SL is not represented in the Table 4 matrix.

Table 4 - Matrix of Biologically Active Peptides From All Possible Combinations of N- and C-Terminal<BR> Truncation Fragments From Tables 2 and 3 Frag 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 A A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 A16 B B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13 B14 B15 B16 C C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 D D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 D14 D15 D16 E E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 E12 E13 E14 E15 E16 F F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 F13 F14 F15 F16 G G1 G2 G3 G4 G5 G6 G7 G8 G9 G10 G11 G12 G13 G14 G15 G16 H H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 I I1 I2 I3 I4 I5 I6 I7 I8 I9 I10 I11 I12 I13 I14 I15 I16 J J1 J2 J3 J4 J5 J6 J7 J8 J9 J10 J11 J12 J13 J14 J15 J16 K K1 K2 K3 K4 K5 K6 K7 K8 K9 K10 K11 K12 K13 K14 K15 K16 L L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 L11 L12 L13 L14 L15 L16 M M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16 N N1 N2 N3 N4 N5 N6 N7 N8 N9 N10 N11 N12 N13 N14 N15 N16 O O1 O2 O3 O4 O5 O6 O7 O8 O9 O10 O11 O12 O13 O14 O15 O16 P P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 Q Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 R R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 S S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 T T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 T14 T15 T16 U U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 V V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13 V14 V15 V16

Thus, referring to Table 4, biologically active peptide M9 is formed from N- terminal fragment M (KPAGTTCWKT ; SEQ ID NO : 45) and C-terminal fragment 9 (TSHYCTGK; SEQ ID NO : 46) attached to the core sequence SL, to give KPAGTTCWKTSLTSHYCTGK (SEQ ID NO : 47). Likewise, biologically active peptide S13 is formed from N-terminal fragment S (CWKT; SEQ ID NO : 48) and C- terminal fragment 13 (TSHY; SEQ ID NO : 49) attached to the core sequence SL, to give CWKTSLTSHY (SEQ ID NO : 50). It is apparent that all combinations of biologically active peptides may be identified by reference to Tables 2,3 and 4.

A preferred biologically active peptide is U19 (KTSLT; SEQ ID NO : 51).

The peptides and fragments of the present invention may be synthesized de novo using conventional solid phase synthesis methods. In such methods, the peptide chain is prepared by a series of coupling reactions in which the constituent amino acids are added to the growing peptide chain in the desired sequence. The use of various N-protecting groups, e. g., the carbobenzyloxy group or the t- butyloxycarbonyl group; various coupling reagents e. g., dicyclohexylcarbodiimide or carbonyldimidazole ; various active esters, e. g., esters of N-hydroxyphthalimide or N- hydroxy-succinimide ; and the various cleavage reagents, e. g., trifluoroactetic acid (TFA), HC1 in dioxane, boron tris- (trifluoracetate) and cyanogen bromide; and reaction in solution with isolation and purification of intermediates are methods well- known to those of ordinary skill in the art. A preferred peptide synthesis method follows conventional Merrifield solid phase procedures well known to those skilled in the art. Additional information about solid phase synthesis procedures can be had by reference to Steward and Young, Solid Phase Peptide Synthesis, W. H. Freeman & Co., San Francisco, 1969; the review chapter by Merrifield in Advances in Enzymology 32: 221-296, F. F. Nold, Ed., Interscience Publishers, New York, 1969; and Erickson and Merrifield, The Proteins 2: 61-64 (1990), the entire disclosures of which are incorporated herein by reference. Crude peptide preparations resulting from solid phase syntheses may be purified by methods well known in the art, such as preparative HPLC. The amino-terminus may be protected according to the methods

described for example by Yang et al., FEBS Lett. 272: 61-64 (1990), the entire disclosure of which is herein incorporated by reference.

The alpl integrin-inhibiting compounds of the present invention also include derivatives of obtustatin. The techniques for obtaining these derivatives are known to persons having ordinary skill in the art and include, for example, standard recombinant nucleic acid techniques, solid phase peptide synthesis techniques, and chemical synthetic techniques. Linking groups may be used to join or replace portions of obtustatin. Linking groups include, for example, cyclic compounds capable of connecting an amino-terminal portion and a carboxyl terminal portion of obtustatin. Techniques for generating derivatives are also described in U. S. patent 6,030,942 the entire disclosure of which is herein incorporated by reference (derivatives are designated"peptoids"in the 6,030,942 patent). Derivatives may incorporate labels such as radioisotopes into their structure and may be in the form of salts such as pharmaceutically acceptable salts.

Examples of obtustatin derivatives include, for example, synthetic mutants of obtustatin. Derivatives of obtustatin may also include, for example, fusion peptides in which a portion of the fusion peptide has a substantially similar amino acid sequence to obtustatin. Fusion peptides may be generated by any means which permits linking two or more peptide sequences including, for example, standard recombinant nucleic acid techniques, solid phase peptide synthesis techniques, or other techniques which are well known to those skilled in the art.

