TIE-2 ligands, methods of making and uses thereof

INTRODUCTION

[0002] The present invention relates generally to the field of genetic engineering and more particularly to genes for receptor tyrosine kinases and their cognate ligands, their insertion into recombinant DNA vectors, and the production of the encoded proteins in recipient strains of microorganisms and recipient eukaryotic cells. More specifically, the present invention is directed to novel ligands, known as the TIE-2 ligands, that bind the TIE-2 receptor, as well as to methods of making and using the TIE-2 ligands. The invention further provides nucleic acid sequences encoding TIE-2 ligands, and methods for the generation of nucleic acids encoding TIE-2 ligands and their gene products. The TIE-2 ligands, as well as nucleic acids encoding them, may be useful in the diagnosis and treatment of certain diseases involving endothelial cells and associated TIE receptors, such as neoplastic diseases involving tumor angiogenesis, wound healing, thromboembolic diseases, atherosclerosis and inflammatory diseases. In addition, the ligands may be used to promote the proliferation and/or differentiation of hematopoietic stem cells.

[0003] More generally, biologically active TIE-2 ligands may be used to promote the growth, survival, migration, and/or differentiation and/or stabilization or destabilization of cells expressing the TIE-2 receptor. Biologically active TIE-2 ligand may be used for the in vitro maintenance of TIE-2 receptor expressing cells in culture. Cells and tissues expressing TIE-2 receptor include, for example, cardiac and vascular endothelial cells, lens epithelium and heart epicardium and early hematopoietic cells. Alternatively, such ligand may be used to support cells which are engineered to express TIE-2 receptor. Further, TIE-2 ligands and their cognate receptor may be used in assay systems to identify agonists or antagonists of the TIE-2 receptor.

BACKGROUND OF THE INVENTION

[0004] The cellular behavior responsible for the development, maintenance, and repair of differentiated cells and tissues is regulated, in large part, by intercellular signals conveyed via growth factors and similar ligands and their receptors. The receptors are located on the cell surface of responding cells and they bind peptides or polypeptides known as growth factors as well as other hormone-like ligands. The results of this interaction are rapid biochemical changes in the responding cells, as well as a rapid and a long-term readjustment of cellular gene expression. Several receptors associated with various cell surfaces may bind specific growth factors.

[0005] The phosphorylation of tyrosine residues in proteins by tyrosine kinases is one of the key modes by which signals are transduced across the plasma membrane. Several currently known protein tyrosine kinase genes encode transmembrane receptors for polypeptide growth factors and hormones such as epidermal growth factor (EGF), insulin, insulin like growth factor-I (IGF-I), platelet derived growth factors (PDGF-A and -B), and fibroblast growth factors (FGFs). (Heldin et al., Cell Regulation, 1: 555-566 (1990); Ullrich, et al., Cell, 61: 243-54 (1990)). In each instance, these growth factors exert their action by binding to the extracellular portion of their cognate receptors, which leads to activation of the intrinsic tyrosine kinase present on the cytoplasmic portion of the receptor. Growth factor receptors of endothelial cells are of particular interest due to the possible involvement of growth factors in several important physiological and pathological processes, such as vasculogenesis, angiogenesis, atherosclerosis, and inflammatory diseases. (Folkman, et al. Science, 235: 442-447 (1987)). Also, the receptors of several hematopoietic growth factors are tyrosine kinases; these include c-fms, which is the colony stimulating factor 1 receptor, Sherr, et al., Cell, 41: 665-676 (1985), and c-kit, a primitive hematopoietic growth factor receptor reported in Huang, et al., Cell, 63: 225-33 (1990).

