WO2000059532A1

WO2000059532A1 - The use of domains of type iv collagen t inhibit angiogenesis an tumour growth

Info

Publication number: WO2000059532A1
Application number: PCT/US2000/008678
Authority: WO
Inventors: Peter Brooks; Billy Hudson
Original assignee: Biostratum, Inc.
Priority date: 1999-04-01
Filing date: 2000-03-31
Publication date: 2000-10-12
Also published as: AU4187200A; WO2000059532A9

Abstract

The instant invention provides methods and kits for inhibiting angiogenesis, tumor growth and metastasis, and endothelial cell interactions with the extracellular matrix, involving contacting the tumor, animal tissue, or endothelial cells with an amount effective to inhibit angiogenesis, tumor growth and metastasis, or endothelial cell interactions with the extracellular matrix of an antagonist of specific integrin receptors.

Description

THE USE OF DOMAINS OF TYPE IV COLLAGEN T INHIBIT ANGIOGENESIS AN

TUMOUR GROWTH

5

Cross Reference

This application claims priority to U.S. Provisional Application Serial No.

60/127,391 filed April 1, 1999, and is a continuation in part of U.S. Application Serial

No. 09/277,665, filed March 26, 1999, which is a continuation in part of U.S. 0 Application Serial No. 09/183,548 filed October 30, 1998, which is a continuation of

08/800,965 filed February 18, 1997, now U.S. Patent No. 5,856,184.

Field of the Invention

This invention relates to methods and kits for inhibiting angiogenesis, tumor 5 growth and metastasis, and endothelial cell interactions with the extracellular matrix.

Background of the Invention

Angiogenesis, the process of formation of new blood vessels, plays an important role in physiological processes such as embryonic and postnatal 0 development, as well as in wound repair. Formation of blood vessels can also be induced by pathological processes involving inflammation (e.g., diabetic retinopathy and arthritis) or neoplasia (e.g., cancer) (Folkman, 1985, Perspect, Biol. Med., 29, 10). Neovascularization is regulated by angiogenic growth factors secreted by tumor or normal cells as well as the composition of the extracellular matrix and by the activity of

25 endothelial enzymes (Nicosia and Ottinetti, 1990, Lab. Invest., 63, 115).

During the initial stages of angiogenesis, endothelial cell sprouts appear through gaps in the basement membrane of pre-existing blood vessels (Nicosia and Ottinetti, 1990, supra; Schoefl, 1963, Virehous Arch, Pathol. Anat. 337, 97-141; Ausprunk and Folkman, 1977, Microvasc. Res. 14, 53-65; Paku and Paweletz, 1991, Lab. Invest. 63, 334-346). As new vessels form, their basement membrane undergoes complex structural and compositional changes that are believed to affect the angiogenic response (Nicosia, et. al., 1994, Exp. Biology, 164, 197-206). Early planar culture models have shown that basement membrane molecules modulate the attachment, migration, proliferation, and organizational behavior of endothelial cells (Nicosia, et. al., 1994, supra). More recent studies with three-dimensional aortic culture models that more closely simulate angiogenic conditions during wound healing in vivo suggest that the basement membrane is a dynamic regulator of angiogenesis, and its function varies according to its molecular components (Nicosia, 1994, supra).

A common feature of all solid tumor growth is the requirement for a blood supply. Therefore, numerous laboratories have focused on developing anti-angiogenic compounds based on growth factors and their receptors. While this approach has led to some success, the number of growth factors known to play a role an angiogenesis is large. Therefore, the possibility exists that growth factor antagonists may have only limited use in treating cancer since tumors and associated inflammatory cells likely produce a wide variety of factors that can induce angiogenesis.

In this regard, a strategy that targets a common feature of angiogenesis, such as endothelial cell adhesion to the extracellular matrix (ECM), might be expected to have a profound physiological impact on tumor growth in humans. This notion is supported by the fact that RGD-containing antagonists of the vβ3 integrin ECM cell adhesion

receptor can block angiogenesis. (U.S. Patent No. 5,766,591) Furthermore, the αvβ3

integrin is expressed most prominently on cytokine -activated endothelial and smooth muscle cells and has been shown to be required for angiogenesis. (Varner et al., Cell Adhesion and Communication 3:367-374 (1995); Brooks et al., Science 264:569-571 (1994)). Based on these findings, a potentially powerful new approach to anti- angiogenic therapy might be to specifically target critical regulatory domains within distinct ECM components.

The basement membrane (basal lamina) is a sheet-like extracellular matrix (ECM), which is a basic component of all tissues. The basal lamina provides for the compartmentalization of tissues, and acts as a filter for substances traveling between tissue compartments. Typically the basal lamina is found closely associated with an epithelium or endothelium in all tissues of an animal including blood vessels and capillaries. The basal lamina components are secreted by cells and then self assemble to form an intricate extra-cellular network. The formation of biologically active basal lamina is important to the development and differentiation of the associated cells.