The alpl integrin-inhibiting compounds of the invention also include homologs of obtustatin. Homologs have substantially similar amino acid sequence to obtustatin and may be identified on this basis. It is particularly preferred that homologs of obtustatin contain the core amino acid sequences KTSLT, SL, SLT or KTS.

The al p1 integrin-inhibiting compounds of the invention also include analogs of obtustatin. The analogs of the invention may, for example, be small organic molecules capable of inhibiting alpl integrin activity. Analogs may incorporate

labels such as radioisotopes into their structure and may be in the form of salts such as pharmaceutically acceptable salts.

Without wishing to be bound by a particular theory, it is believed that obtustatin analogs comprise a structure, called a pharmacophore, that mimics the physico-chemical and spatial characteristics of the obtustatin binding site.

Consequently, pro-analogs can, for example, be designed based on variations in the molecular structure of the a 1 (31 integrin's obtustatin binding sites or portions of obtustatin. The structure of the various portions of obtustatin or a 1 (31 integrin may be determined, for example, using NMR (nuclear magnetic resonance), crystallographic, or computational methods which permit the electron density, electrostatic charges or molecular structure of these peptides to be mapped; these methods are well known to those skilled in the art.

Alternatively, pro-analogs can be designed, for example, by using the retrosynthetic, target oriented, or diversity-oriented synthesis strategies described by Schreiber, Science 287: 1964-1969 (2000) the entire disclosure of which is herein incorporated by reference. Retrosynthetic strategies, for example, require that key structural elements in a molecule, such as obtustatin, which interacts with a target molecule, such as the alpl integrin, be identified and then incorporated into the structure of otherwise distinct pro-analogs generated by organic syntheses. U. S. patent 6,030,942, in particular Example 4 therein, describes retrosynthetic methods for the design and selection of analogs based on key structural elements in an inhibitory peptide and is incorporated herein in its entirety (analogs are designated "peptidomimetics"in the 6,030,942 patent).

The solid-phase synthesis methods described by Schreiber supra can be used to generate a library of distinct pro-analogs generated by organic syntheses. Briefly, a suitable synthesis support, for example a resin, is coupled to a pro-analog precursor and the pro-analog precursor is subsequently modified by organic reactions such as, for example, Diels-Alder cyclization. The immobilized pro-analog may then be released from the solid substrate. Pools and subpools of pro-analogs may be generated by automated synthesis techniques in parallel, such that all synthesis and

resynthesis may be performed in a matter of days; pools and subpools of pro-analogs are said to comprise libraries. Once generated, pro-analog libraries can be screened for analogs; i. e. compounds exhibiting one or more biological activities of obtustatin.

Analogs can be identified by, for example, automated screening assays performed in 96 well plates in which the ability of one or more pro-analogs present in solution to inhibit alpl integrin activity is assayed via a cell adhesion assay of the type described in Example 3.

The compounds of the invention may be natural or synthetic peptides produced by any known means, including synthesis by biological systems and chemical methods. Biological synthesis of peptides is well known in the art, and includes the transcription and translation of a synthetic gene encoding obtustatin or biologically active fragments thereof. Chemical peptide synthesis includes manual and automated techniques well known to those skilled in the art. For example, automated synthesis can be performed with commercially available peptide synthesizers. Biologically active fragments according to the invention may also be obtained by the digestion or fragmentation of larger natural or synthetic peptides.

Techniques to synthesize or otherwise obtain peptides and peptide fragments are well known in the art.

The compounds of the invention may be modified with other substances. The modifying substance may be joined to the alpl integrin-inhibiting compound, for example, by chemical means (e. g., by covalent bond, electrostatic interaction, Van der Waals forces, hydrogen bond, ionic bond, chelation, and the like) or by physical entrapment.

For example, the compounds of the invention may be modified with a label (e. g., substances which are magnetic resonance active; radiodense; fluorescent; radioactive; detectable by ultrasound; detectable by visible, infrared or ultraviolet light). Suitable labels include, for example, fluorescein isothiocyanate, peptide chromophores such as phycoerythrin or phycocyanin and the like; bioluminescent peptides such as the luciferases originating from Photinus pyrali ; or fluorescent proteins originating from Renilla renformi. The compounds of the invention may

also be modified with polymeric and macromolecular structures (e. g., liposomes, zeolites, dendrimers, magnetic particles, and metallic beads) or targeting groups (e. g., signal peptide sequences, ligands, lectins, or antibodies). Peptides or peptide fragments may further be modified with end protecting groups at the carboxyl or amino-terminal ends, amino-acid side chain modifying groups, and the like.

Modification of the alpl integrin-inhibiting compounds may alter their activity, for example by altering characteristics such as in vivo tissue partitioning, peptide degradation rate, integrin binding or integrin specificity. The modifications may also confer additional characteristics to the compound, such as the ability to be detected, manipulated or targeted.