[0006] The receptor tyrosine kinases have been divided into evolutionary subfamilies based on the characteristic structure of their ectodomains. (Ullrich, et al. Cell, 61: 243-54 (1990)). Such subfamilies include, EGF receptor-like kinase (subclass I) and insulin receptor-like kinase (subclass II), each of which contains repeated homologous cysteine-rich sequences in their extracellular domains. A single cysteine-rich region is also found in the extracellular domains of the eph-like kinases. Hirai, et al., Science, 238: 1717-1720 (1987); Lindberg, et al. Mol. Cell. Biol., 10: 6316-24 (1990); Lhotak, et al., Mol. Cell. Biol. 11: 2496-2502 (1991). PDGF receptors as well as c-fms and c-kit receptor tyrosine kinases may be grouped into subclass III; while the FGF receptors form subclass IV. Typical for the members of both of these subclasses are extracellular folding units stabilized by intrachain disulfide bonds. These so-called immunoglobulin (Ig)-like folds are found in the proteins of the immunoglobulin superfamily which contains a wide variety of other cell surface receptors having either cell-bound or soluble ligands. Williams, et al., Ann. Rev. Immunol., 6: 381-405 (1988).

[0007] Receptor tyrosine kinases differ in their specificity and affinity. In general, receptor tyrosine kinases are glycoproteins which consist of (1) an extracellular domain capable of binding the specific growth factor(s); (2) a transmembrane domain which usually is an alpha-helical portion of the protein; (3) a juxtamembrane domain where the receptor may be regulated by, e.g., protein phosphorylation; (4) a tyrosine kinase domain which is the enzymatic component of the receptor; and (5) a carboxyterminal tail which in many receptors is involved in recognition and binding of the substrates for the tyrosine kinase.

[0008] Processes such as alternative exon splicing and alternative choice of gene promoter or polyadenylation sites have been reported to be capable of producing several distinct polypeptides from the same gene. These polypeptides may or may not contain the various domains listed above. As a consequence, some extracellular domains may be expressed as separate, secreted proteins and some forms of the receptors may lack the tyrosine kinase domain and contain only the extracellular domain inserted in the plasma membrane via the transmembrane domain plus a short carboxyl terminal tail.

[0009] A gene encoding an endothelial cell transmembrane tyrosine kinase, originally identified by RT-PCR as an unknown tyrosine kinase-homologous cDNA fragment from human leukemia cells, was described by Partanen, et al., Proc. Natl. Acad. Sci. USA, 87: 8913-8917 (1990). This gene and its encoded protein are called “TIE” which is an abbreviation for “tyrosine kinase with Ig and EGF homology domains.” Partanen, et al. Mol. Cell. Biol. 12: 1698-1707 (1992).

[0010] It has been reported that tie mRNA is present in all human fetal and mouse embryonic tissues. Upon inspection, tie message has been localized to the cardiac and vascular endothelial cells. Specifically, tie mRNA has been localized to the endothelia of blood vessels and endocardium of 9.5 to 18.5 day old mouse embryos. Enhanced tie expression was shown during neovascularization associated with developing ovarian follicles and granulation tissue in skin wounds. Korhonen, et al. Blood 80: 2548-2555 (1992). Thus the TIEs has been suggested to play a role in angiogenesis, which is important for developing treatments for solid tumors and several other angiogenesis-dependent diseases such as diabetic retinopathy, psoriasis, atherosclerosis and arthritis.

[0011] Two structurally related rat TIE receptor proteins have been reported to be encoded by distinct genes with related profiles of expression. One gene, termed tie-1, is the rat homolog of human tie. Maisonpierre, et al., Oncogene 8: 1631-1637 (1993). The other gene, tie-2, may be the rat homolog of the murine tek gene, which, like tie, has been reported to be expressed in the mouse exclusively in endothelial cells and their presumptive progenitors. Dumont, et al. Oncogene 8: 1293-1301 (1993). The human homolog of tie-2 is described in Ziegler, U.S. Pat. No. 5,447,860 which issued on Sep. 5, 1995 (wherein it is referred to as “ork”), which is incorporated in its entirety herein.

[0012] Both genes were found to be widely expressed in endothelial cells of embryonic and postnatal tissues. Significant levels of_tie-2 transcripts were also present in other embryonic cell populations, including lens epithelium, heart epicardium and regions of mesenchyme. Maisonpierre, et al., Oncogene 8: 1631-1637 (1993).