Type IV collagen has been shown to be a major structural component of basement membranes. The protomeric form of type IV collagen is formed as a heterotrimer made up from a number of different subunit chains called αl(IV) through

α6(IV). Up to now, six genetically distinct α-chains belonging to two classes with

extensive homology have been identified, and their relative abundance has been demonstrated to be tissue specific. The type IV collagen heterotrimer is characterized by three distinct structural domains: the non-collagenous (NCI) domain at the carboxyl terminus; the triple helical collagenous domain in the middle region; and the 7S collagenous domain at the amino terminus. (Martin, et. al., 1988, Adv. Protein Chem. 39:1-50; Gunwar, et. al. 1991, J. Biol. Chem. 266:14088-14094). The ability to express recombinant α(IV) NCI domains provides the opportunity to study the effect of specific domains on many biological processes, such as angiogenesis, tumor metastasis, cell binding to basement membranes, and assembly of Type IV collagen molecules.

Summary of the Invention

The instant invention provides methods and kits for inhibiting angiogenesis, tumor growth and metastasis, and endothelial cell interaction with the extracellular matrix, each method comprising contacting the tumor, animal tissue, or endothelial cells with antagonists of specific integrin receptors.

Brief Description of the Drawings

Figure 1 illustrates the effects of NCI (Hexamer) and 7S domains of Type IV collagen at a 50 μg/ml concentration on angiogenesis from mouse thoracic aorta organ cultures. Figure 2 illustrates the effects of 7S domain of Type IV collagen on angiogenesis from mouse thoracic aorta organ cultures. The domain concentrations employed in this experiment were 0 μg/ml (control); 0.5 μg/ml; 5 μg/ml and 50 μg/ml. Figure 3 illustrates the effects of NCI (Hexamer) domain of Type IV collagen on angiogenesis from mouse thoracic aorta organ cultures. The domain concentrations employed in this experiment were 0 μg/ml (control); 5 μg/ml and 5 μg/ml and 50 μg/ml.

Figure 4 are photographs of mouse thoracic aorta segments embedded in Matrigel (EHS basement membrane matrix, Collaborative Biomedical Products, Bedford, MA) at 5 days of culture. Control specimen (0 μg/ml of NCI (Hexamer) and 7S domains) exhibited growth of microvessels from the cultured tissue into the matrix (Figure 4A).

In contrast, angiogenesis was inhibited in specimens cultured with 50 μg/ml of 7S domain (Figure 4B) and NCI (Hexamer) domain (Figure 4C).

Figure 5 is a graphical representation of data demonstrating the in vivo effect of IV

injection of recombinant (αl) type IV collagen monomer on angiogenesis using fibrin implants in rats.

Figure 6 is a graphical representation of data demonstrating that the recombinant (αl)

and (α2) NCI monomers inhibit the bFGF-induced increase in angiogenic index in

vivo. Figure 7 is a graphical representation of demonstrating the dose response effect of

recombinant (α2) NCI monomer on the bFGF-induced increase in total blood vessel

branch points in vivo.

Figure 8 is a graphical representation of data demonstrating the dose response effect of

recombinant (α2) NCI monomer on the bFGF-induced increase in angiogenic index in

vivo.

Figure 9 is a graphical representation of data demonstrating the dose response effect of

vivo.

Figure 10 is a graphical representation of data demonstrating the effect of recombinant

(αl) and (α2) NCI monomers on mean CS-1 melanoma tumor weight in vivo.

Figure 11 is a graphical representation of data demonstrating the dose response effect

of recombinant (α2) NCI monomer on mean CS-1 melanoma tumor weight in vivo.

Figure 12 is a graphical representation of data demonstrating the effect of recombinant

(αl), (α2), and (α4) NCI monomers on mean HT1080 tumor weight in vivo. Figure 13 is a graphical representation of data demonstrating the effect of recombinant

(αl), (α2), (α3) and (α5) NCI monomers on mean HEP-3 tumor weight in vivo.

Figure 14 is a graphical representation of data demonstrating human endothelial cell

adhesion to immobilized NCI α monomers.

Figure 15 is a graphical representation of data demonstrating the effect of soluble αl

and α2 NCI monomers on human endothelial cell adhesion to pepsinized collagen type

IV.

Figure 16 is a graphical representation of data demonstrating the effect of isolated recombinant NCI monomers on human endothelial cell migration in vitro.

Figure 17 A-F provides the sequences of each type IV collagen α chain monomer.

Figure 18 is a graphical representation of data demonstrating the effect of monoclonal antibodies against various integrins on human endothelial cell adhesion to recombinant

the (α2) NCI domain.

Figure 19 is a graphical representation of data demonstrating human endothelial cell

adhesion to the recombinant (αl) NCI domain.