Methods of modifying the alßl iregrin-inhibiting compounds with other substances, in particular labels, are well known to those skilled in the art. For example, methods of conjugating fluorescent compounds such as fluorescein isothiocyanate to the short disintegrin eristostatin are described in Danen et al., Exp.

Cell Res., 238: 188-86 (1998), the entire disclosure of which is incorporated herein by reference. Methods of radiolabeling peptides with 125I are disclosed by Sambrook et al. in Molecular Cloning : A Laboratory Manual, Cold Spring Harbor Laboratories, Second Ed., (1989), the disclosure of which is incorporated herein by reference.

The invention also provides a method of inhibiting alpl integrin from binding to their adhesive ligands. The alpl integrin may be free or bound to a cell membrane. Generally, inhibition of alpl integrin binding where alpl integrin is free will be performed in vitro. The inhibition of alpl integrin binding where alpl integrin is bound to a cell membrane may be performed in vitro or in vivo. The method comprises contacting a sample with an effective amount of one or more alpl integrin-inhibiting compounds, so that the binding of alpl integrin to its adhesive ligand is inhibited.

An effective amount of alpl integrin-inhibiting compound sufficient to prevent binding of the integrin to a ligand may be determined by cell adhesion assays or epitope exposure assays, as discussed above.

The compositions according to the invention may be administered in vivo to a subject suffering from a disease or biological condition associated with the binding of alpl integrin to its adhesive ligand. The subject may be any animal, preferably a mammal, and most preferably a human being. Diseases associated with the binding of alpl integrin to its adhesive ligand include, but are not limited to, insulin dependent diabetes mellitus, multiple sclerosis, rheumatoid arthritis, ulcerative colitis, arteriosclerosis, and cancer. Biological conditions associated with the binding of alpl integrin to its adhesive ligand include, but are not limited to, asthma, allergy, organ rejection, and restenosis of arteries after surgery or angioplasty, and angiogenesis.

For example, the al (3l integrin-inhibiting compounds of present invention are useful in the treatment of cancers which express alpl integrins on the surface of the cancer or tumor cells. Such cancers may include, for example, leukemias, melanomas, lymphomas, and sarcomas.

Moreover, the al (31 integrin is known to be involved in angiogenesis and neovascularization, two biological processes which occur during tumorigenesis.

Inhibition of al (31 containing integrins at tumor sites with the compounds of the invention may therefore limit tumor growth, metastasis, and vascularization regardless of whether the tumor cells express alpl integrin.

The al (31 integrin-inhibiting compounds of the present invention may also be useful in the treatment of angiogenesis-mediated disorders other than cancer. Thus, a method for treating, inhibiting or delaying the onset of an angiogenesis-mediated disorder in a subject is provided comprising administering to a subject in need of such treatment an effective amount of an al (3l integrin-inhibiting compound. Such disorders include, for example, metastasis, corneal graft rejection, ocular neovascularization, retinal neovascularization, diabetic retinopathy, retrolental fibroplasia, neovascular glaucoma, gastric ulcer, infantile hemangiomas, angiofibroma of the nasopharynx, avascular necrosis of bone, and endometriosis.

The al (31 integrin is also involved in wound healing and blood clot formation.

Inhibition of al (31 integrin binding may therefore be used to control thrombic occlusion formation, blood clot formation and wound healing.

The invention thus provides a method of treating diseases or biological conditions associated with the binding of alpl integrin to its adhesive ligand, comprising administering to a subject an amount of al (31 integrin-inhibiting compound sufficient to inhibit the binding of a 10 1 integrin with its adhesive ligand.

In preferred embodiments, the alpl integrin-inhibiting compounds are administered as pharmaceutically acceptable salts or a pharmaceutical composition.

Generally, the amount of peptide administered in vivo depends upon the degree of integrin inhibition that is desired. Those skilled in the art may derive appropriate dosages and schedules of administration to suit the specific circumstances and needs of the patient. For example, suitable doses of alpl integrin-inhibiting compound to be administered may be estimated from the cell adhesion assays or epitope exposure assays discussed above. Typically, dosages are between about 0.001 mg/kg and about 100 mg/kg body weight. In some embodiments dosages are between about 0.01 mg/kg and about 10 mg/kg body weight. In other embodiments dosages are between about 0.05 mg/kg and about 5 mg/kg body weight.

For in vivo applications, the compounds of the present invention may comprise a pharmaceutically acceptable salt. Suitable acids which are capable of forming such salts with the compounds of the present invention include inorganic acids such as hydrochloric acid, hydrobromic acid, perchloric acid, nitric acid, thiocyanic acid, sulfuric acid, phosphoric acid and the like ; and organic acids such as formic acid, acetic acid, propionic acid, glycolic acid, lactic acid, anthranilic acid, cinnamic acid, naphthalene sulfonic acid, sulfanilic acid and the like.