[0013] The predominant expression of the TIE receptor in vascular endothelia suggests that the TIEs plays a role in the development and maintenance of the vascular system. This could include roles in endothelial cell determination, proliferation, differentiation and cell migration and patterning into vascular elements. Analyses of mouse embryos deficient in TIE-2 illustrate its importance in angiogenesis, particularly for vascular network formation in endothelial cells. Sato, T. N., et al., Nature 376:70-74 (1995). In the mature vascular system, the TIEs could function in endothelial cell survival, maintenance and response to pathogenic influences.

[0014] The TIE receptors are also expressed in primitive hematopoietic stem cells, B cells and a subset of megakaryocytic cells, thus suggesting the role of ligands which bind these receptors in early hematopoiesis, in the differentiation and/or proliferation of B cells, and in the megakaryocytic differentiation pathway. Iwama, et al. Biochem. Biophys. Research Communications 195:301-309 (1993); Hashiyama, et al. Blood 87:93-101 (1996), Batard, et al. Blood 87:2212-2220 (1996).

SUMMARY OF THE INVENTION

[0015] The present invention provides for a composition comprising a TIE-2 ligand substantially free of other proteins. The invention also provides for an isolated nucleic acid molecule encoding a TIE-2 ligand. The isolated nucleic acid may be DNA, cDNA or RNA. The invention also provides for a vector comprising an isolated nucleic acid molecule encoding a TIE-2 ligand. The invention further provides for a host-vector system for the production in a suitable host cell of a polypeptide having the biological activity of a TIE-2 ligand. The suitable host cell may be bacterial, yeast, insect or mammalian. The invention also provides for a method of producing a polypeptide having the biological activity of a TIE-2 ligand which comprises growing cells of the host-vector system under conditions permitting production of the polypeptide and recovering the polypeptide so produced.

[0016] The invention herein described of an isolated nucleic acid molecule encoding a TIE-2 ligand further provides for the development of the ligand, a fragment or derivative thereof, or another molecule which is a receptor agonist or antagonist, as a therapeutic for the treatment of patients suffering from disorders involving cells, tissues or organs which express the TIE receptor. The present invention also provides for an antibody which specifically binds such a therapeutic molecule. The antibody may be monoclonal or polyclonal. The invention also provides for a method of using such a monoclonal or polyclonal antibody to measure the amount of the therapeutic molecule in a sample taken from a patient for purposes of monitoring the course of therapy.

[0017] The present invention also provides for an antibody which specifically binds a TIE-2 ligand. The antibody may be monoclonal or polyclonal. Thus the invention further provides for therapeutic compositions comprising an antibody which specifically binds a TIE-2 ligand in a pharmaceutically acceptable vehicle. The invention also provides for a method of blocking blood vessel growth in a mammal by administering an effective amount of a therapeutic composition comprising an antibody which specifically binds a TIE-2 ligand in a pharmaceutically acceptable vehicle.

[0018] The invention further provides for therapeutic compositions comprising a TIE-2 ligand in a pharmaceutically acceptable vehicle. The invention also provides for a method of promoting neovascularization in a patient by administering an effective amount of a therapeutic composition comprising a TIE-2 ligand in a pharmaceutically acceptable vehicle. In one embodiment, the method may be used to promote wound healing. In another embodiment, the method may be used to treat ischemia. In yet another embodiment, a TIE-2 ligand is used, alone or in combination with other hematopoietic factors, to promote the proliferation or differentiation of hematopoietic stem cells, B cells or megakaryocytic cells.

[0019] Alternatively, the invention provides that a TIE-2 ligand may be conjugated to a cytotoxic agent and a therapeutic composition prepared therefrom. The invention further provides for a receptorbody which specifically binds a TIE-2 ligand. The invention further provides for therapeutic compositions comprising a receptorbody which specifically binds a TIE-2 ligand in a pharmaceutically acceptable vehicle. The invention also provides for a method of blocking blood vessel growth in a mammal by administering an effective amount of a therapeutic composition comprising a receptorbody which specifically binds a TIE-2 ligand in a pharmaceutically acceptable vehicle.