Description of the Preferred Embodiments

Within this application, unless otherwise stated, the techniques utilized may be found in any of several well-known references such as: Molecular Cloning: A Laboratory Manual (Sambrook, et al., 1989, Cold Spring Harbor Laboratory Press), Gene Expression Technology (Methods in Enzymology, Vol. 185, edited by D. Goeddel, 1991. Academic Press, San Diego, CA), "Guide to Protein Purification" in Methods in Enzymology? (M.P. Deutshcer, ed., (1990) Academic Press, Inc.); PCR Protocols: A Guide to Methods and Applications (Innis, et al. 1990. Academic Press, San Diego, CA), Culture of Animal Cells: A Manual of Basic Technique, 2^nd Ed. (R.L Freshney. 1987. Liss, Inc. New York, NY), and Gene Transfer and Expression Protocols, pp. 109-128, ed. EJ. Murray, The Humana Press Inc., Clifton, NJ.).

As used herein, the term Type IV collagen domain encompasses the group of molecules including the non-collagenous NCI domain (Hexamer) and 7S collagenous

domains, as well as NCI α chain monomers.

The invention comprises methods for using Type IV collagen NCI α-monomers

(ie: αl, α2, α3, and α6), which are defined to include such monomers isolated from

any multicellular organism or produced via recombinant protein expression from a gene encoding such a monomer from any multicellular organism, and also to encompass various modifications, additions, and/or deletions to such monomers.

In one aspect, the present invention provides methods and kits for inhibiting angiogenesis in an animal tissue comprising contacting the tumor or animal tissue with an amount effective to inhibit angiogenesis of a polypeptide composition comprising one or more isolated type IV collagen NCI α chain monomers selected from the group

consisting of αl, α2, α3, and α6 NCI chain monomers.

In another aspect, the present invention provides methods and kits for inhibiting tumor growth in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor growth of a polypeptide composition comprising one or more

isolated type IV collagen NCI α chain monomers selected from the group consisting of

αl, α2, α3, and α6 NCI chain monomers.

In another aspect, the present invention provides methods and kits for inhibiting tumor metastasis in tissue comprising contacting the tumor or tissue with an amount effective to inhibit metastasis of a polypeptide composition comprising one or more isolated type IV collagen NCI α chain monomers selected from the group consisting of

αl, α2, α3, and α6 NCI chain monomers.

In a further aspect, the present invention provides methods and kits for inhibiting endothelial cell interactions with the extracellular matrix in tissue comprising contacting the tumor or tissue with an amount effective to inhibit endothelial cell interactions with the extracellular matrix of a polypeptide composition comprising one or

more isolated type IV collagen NCI α chain monomers selected from the group

consisting of αl, α2, α3, and α6 NCI chain monomers.

The NCI -encoding domain of each of the six α chain cDNAs has been cloned into a vector for recombinant protein expression as previously described (Sado et al., Kidney Intl. 53:664-671 (1998), incorporated by reference herein in its entirety). The vectors are used to stably transfect human kidney 293 cells, which produce the recombinant protein. The DNA and deduced amino acid sequences of the recombinant type IV collagen alpha chain monomers produced as described are shown in Figure 17A-F. The first 17 amino acids correspond to a BM40 signal sequence (which is cleaved from the mature protein), to facilitate protein secretion. All the secreted proteins (ie: mature proteins) start with the sequence APLA followed by the affinity tag, DYKDDDDK at the amino terminus. This tag facilitates purification and identification of the material, and does not interfere with biological activity of the

recombinant NCI α chain monomers.

The type IV collagen NCI α chain monomers can be produced by any method

known in the art, including using recombinant DNA technology or biochemical peptide synthesis technology, or by isolating the NCI domains from animal sources, such as from basement membrane sources such as bovine lens capsule and bovine kidney glomeruli. (Peczon et al., Exp. Eye Res. 30:155-165 (1980); Langeveld et al., J. Biol. Chem. 263:10481-10488 (1988); Gunwar et al, J. Biol. Chem. 266:14088-14094 (1991))

In practicing the invention, the amount or dosage range of type IV collagen

NCI α chain monomers employed is one that effectively inhibits angiogenesis, tumor growth, tumor metastasis, and/or endothelial cell-extracellular matrix interactions. An

inhibiting amount of NCI α chain monomers that can be employed ranges generally between about 0.01 μg/kg body weight and about 10 mg/kg body weight, preferably ranging between about 0.05 μg/kg and about 5 mg/kg body weight.

The NCI α chain monomers may be administered by any suitable route,

including orally, parentally, by inhalation spray, rectally, or topically in dosage unit formulations containing conventional pharmaceutically acceptable carriers, adjuvants, and vehicles. The term parenteral as used herein includes, subcutaneous, intravenous, intraarterial, intramuscular, intrasternal, intratendinous, intraspinal, intracranial, intrathoracic, infusion techniques or intraperitoneally. In preferred embodiments, the NCI α chain monomers are administered intravenously or subcutaneously.