When used in vivo, the compounds of the invention are preferably administered as a pharmaceutical composition. The invention thus provides pharmaceutical compositions comprising an alpl integrin-inhibiting compound, or a pharmaceutically acceptable salt thereof, and a pharmaceutically acceptable carrier.

Pharmaceutically acceptable carriers include physiologically tolerable or acceptable diluents, excipients, solvents or adjuvants. The compositions are preferably sterile and nonpyrogenic. Examples of suitable carriers include, but are not limited to, water, normal saline, dextrose, mannitol, lactose or other sugars, lecithin, albumin, sodium glutamate cysteine hydrochloride, ethanol, polyols (propylene glycol, polyethylene glycol, glycerol, and the like), vegetable oils (such as olive oil), injectable organic esters such as ethyl oleate, ethoxylated isosteraryl alcohols, polyoxyethylene sorbitol and sorbitan esters, microcrystalline cellulose, aluminum methahydroxide, bentonite, kaolin, agar-agar and tragacanth, or mixtures of these substances, and the like.

The pharmaceutical compositions may also contain minor amounts of nontoxic auxiliary substances or excipients such as wetting agents, emulsifying agents, pH buffering agents, antibacterial and antifungal agents (such as parabens, chlorobutanol, phenol, sorbic acid, and the like). If desired, absorption enhancing or delaying agents (such as liposomes, aluminum monostearate, or gelatin) may be used.

The compositions can be prepared in conventional forms, either as liquid solutions or suspensions, solid forms suitable for solution or suspension in liquid prior to injection, or as emulsions. Pharmaceutical compositions according to the present invention can be prepared in a manner fully within the skill of the art.

The alpl integrin-inhibiting compounds of the invention, pharmaceutically acceptable salts thereof, or pharmaceutical compositions comprising these compounds, may be administered by any method designed to expose al, Bl integrin expressing cells of a subject to the compounds, so that the compounds may have a physiological effect. Administration may occur enterally or parenterally; for example orally, rectally, intracisternally, intravaginally, intraperitoneally, locally (e. g., with powders, ointments or drops), or as a buccal or nasal spray or aerosol. Parenteral administration is preferred. Particularly preferred parenteral administration methods include intravascular administration (e. g. intravenous bolus injection, intravenous infusion, intra-arterial bolus injection, intra-arterial infusion and catheter instillation into the vasculature), peri-and intra-target tissue injection (e. g. peri-tumoral and intra-

tumoral injection), subcutaneous injection or deposition including subcutaneous infusion (such as by osmotic pumps) and direct application to the target area, for example by a catheter or other placement device. For example, if a subject is being treated for restenosis of coronary arteries after balloon angioplasty, the alp 1 integrin- inhibiting compounds may be administered by intra-arterial infusion at the site of catheterization.

Where the administration of the alpl integrin-inhibiting compound is by injection or direct application, the injection may be in a single dose or in multiple doses. Where the administration of the compound is by infusion, the infusion may be a single sustained dose over a prolonged period of time or multiple infusions.

The alpl integrin-inhibiting compounds of the invention may be used to detect alpl integrin in a variety of samples. Such samples include substances, matrixes, solutions, tissues, cells, organisms, and anything else which may contain, express or be associated with alpl integrin. According to one embodiment, samples which carry alpl integrin may comprise a plurality of immobilized peptides or cells.

Methods for immobilizing peptides are well known to those skilled in the art and include, for example, immobilization of peptides on nitrocellulose. Similarly, methods for immobilizing cells are well known to those skilled in the art and include, for example, immobilization of cells by cross-linking to a solid substrate. A sample may comprise a population of cells or a tissue expressing alpl integrin. Detection of alpl integrin may occur in vitro or in vivo.

The invention thus provides methods of detecting alpl integrin in a sample.

In one embodiment, a sample is immobilized on a solid support, such as a polyacrylamide gel or nitrocellulose filter. A labeled alpl integrin-inhibiting compound is then contacted with the immobilized sample for a time sufficient to allow binding of the labeled compound to any alpl integrin that may be present in the sample. The label may be a substance which is radioactive; emits fluorescent light ; or is detectable, for example by visible, infrared or UV light; by exposure of photographic or X-ray film; by gamma camera, scintillation counter, or other device

capable of detecting radioactive decay. Fluorescent (e. g., fluorescein isothiocyanate) or radioactive (e. g., 35S or l25I) labels are preferred.

Labeled algol integrin-inhibiting compounds bound to al (31 integrin may be detected by any appropriate means, including, for example, fluorescence microscopy, light microscopy, confocal microscopy, electron microscopy, phosphorimaging, autoradiography, scintillation counting, and nuclear magnetic resonance. Such detection techniques are well known to those skilled in the art. Various techniques for detecting the labeled compounds of the invention may also be used in conjunction with other approaches such as fluorescence activated cell sorting (FACS), flow cytometry, or endoscopic techniques.