The NCI α chain monomers may be made up in a solid form (including

granules, powders or suppositories) or in a liquid form (e.g., solutions, suspensions, or

emulsions). The NCI α chain monomers of the invention may be applied in a variety

of solutions. Suitable solutions for use in accordance with the invention are sterile,

dissolve sufficient amounts of the NCI α chain monomers, and are not harmful for the proposed application.

The NCI α chain monomers may be subjected to conventional pharmaceutical

operations such as sterilization and/or may contain conventional adjuvants, such as preservatives, stabilizers, wetting agents, emulsifiers, buffers etc. For administration, the NCI α chain monomers are ordinarily combined with

one or more adjuvants appropriate for the indicated route of administration. The compounds may be admixed with lactose, sucrose, starch powder, cellulose esters of alkanoic acids, stearic acid, talc, magnesium stearate, magnesium oxide, sodium and calcium salts of phosphoric and sulphuric acids, acacia, gelatin, sodium alginate, polyvinylpyrrolidine, and/or polyvinyl alcohol, and tableted or encapsulated for conventional administration. Alternatively, the compounds of this invention may be dissolved in saline, water, polyethylene glycol, propylene glycol, carboxymethyl cellulose colloidal solutions, ethanol, corn oil, peanut oil, cottonseed oil, sesame oil, tragacanth gum, and/or various buffers. Other adjuvants and modes of administration are well known in the pharmaceutical art. The carrier or diluent may include time delay material, such as glyceryl monostearate or glyceryl distearate alone or with a wax, or other materials well known in the art.

The present invention may be better understood with reference to the accompanying examples that are intended for purposes of illustration only and should not be construed to limit the scope of the invention, as defined by the claims appended hereto.

Example 1 - In Vitro Effect on Angiogenesis With modifications, the procedures of Nicosia and Ottinetti (1990), supra, and

Nicosia, et. al. (1994), supra, were utilized for experiments designed to test the effect of Type IV collagen on angiogenesis under in vitro conditions. The model has been used to study the effects of growth factors and extracellular matrix molecules on the angiogenic response and employs aortic rings cultures in three-dimensional collagen gels under serum-free conditions. These experiments are outlined below. A. Methods

Experiments were performed with 1-3 month old Swiss Webster male mice. Following anesthesia, the thoracic aorta was excised under aseptic conditions and transferred to sterile MCDB 131 sterile growth medium (Clonetics, San Diego, CA) containing antibiotics. Fat was dissected away from the aorta and approximately six to eight 1 mm thoracic segments were obtained from each specimen. Segments were transferred to 48 well tissue culture plates. The wells of these plates were layered with 100 microliters of Matrigel (EHS basement membrane, Collaborative Biomedical Products, Bedford, MA) prior to transfer of the aortic segments. The Matrigel was diluted 1 :1 with MCDB 131 growth medium prior to use. The segments were centered in the wells and an additional 100 microliters of Matrigel was then placed over the specimens. The aortic segments were therefore embedded in the basement membrane matrix. Each well then received 300 microliters of MCDB 131 growth medium. The

plates were placed in an incubator maintained at 37° C with 5% CO₂. Specimens were observed daily over a 7 day period. Newly growing microvessels were counted using an inverted phase microscope at various times during the culture period, but data is expressed at 3 and 5 days of culture. To test for the effect of Type IV collagen on angiogenesis, domains at known concentrations were mixed with the Matrigel and with the MCDB 131 growth medium. Fresh MCDB 131 growth medium (plus and minus collagen domains) was changed every 3 days. B. Results

After establishing the time course of angiogenesis under control conditions (Matrigel plus MCDB 131 growth medium), experiments were performed using various concentrations of Type IV collagen (isolated from bovine lens) NCI (hexamer) and 7S domains. Data represents the analysis of at least 3 specimens per experimental condition. In the first experiment (Figure 1), analysis indicated that at a concentration of 50 μg/ml, NCI domain and 7S domain significantly inhibited angiogenesis as monitored at 3 and 5 days of culture. In the second experiment, various concentrations of these domains were analyzed. As indicated in Figure 3, 7S domain at 50 μg/ml again significantly inhibited angiogenesis at 3 and 5 days. Inhibition was reduced at 5 and 0.5 μg/ml concentrations. As indicated in Figure 2, NCI domain was less effective in blocking angiogenesis as compared to that observed in the first experiment (Figure 1), although it was still effective. In addition, as compared to the 7S domain, there was less of a correlation between concentration and inhibitory action. Figure 4A-C are photographs of mouse thoracic aorta segments embedded in

Matrigel (EHS basement membrane matrix, Collaborative Biomedical Products, Bedford, MA) at 5 days of culture in the presence or absence of 50 μg/ml of Type IV collagen domains. The control specimen (no domains) exhibited growth of microvessels from the cultured tissue into the matrix (Figure 4A). In contrast, angiogenesis inhibition was observed in tissues cultured in the presence of 50 μg/ml of 7S (Figure 4B) and NCI (Hexamer) domain (Figure 4C).