Labeled compounds of the invention may also be used to map the interactions of alpl integrin with extracellular matrix proteins through techniques such as fluorescence resonance energy transfer (for example as described in Golbik et al., J.

Mol. Biol. 2000,297: 501-509, the entire disclosure of which is incorporated herein by reference).

The invention also provides a method of isolating alpl integrin from a sample. The al (31 integrin may be free or bound to a cell membrane. Samples containing, or suspected of containing, alpl integrin may be contacted with an al (31 integrin-inhibiting compound that has been modified with a selectable label which allows the compound (and any bound alpl integrin) to be separated from the sample.

The alpl integrin-inhibiting compound modified with a selectable label is then contacted with the sample for a time sufficient to allow it to bind to any alpl integrin present in the sample. Examples of suitable selectable labels include, for example, fluorescent labels (e. g., fluorescein isothiocyanate), magnetic particles or beads, ligands, antibodies, and polymeric or macromolecular structures (including solid supports). In preferred embodiments, the conjugated alpl integrin-inhibiting compound is immobilized on a solid support such as nitrocellulose, a polyacrylamide gel or a chromatography column. Once bound to the alpl integrin, the modified alpl integrin-inhibiting compound may be removed from the sample by the

appropriate means. For example, al (31 integrin-inhibiting compounds conjugated to magnetic beads, and having bound a 1 (31 integrin, may be removed from a sample by application of a magnetic field. Alternatively, (xlpl integrin-inhibiting compounds conjugated to insoluble substances, and having bound al (31 integrin, may be removed from a sample by column chromatography, filtration, or by centrifugation. Any al (31 integrin bound to the immobilized al (31 integrin-inhibiting compound may be isolated by recovering and analyzing the complex, or by dislodging the al (31 integrin from the immobilized al (31 integrin-inhibiting compound after unbound material has been removed. Additionally, cells bound to al (31 integrin-inhibiting compounds modified with a fluorescent label such as fluorescein isothiocyanate, may be isolated with flow cytometry techniques. A preferred flow cytometry technique is FACS.

Unknown peptides that bind to the cclol integrin-inhibiting compound may be isolated and identified by techniques well known to those skilled in the art.

Techniques for protein identification include, for example, SDS-PAGE separation of peptides in conjunction with silver staining or other means for detecting proteins in a gel. Techniques for detecting proteins in a gel are disclosed by Sambrook et al., Molecular Cloning : A Laboratory Manual, Cold Spring Harbor Laboratories, Second Ed., (1989), the disclosure of which is herein incorporated by reference. Unknown proteins may also be identified by sequencing, for example with automated sequencing techniques well known to those of ordinary skill in the art.

The present invention also provides antibodies against the compounds of the invention. The antibody of the invention may, for example, specifically bind an epitope of obtustatin. The antibody may be a monoclonal antibody or a polyclonal antibody or an antibody fragment that is capable of binding antigen. One aspect of the invention is a hybridoma that produces a monoclonal antibody which specifically binds to a peptide or peptide fragment according to the invention. The antibodies of the invention may, for example, comprise antibodies, and preparations thereof, produced by immunizing an animal with substantially pure obtustatin or an immunogenic fragment thereof. The present invention includes chimeric, single

chain, and humanized antibodies, as well as Fab fragments and the products of a Fab expression library. Antibody fragments, such as Fab antibody fragments, which retain some ability to selectively bind to the antigen of the antibody from which they are derived, can be made using well known methods in the art. Such methods are generally described in U. S. patent 5,876,997 the entire disclosure of which is incorporated herein by reference.

Polyclonal antibodies may be generated against the compounds of the invention. Antibodies may be obtained following the administration of one or more peptides, fragments, derivatives, or homologs to an animal, using the techniques and procedures known in the art.

Monoclonal antibodies may be prepared using the method of Mishell, B. B., et al., Selected Methods In Cellular Immunology, (W. H. Freeman, ed.) San Francisco (1980), the disclosure of which is herein incorporated by reference. Briefly, a peptide of the present invention is used to immunize spleen cells of Balb/C mice. The immunized spleen cells are fused with myeloma cells. Fused cells containing spleen and myeloma cell characteristics are isolated by growth in HAT medium, a medium which kills both parental cells, but allows the fused products to survive and grow.

Antibodies may be used to purify the compounds of the invention, using immunoaffinity techniques which are well known by those of skill in the art.

The invention will now be illustrated with the following non-limiting examples.

Example 1-Primary purification of obtustatin Lyophilized Vipera lebetina obtusa venom was purchased from Latoxen (Valance, France). Lyophilized venom was dissolved in 0.1% trifluoroacetic acid to a final concentration of 30 mg/ml. The solution was then centrifuged for 5 minutes at 5000 rpm to remove insoluble matter. The supernatant was next applied to a C18 HPLC column and the pellet was discarded. Peptides were eluted from the column with a linear 0-80% acetonitrile in H20 gradient run which ran for 45 minutes.