Example 2. Subcutaneous fibrin implant angiogenesis Recombinant human type IV collagen NCI (α3) monomer (Sado et al., Kidney International 53:664-671 (1998)) was injected intravenously in Fisher 344 rats containing fibrin implants surgically placed subcutaneously, a modified version of the method described by Dvorak et al ( Lab. Invest. 57(6):673-686 (1987)). The implants were then removed and directly analyzed using an inverted microscope. The analysis involved counting the number of blood vessels that had grown into the fibrin in the control and experimental group.

Briefly, 4 fibrin implants were surgically implanted subcutaneously into Fisher 344 rats (2 dorsal and 2 ventral sides). The average rat weight was approximately 125 grams.

Three rats (EXP) were given tail vein injections of either control (fibrin alone),

100 μl of 100 μg/ml of 7S domain of type IV collagen (approximately 0.80 mg/kg body weight), 100 μl of 100 μg/ml of type IV collagen hexamer (approximately 0.80 mg/kg

body weight), or recombinant collagen type IV NCI (α3) monomer at a concentration of 1.26 mg/ml in PBS (120 μg protein, or approximately 0.96 mg/kg body weight) and

3 rats (C) were given 100 μl tail vein injections of PBS. Injections of recombinant protein were given every other day for five doses. The injection schedule was as follows:

Day 1 : (implant day) injection and remove blood sample (EXP and C) Day 3 : Injection (EXP and C)

Day 5: Injection and remove blood sample (EXP and C)

Day 7: Injection (EXP and C)

Day 9: Injection and remove blood sample (EXP and C)

Day 11 : Remove and fix implants (save blood sample) (EXP and C)

The results of one experiment were as follows:

2 week in vivo experiment:

Control (fibrin alone) about 66 BV 7S domain of type IV lens collagen (100 μg/ml) None Hexamer of type IV lens collagen (100 μg/ml) None

Monomer (α3) None

The results are shown as the mean number of blood vessels per implant. The results of this study demonstrate that isolated domains of type IV collagen, including the α3 monomer, can significantly inhibit capillary growth in the in vivo fibrin clot

implant model. In subsequent experiments, the inhibitory effect was occasionally seen to attenuate with time, suggesting that higher dosages or more frequent injections might be even more effective.

A similar experiment was conducted using recombinant human type IV collagen

NCI (αl) monomer (100 μl of a 1 μg/μl solution; approximately 0.80 mg/kg body

weight) and comparing the number of blood vessels that had grown into the fibrin at day 11 of treatment relative to the control group. Three rats per group were analyzed with each rat having 4 implants. These experiments demonstrated that administration of the αl monomer significantly inhibited capillary growth in the in vivo fibrin clot

implant model (Figure 5).

Example 3. Recombinant NCI (α2) domain inhibits angiogenesis in vivo We next tested the effects of systemic administration of soluble NCI α-chain

monomers in the chick embryo CAM angiogenesis assay.

Angiogenesis was induced in the CAMs of 10 day old chick embryos with bFGF as described (Brooks et al, Cell 92:391-400 (1998)). Twenty four hours later the

embryos were systemically treated with various concentrations of recombinant NCI α-

chain monomers, in a total volume of 100 μl of sterile phosphate buffered saline (PBS).

Two days later the embryos were sacrificed and the filter discs and CAM tissues removed. Angiogenesis was quantitated by counting the number of angigogenic blood vessel branch points in the confined area of the filter disc. The Angiogenic Index is defined as the number of branch points from experimental treatment minus control treatment.

In initial experiments, recombinant αl or α2 NCI domains were injected at a

concentration of 50 μg per embryo. At this concentration, the NCI domains were shown to be highly toxic as demonstrated by greater than 90% embryo cell death. However, at lower doses they were well tolerated and showed potent anti-angiogenic activity. A total of 6 individual angiogenesis experiments were conducted with the NCI domains. However, in two experiments, the bFGF induction was low, making it

difficult to interpret the results. The NCI α2 domain appeared to be more consistent

and potent than the αl NCI domain at inhibiting angiogenesis. In fact, systemic

administration of 30 μg of NCI α2 consistently inhibited angiogenesis by greater than

90% (Figures 6-9), as measured by inhibition of the bFGF-induced increase in the angiogenic index and the mean number of blood vessel branch points. In contrast, NCI

αl domain showed variable inhibitory activity (0%-50%) throughout the experiments.

Example 4 . Recombinant NCI domain inhibits melanoma tumor growth in vivo:

Since the growth of all solid tumors depends on angiogenesis to provide nutrients for its continued expansion, reagents that have the capacity to inhibit angiogenesis may significantly inhibit tumor growth. Therefore, we tested the effects of recombinant NCI domains of type IV collagen for their effects on tumor growth in vivo. To test the effects of NCI domains on tumor growth in vivo, we utilized the chick embryo tumor growth assay. Briefly, single cell suspensions of 3 distinct tumor types were applied to the CAM of 10 day old chick embryos. The tumors included CS- 1 Melanoma cells (5 x 10⁶), HT1080 human fibrosarcoma cells (4 x 10⁵) and Hep-3 human epidermoid carcinoma cells (2 x 10⁵). The embryos were injected systemically

with varying concentrations of NCI α-chain monomers 24 hours later. The embryos

were next allowed to incubate for a total of 7 days, at which time they were sacrificed. The resulting tumors were resected and wet weights determined. A total of 6 tumor growth assays were conducted with the 3 distinct tumor types. A single injection of 10 μg NCI α2 domain inhibited CSl melanoma tumor growth by approximately 70%

relative to control (Figure 10). In similar experiments, dose response curves were completed with CS-1 tumors. Systemic administration of NCI α2 resulted in a dose-

dependent inhibition of CS-1 melanoma tumor growth in vivo with a maximum inhibition following a single dose at 30 μg (Figure 11). Systemic administration of

NCI αl also inhibited CS-1 tumor growth but it was variable and in some experiments

failed to inhibit tumor growth (See Figure 10). In similar experiments, NCI α2

inhibited HT1080 human fibrosarcoma tumor growth by approximately 50% after a

single systemic injection of 30 μg, while NCI αl and α4 had no effect (Figure 12).

Finally, systemic administration of NCI α2 (30.0 μg) and α3 inhibited Hep-3 human epidermoid carcinoma tumor growth by approximately 40% and 60% respectively, and

αl inhibited Hep-3 tumor growth by approximately 30%, while NCI α5 domain failed

to inhibit tumor growth (Figure 13).

We conclude from these in vivo studies that tumor growth can be inhibited by isolated NCI α-chain monomers. These molecules can thus be used alone, or to complement the use of existing anti-tumor agents, in providing enhanced and more effective anti-tumor therapy.

Example 5. Immobilized NCI domains support human endothelial cell adhesion In order for new blood vessels to form, endothelial cells must have the capacity to adhere and migrate through the ECM. Moreover, this endothelial cell-ECM interaction may facilitate signal transduction events required for new blood vessel formation. Therefore, since type IV-collagen is an ECM protein which is known to support cell adhesion, we tested the ability of the NCI domains to support endothelial cell attachment.

Microtiter plates were coated with 25 μg/ml of purified NCI domains followed by incubation with 1% bovine serum albumin (BSA) to block non-specific interactions. Human endothelial cells (ECV304) were then allowed to attach to the immobilized NCI domains for 1 hour. Non-adherent cells were removed by washing and attached cells were quantified by measuring the optical density (O.D.) of crystal violet eluted from attached cells. Data bars represent the mean +/- standard error of the O.D. from triplicate wells.

Immobilized NCI α2, α3, and α6 domains supported endothelial cell adhesion

while NCI αl, α4, and α5 domains promoted little if any cell adhesion (Figure 14).

Soluble NCI αl (al) and α2 (a2) inhibited endothelial cell adhesion to pepsinized

collagen type IV by approximately 50% (Figure 15).

Taken together, these findings demonstrate that isolated, recombinant NCI

domains from the αl, α2 , α3, and α6 chains of collagen type IV can mediate human endothelial cell adhesion and/or inhibit endothelial cell adhesion to ECM proteins in vitro, and suggest that the potent anti-angiogenic and anti-tumor activity of the isolated NCI domains is due to disruption of endothelial cell interaction with the extracellular matrix that are necessary for angiogenesis.

Example 6. Endothelial Cell Migration

Invasive cellular processes such as angiogenesis and tumor metastasis also require cellular motility. Thus we evaluated the ability of isolated NCI domains to support human endothelial cell migration in vitro. These experiments were conducted essentially according to the methods in Brooks et al., J. Clin. Invest. 99:1390-1398 (1997).

The results of these experiments indicate that NCI α2, α3, and α6 domains can

support human endothelial cell migration in vitro, while αl, α4, and α5 domains

showed little if any capacity to support endothelial cell migration (FIG 16).

Example 7. Efficacy in Lewis lung in vivo tumor

The above studies indicated that specific domains of collagen type IV can promote cell migration in vitro. Thus, we evaluated the ability of NCI domains to support endothelial cell migration in vivo.

The α (IV) NCI domain hexamer, isolated by enzymatic digestion of bovine

lens capsule basement membrane by known protocols (Peczon et al., Exp. Eye Res.

30:155-165 (1980)) was tested in the metastatic Lewis lung mouse tumor model using a standard protocol which is considered to be a good model of both metastasis and angiogenesis of lung tumors. (See for example, Teicher et al., Anticancer Res. 18:2567-2573 (1998); Guibaud et al., Anticancer Drugs 8:276-282 (1997); Anderson et al., Cancer Res. 56:715-718 (1996)).