Peptide elution was followed by monitoring A206nm and 19 fractions were collected.

Each fraction was lyophylized and resuspended in H20. Protein concentration in each resuspended fraction was measured using the bicinchoninic acid (BCA) assay. Five zug of protein from each fraction was then assayed for the ability to disrupt alpl integrin-mediated adhesion of alK562 cells to immobilized collagen IV by the method described in Example 3. The fifth fraction contained a disintegrin activity which inhibited al J31 integrin-mediated adhesion of alK562 cells to immobilized collagen IV. The disintegrin activity in this fraction elutes at approximately 21 minutes and 37% acetonitrile as shown in Figure 1. This disintegrin containing fraction was collected and lyophilized.

Example 2-Secondary purification of obtustatin One mg of the lyophilized, disintegrin containing fifth fraction collected as described in Example 1 was dissolved in 500 uL of 0.1% trifluoroacetic acid solution and applied to a Cl8 HPLC column. Peptides were then eluted from the column with a linear 20-70% acetonitrile in H20 gradient run which ran for 70 minutes as shown in Figure 2. Peptide elution was followed by monitoring A206nm. The first major peak eluted at approximately 23 minutes and contained a disintegrin activity which inhibited al (31 integrin-mediated adhesion of alK562 cells to immobilized collagen IV, as described in Example 3. SDS-PAGE and mass spectrometry confirmed that only one major peptide species was present in this fraction. Mass spectrometry revealed the eluted obtustatin peptide had a molecular mass of 4395 Da. Automated

Edman degradation revealed that the obtustatin peptide has a primary amino acid sequenceof : CTTGPCCRQCKLPAGTTCWKTSLTSHYCTGKSCDCPLYPG (SEQ ID NO : 1).

The yield after the primary and secondary purifications was approximately 12 mg substantially purified obtustatin per 1 g crude Vipera lebetina obtusa venom.

Example 3-The effect of obtustatin, EP-obtustatin and fragments of obtustatin on the adhesion of alK562 cells to collagen IV alK562 cells express the alpl integrin which binds collagen IV.

Ethylpyridylated obtustatin (EP-obtustatin) is a form of full-length obtustatin in which the S-S bonds between cysteine residues have been reduced and exposed thiol groups have been alkylated by reaction with 4-vinylpyridine.

EP-obtustatin was generated by incubating an EP-obtustatin reaction cocktail for 12 hours in the dark at room temperature. The EP-obtustatin reaction cocktail consisted of 100 jig of obtustatin in a solution of 6 M guanidine HCl, 4 mM EDTA, 0.1 M Tris-HCl (pH 8.5), 3.2 mM dithiothreitol (DTT), and 1, uL of a 95% pure 4- vinylpyridine solution added per 100 uL of each reaction cocktail to start the reaction.

Purified EP-obtustatin was obtained post-reaction by applying the reaction cocktail to a C18 HPLC column followed by elution from the column with a linear 0-80% acetonitrile in 0.1% (v/v) trifluoroacetic acid/H20 gradient which ran for 45 minutes.

Peptide elution was followed by monitoring A206nm and the EP-obtustatin containing fraction was collected, lyophilized and resuspended in H20.

Peptides with the primary amino acid sequences of CWKTSLTSHYC (SEQ ID NO : 2), TSLTS (SEQ ID NO : 3), and CKLKPAGTTC (SEQ ID NO : 4) were generated with standard synthetic techniques. These peptides were based on the primary sequence of full-length obtustatin.

The ability of EP-obtustatin and the synthetic peptides described above to inhibit activity of the al (31 integrin was then assayed. Collagen IV (0.2 Rg/well) in 0.02 M acetic acid was immobilized by incubation overnight at 4°C on a 96-well plate. Plates were then blocked with 1% (w/v) BSA (bovine serum albumin) in Hank's Balanced Salt Solution (HBSS) containing 3 mM Mg2+ at room temperature for 1-2 hours to prevent cells from non-specifically binding to the plates. alK562 cells were labeled by incubation with 12.5 uM 5-chloromethylfluorescein diacetete (CMFDA) in HBSS for 15 minutes at 37°C. CMFDA labeled alK562 cells were then pelleted, washed and resuspended in HBSS buffer containing 3mM Mg2+ and 1% BSA. 1 x 105 CMFDA labeled cells were then added to each well in the presence or absence of full-length obtustatin, EP-obtustatin, or the synthetic peptides and incubated at 37°C for 30 minutes. The carrier for full-length obtustatin, EP- obtustatin, and the synthetic peptides was H20 containing 1% (w/v) BSA. The plates were washed three times with HBSS containing 3 mM Mg2+ and 1% (w/v) BSA to remove unbound cells. Bound cells were lysed using 0.5% (v/v) Triton X-100 in H20. Fluorescent CMFDA released by lysis of adherent cells in a given well was read using a Cytofluor 2350 fluorescence plate reader.