Each study consisted of an untreated control group and six treatment groups. There were ten animals per treatment group with 40 mice in the control. In each study, all treatment was administered intravenously once every 2 days for 7 doses starting one

day after tumor inoculation. Dosages of α (IV) NCI hexamer were either 100 μg/mouse or 200 μg/mouse. In the Lewis lung study, the tumor cell inoculum was 1 xlO⁶ viable cells. All animals were weighed twice a week throughout the study. Starting one day after the last treatment, 5 mice were periodically sacrificed from each control group to measure pulmonary tumor burden. The experiment was terminated at day 14 when the lungs of the control animals had sufficient tumor mass to provide meaningful evaluation. At that time, the lungs of all remaining animals were excised, weighed, and the number of tumor foci greater than 2 mm in diameter counted. The

resulting data showed that both dosages of α (IV) NCI hexamer significantly reduced the number of visible lung metastases (Mann- Whitney Rank Sum Test, p < 0.05), with 8 visible lung metastases in the control, vs. 5 (100 μg/mouse) and 4 (200 μg/mouse), and the 100 μg/mouse dosage reduced the lung weights from a median of 520 mg in controls to a median of 462 mg in experimental, while the median lung weight of mice treated with 200 μg/mouse was 620 mg. Other in vivo studies demonstrated that tumor cell metastasis to the lung can be reduced by 50% or more using intravenous injections of the Type IV collagen domains in murine B16 melanoma, human A375SM melanoma xenografts. Furthermore, injection of the NCI hexamer also significantly reduced the number of lung tumors in separate Lewis Lung tumor studies. Example 8. Defining the Integrin Receptor Mediating Cellular Adhesion to the NCI domains

To define the integrin receptors that mediated cellular adhesion to the NCI αl

and α2 domains, adhesion assays were performed as described in Example 5 in the

presence or absence of function blocking monoclonal antibodies directed to specific

integrins (Figures 18 (α2); Fig. 19 (αl)). These antibodies were directed against α5B3

integrin (anti-avb3), the α5B5 integrin (anti-avb5), the Bl integrin (bl) (all described in U.S. Patent No. 5,766,591, incorporated by reference herein in its entirety), and

monoclonal antibodies directed against the αl (anti-al), α2 (anti-a2), and α3 (anti-a3)

integrins (purchased from Chemicon, California). These studies indicated that human endothelial cells interact with NCI α2 domain primarily through αvB5 and αvB3 integrins with variable contribution from Bl integrins (Figure 18). In similar experiments, anti-Bl integrin antibodies showed a lesser effect on endothelial cell

adhesion to NCI α2, suggesting a lesser contribution of Bl integrins to this adhesive

activity. In contrast, endothelial cell adhesion promoted by NCI αl domain was

mediated by integrin α3Bl (Figure 19).

Previous studies have demonstrated that RGD-containing antagonists of the

αvβ3 receptor can block angiogenesis (U.S. Patent No. 5,766,591), but the instant invention provides the first demonstration of a non-RGD containing antagonist of the

αvβ3 integrin that can block angiogenesis. The present study also demonstrates that

antagonists of the αvβ5 integrin and the α3βl integrins can block angiogenesis.

Thus, the instant invention also provides methods and kits for inhibiting angiogenesis, tumor growth and metastasis, and endothelial cell interaction with the extracellular matrix, each method comprising contacting the tumor, animal tissue, or endothelial cells with antagonists of specific integrin receptors. Specifically, the methods comprise contacting the tumor, animal tissue, or endothelial cells with one or more of the following polypeptide compositions:

(a) a polypeptide composition comprising one or more non-RGD containing

integrin αvβ3 antagonists; or

(b) a polypeptide composition comprising one or more antagonists of αvβ5

integrin; or

(c) a polypeptide composition comprising one or more antagonists of βl

integrins; or (d) a polypeptide composition comprising one or more antagonists of α3βl

integrins.

We conclude from all of the above studies that angiogenesis, tumor growth and metastasis, and endothelial cell adhesion to the ECM, can be inhibited by isolated, recombinant domains of type IV collagen, or by antagonists of specific integrin receptors. The present invention is thus broadly applicable to a variety of uses which include inhibition of angiogenesis and treatment of diseases and conditions with accompanying undesired angiogenesis, such as solid and blood-borne tumors including but not limited to melanomas, carcinomas, sarcomas, rhabdomyosarcoma, retinoblastoma., Ewing sarcoma, neuroblastoma, osteosarcoma, and leukemia.

The invention is further applicable to treating non-tumorigenic diseases and conditions with accompanying undesired angiogenesis, including but not limited to diabetic retinopathy, rheumatoid arthritis, retinal neovascularization, choroidal neovascularization, macular degeneration., corneal neovascularization, retinopathy of prematurity., corneal graft rejection, neovascular glaucoma., retrolental fibroplasia, epidemic keratoconjunctivitis, Vitamin A deficiency, contact lens overwear, atopic keratitis, superior limbic keratitis, pterygium keratitis sicca, sogrens, acne rosacea, phylectenulosis, syphilis, Mycobacteria infections, lipid degeneration, chemical burns, bacterial ulcers, fungal ulcers, Herpes simplex infections, Herpes zoster infections, protozoan infections, Kaposi's sarcoma, Mooren ulcer, Terrien's marginal degeneration, marginal keratolysis, traum, systemic lupus, polyarteritis, Wegeners sarcoidosis, scleritis, Steven's Johnson disease, radial keratotomy, sickle cell anemia, sarcoid, pseudoxanthoma elasticum, Pagets disease, vein occlusion, artery occulsion, carotid obstructive disease, chronic uveitis, chronic vitritis, Lyme's disease, Eales disease, Bechets disease, myopia, optic pits, Stargarts disease, pars planitis, chronic retinal detachment, hyperviscosity syndromes, toxoplasmosis, post-laser complications, abnormal proliferation of fibrovascular tissue, hemangiomas, Osier- Weber-Rendu, acquired immune deficiency syndrome, ocular neovascular disease, osteoarthritis, chronic inflammation, Crohn's disease, ulceritive colitis, psoriasis., atherosclerosis, and pemphigoid. See U.S. Patent No. 5,712,291)

The invention is also broadly applicable to methods for inhibiting tumor growth and metastasis, reduction of scar tissue formation, reduction of complications due to cell adhesion in organ transplants, and the inhibition of lymphocyte adhesion and mobility.

While the fundamental novel features of the invention have been shown and described, it will be understood that various omissions, substitutions, and changes in the form and details illustrated may be made by those skilled in the art without departing from the spirit of the invention. For example, various modifications, additions, and/or substitutions can be made to the type IV collagen α monomer chains that would be encompassed by the invention.

Claims

We claim

1. A method for inhibiting angiogenesis in an animal tissue comprising contacting the tumor or animal tissue with an amount effective to inhibit angiogenesis of a

polypeptide composition comprising one or more non-RGD containing integrin αvβ3 antagonists.

2. A method for inhibiting angiogenesis in an animal tissue comprising contacting the tumor or animal tissue with an amount effective to inhibit angiogenesis of a

polypeptide composition comprising one or more antagonists of αvβ5 integrin.

3. A method for inhibiting angiogenesis in an animal tissue comprising contacting the tumor or animal tissue with an amount effective to inhibit angiogenesis of a

polypeptide composition comprising one or more antagonists of βl integrins.

4. A method for inhibiting angiogenesis in an animal tissue comprising contacting the tumor or animal tissue with an amount effective to inhibit angiogenesis of a

polypeptide composition comprising one or more antagonists of α3βl integrin.

5. A method for inhibiting endothelial cell adhesion to extracellular matrix, comprising contacting the endothelial cell with an amount effective to inhibit endothelial cell adhesion to extracellular matrix of a polypeptide composition comprising one or

more non-RGD containing integrin αvβ3 antagonists.

6. A method for inhibiting endothelial cell adhesion to extracellular matrix, comprising contacting the endothelial cell with an amount effective to inhibit endothelial cell adhesion to extracellular matrix of a polypeptide composition comprising one or

more antagonists of αvβ5 integrin.

7. A method for inhibiting endothelial cell adhesion to extracellular matrix, comprising contacting the endothelial cell with an amount effective to inhibit endothelial cell adhesion to extracellular matrix of a polypeptide composition comprising one or

more antagonists of α3βl integrin.

8. A method for inhibiting endothelial cell adhesion to extracellular matrix, comprising contacting the endothelial cell with an amount effective to inhibit endothelial cell adhesion to extracellular matrix of a polypeptide composition comprising one or

more antagonists of βl integrins.

9. A method for inhibiting tumor metastasis in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor metastasis of a polypeptide

composition comprising one or more non-RGD containing integrin αvβ3 antagonists.

10. A method for inhibiting tumor metastasis in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor metastasis of a polypeptide

composition comprising one or more antagonists of αvβ5 integrin.

11. A method for inhibiting tumor metastasis in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor metastasis of a polypeptide

composition comprising one or more antagonists of α3βl integrin.

12. A method for inhibiting tumor metastasis in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor metastasis of a polypeptide

composition comprising one or more antagonists of βl integrins.

13. A method for inhibiting tumor growth in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor growth of a polypeptide composition

comprising one or more non-RGD containing integrin αvβ3 antagonists.

14. A method for inhibiting tumor growth in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor growth of a polypeptide composition comprising one or more antagonists of αvβ5 integrin.

15. A method for inhibiting tumor growth in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor growth of a polypeptide composition

comprising one or more antagonists of α3βl integrin.

16. A method for inhibiting tumor growth in tissue comprising contacting the tumor or tissue with an amount effective to inhibit tumor growth of a polypeptide composition

comprising one or more antagonists of β 1 integrins.