Under these assay conditions full-length, native obtustatin had an ICso value of 2 nM (Figure 4) and full-length, reduced EP-obtustatin had an ICso value of 30 uM.

The synthetic peptides limited oclßl integrin-mediated adhesion of aIK562 cells to collagen IV to lesser extent (see Figure 4): The peptide of sequence CWKTSLTSHYC (SEQ ID NO : 2) had an ICso value of 600, uM ; the peptide of sequence TSLTS (SEQ ID NO : 3) had an ICso value of 3.5 mM ; and the peptide of sequence CKLKPAGTTC (SEQ ID NO : 4) did not appreciably inhibit alpl integrin- mediated adhesion of alK562 cells to collagen IV.

Example 4-The effect of obtustatin on adhesion of MV3 cells to immobilized collagen IV The alpl integrin specific inhibitory effect of obtustatin was confirmed using human melanoma MV3 cells. MV3 cells express both the al (31 and a2ol integrins.

MV3 cells adhere strongly to collagen IV through al (31 and a2 (31 integrin-mediated interactions.

Collagen IV (0.2 ug/well) in 0.02 M acetic acid was immobilized overnight at 4°C on a 96-well plate. Plates were then blocked with 1% (w/v) BSA (bovine serum albumin) in Hank's Balanced Salt Solution (HBSS) containing 3 mM Mg2+ at room temperature for 1-2 hours to prevent cells from non-specifically binding to the plates. alK562 cells were labeled by incubation with 12.5 uM 5-chloromethylfluorescein diacetete (CMFDA) in HBSS for 15 minutes at 37°C. CMFDA labeled MV3 cells were then pelleted, washed and resuspended in HBSS buffer containing 3mM Mg2+ and 1% (w/v) BSA. 1 x 105 cells were then added to each well in the presence (10 llg/ml) or absence of the al integrin subunit specific AJH10 mAb and a2 integrin subunit specific P1E6 mAb as indicated in Figure 5. Obtustatin was added at the concentrations indicated in Figure 5 simultaneously with the mAbs. The plates were then incubated at 37°C for 30 minutes and washed three times with HBSS containing 3 mM Mg2+ and 1% (w/v) BSA to remove unbound cells. Bound cells were lysed using 0.5% (v/v) Triton X-100 in H20. Fluorescent CMFDA released by lysis of adherent cells in a given well was read using a Cytofluor 2350 fluorescence plate reader.

As indicated in Figure 5, obtustatin specifically inhibited cclpl integrin- mediated adhesion of MV3 cells to immobilized collagen IV. Adhesion of P1E6 blocked MV3 cells, in which only the orwlßl integrin mediates adhesion to collagen IV, was inhibited by significantly lower obtustatin concentrations than AJH10 blocked cells or unblocked cells. In contrast, unblocked MV3 cells and AJH10 blocked cells, which can adhere to collagen IV via the a2 (31 integrin, were

comparatively insensitive to obtustatin. The extreme obtustatin sensitivity of P1E6 blocked MV3 cells, in combination with the relative insensitivity of unblocked and AJH10 blocked MV3 cells to this peptide, reveals that obtustatin is a potent and selective inhibitor of alpl integrin.

Control experiments confirmed that only the alp 1 and a2p I integrins mediate adhesion of MV3 cells to collagen IV. These control experiments revealed that at final concentrations of 200, ug/ml neither the P1E6 antibody or AJH10 antibody alone could inhibit alpl and a2pl integrin-mediated adhesion of MV3 cells to collagen IV.

Yet simultaneous incubation of MV3 cells with P1E6 and AJH10 antibodies at final concentrations of 10 llg/ml each entirely inhibited alpl and a2pl integrin-mediated adhesion of MV3 cells to collagen IV. Additionally, mAb ASC-1 directed against the a3 integrin subunit present in the VLA-3 collagen receptor had no effect, either alone or in combination with the P1E6 and AJH10 mAbs, on the binding of MV3 cells to collagen IV. Together the results of these control experiments indicate that only the al (31 and a2pl integrins mediate adhesion of MV3 cells to collagen IV.

Example 5-Comparison of the effects of obtustatin, echistatin, and eristostatin on the binding of selected integrins Adhesive ligands for the selected integrins consisted of collagen IV (0.2 , ug/well), collagen I, fibronectin, or vascular cell adhesion molecule-1 (VCAM-1).

Adhesive ligands in 0.02 M acetic acid were immobilized overnight at 4°C on a 96- well plate. Plates were then blocked with 1% (w/v) BSA (bovine serum albumin) in Hank's Balanced Salt Solution (HBSS) containing 3 mM Mg2+ at room temperature for 1-2 hours to prevent cells from non-specifically binding to the plates. alK562, a2K562, K562, A5, and Jurkat cells expressing the alpl, a2pl, a5pl, alIbp3, and a4pl integrins respectively were labeled by incubation with 12.5, uM 5- chloromethylfluorescein diacetate (CMFDA) in HBSS for 15 minutes at 37°C.

CMFDA labeled cells were then pelleted, washed and resuspended in HBSS buffer containing 3 mM Mg2+ and 1% (w/v) BSA. 1 x 105 cells were then added to the

various wells and the amount of either obtustatin, echistatin, or eristostatin was gradually increased up to determine the ICso values indicated below in Table 5. Plates were then incubated at 37°C for 30 minutes as indicated in Table 5 and washed three times with HBSS containing 3 mM Mg2+ and 1% (w/v) BSA to remove unbound cells. Bound cells were lysed using 0.5% (v/v) Triton X-100 in H20. Fluorescent CMFDA released by lysis of adherent cells in a given well was read using a Cytofluor 2350 fluorescence plate reader.

The ICso values in Table 5 represent the mean of three independent experiments. The results summarized in Table 5 reveal that obtustatin is a potent and selective inhibitor ofalpl mediated binding to collagen IV.

Table 5-Comparison of the effects of obtustatin, echistatin, and eristostatin on the binding of selected integrins. IC50,nom Cell Suspension Integrin Adhesive Ligand Obtustatin Echistatin Eristostatin (xIK562 alpl Collagen IV 2 >10, 000 >10, 000 a2K562 a2 (31 Collagen I >10, 000 >10,000 >10,000 K562 a5pl Fibronectin>10, 000 50 >10, 000 A5 aIIbß3 Fibrinogen > 10, 000 50 5 Jurkat a4 (3l VCAM-1>10, 000 >10,000 >10, 000 Example 6-Effect of obtustatin and eristostatin on angiogenesis in Japanese quail CAM assay The disintegrins obtustatin and eristostatin were evaluated for their effect on angiogenesis via a Japanese quail chorioallantoic membrane (CAM) assay, as is known in the art (see, for example, Parsons-Wingerter P et al.,"A Novel Assay of Angiogenesis in the Quail Chorioallantoic Membrane: Stimulation by bFGF and Inhibition by Angiostatin According to Fractal Dimension and Grid Intersection." Microvascular Research 55 (3): 201-214,1998, the disclosure of which is herein

incorporated by reference). The extent of angiogenesis was assessed by evaluating the pattern of vessel branching in skeletonized quasi-two-dimensional CAM vasculature, represented as the fractal dimension Dr. Fig. 6 shows the results for phosphate buffered saline (control ; bar A) ; 20 u. g eristostatin (bar B) ; 50 u. g eristostatin (bar C); 20 gag obtustatin (bar D) ; and 50 ug obtustatin (bar E).

These results show that obtustatin at 20 pug and 50 ug exhibits potent inhibitory activity (up to 30% of control) on angiogenesis. The assay was performed without any angiogenesis stimulators, such as growth factors, indicating the potency of obtustatin in inhibiting angiogenesis. Eristostatin at either 20 ig and 50 u. g showed no affect on angiogenesis in vivo.

Example 7-Effect of single mutation within synthetic peptides representing the integrin-binding loop of obtustatin on adhesion of alK562 cells to collagen IV.

As discussed above, the motif in obtustatin analogous the RGD sequence in other disintegrins could be theoretically identified as SL, SLT, TSL or KTS. To identify the amino acids involved in the activity of obtustatin, nine peptides were synthesized commercially (Sigma-Genosis) based on CWKTSLTSHYC (SEQ ID NO : 2), which is the native obtustatin sequence containing the integrin binding loop.

Each synthetic peptide contained a single change of one of the native amino acids to alanine, as follows. The C residues on either end of the native peptide were not converted. Peptide 1 had the W converted to alanine with all other amino acids as in the native sequence; peptide 2 had the K adjacent to the W converted to alanine, with all other residues as in the native sequences, etc. The collagen IV binding activity of each synthetic peptide, at a final assay concentration of 1 mM, was measured as described in Example 4 above. The results for each peptide are shown in Fig. 7 under the appropriate converted amino acid, given as % inhibition of adhesion of alK562 cells to collagen IV.

The results show that the K, T and S adjacent to the W are involved in obtustatin binding to collagen IV, and that the T is the most important of the three

with respect to the binding activity of obtustatin. These results suggest that the KTS sequence may be relevant for binding of obtustatin to al (31 integrin.

All references discussed herein are incorporated by reference. One skilled in the art will readily appreciate that the present invention is well adapted to carry out the objects and obtain the ends and advantages mentioned, as well as those inherent therein. The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof and, accordingly, references should be made to the appended claims, rather than to the foregoing specification, as indicating the scope of the invention.