[0020] The invention also provides for a TIE-2 receptor antagonist as well as a method of inhibiting TIE-2 ligand biological activity in a mammal comprising administering to the mammal an effective amount of a TIE-2 antagonist. According to the invention, the antagonist may be the TIE-2 ligand 2 as described herein, an antibody or other molecule capable of specifically binding either TIE-2 ligand 1 or TIE-2 receptor (such as for example TIE-2 receptorbody), or ligandbody comprising the fibrinogen-like domain of TIE-2 ligand 1 or ligand 2.

[0021] Throughout this specification the TIE ligands are variously referred to. As used herein, TIE ligand 1=TL1=Angiopoietin-1=Ang1 and TIE ligand 2=TL2=Angiopoietin-2=Ang2.

[0022] One preferred embodiment of the invention is a method of reducing pathological neovascularization in a human comprising administering to the human an agent capable of interfering with Ang2-mediated biological effects such that pathological neovascularization is reduced.

[0023] Another preferred embodiment is a method wherein the pathological neovascularization is ocular neovascularization, in particular wherein the pathological ocular neovascularization is proliferative retinopathy, such as diabetic retinopathy, ischemic retinopathy, or retinopathy of prematurity; choroidal neovascularization; lens neovasculation; corneal neovascularization; iridial neovascularization; or conjunctival neovascularization.

[0024] Yet another preferred embodiment is a method of reducing the risk of retinal detachment associated with pathological ocular neovascularization in a human comprising administering to the human an agent capable of interfering with Ang2-mediated biological effects such that retinal detachment associated with pathological ocular neovascularization is reduced.

[0025] Also preferred are methods wherein the administration is topical, retrobulbar, subconjunctival, oral, subcutaneous, intraocular, intravitreal, subretinal, intramuscular, intranasal, intrathecal, intraarterial, intravenous, transvaginal, transdermal, or transanal administration.

[0026] Still another preferred embodiment is a method wherein the pathological neovascularization is tumor neovascularization, and in particular wherein the tumor is a solid tumor.

[0027] Still another preferred embodiment is a method wherein the administration is topical, oral, subcutaneous, intramuscular, local, intranasal, intrathecal, intraarterial, intravenous, transvaginal, transdermal, or transanal administration.

[0028] Preferred embodiments of the invention are ones in which the biological effects are Ang2-mediated cell signaling, Ang2 binding to Tie1, Ang2 binding to Tie2, or Ang2 antagonism of Ang1.

[0029] Also preferred are methods wherein the agent is capable of interfering with Ang2 binding to Tie1 or Tie2; is capable of interfering with Ang2 mediated cell signaling; or is capable of antagonizing Ang1 binding to Tie1 or Tie2.

[0030] In preferred embodiments of the invention, the method is one in which the agent is an anti-Ang2 antibody or an anti-Tie1 or anti-Tie2 antibody, including monoclonal antibodies.

[0031] Also preferred is a method wherein the cell signaling is Tie2 phosphorylation and the agent is a kinase inhibitor.

[0032] Another preferred embodiment is a method wherein the agent comprises a fragment of an antibody, and in particular, a scFv.

[0033] Other embodiments are ones in which the agent is a Tie1 or Tie2 receptorbody; an antisense molecule, in particular a DNA molecule or a RNA molecule.

[0034] Also preferred are methods of the invention wherein the agent is a small molecule, lipid, protein, or carbohydrate capable of interfering with Ang2 activity.

BRIEF DESCRIPTION OF THE FIGURES

[0035] FIGS. 1A and 1B—TIE-2 receptorbody (TIE-2 RB) inhibits the development of blood vessels in the embryonic chicken chorioallantoic membrane (CAM). A single piece of resorbable gelatin foam (Gelfoam) soaked with 6 μg of RB was inserted immediately under the CAM of 1-day chick embryos. After 3 further days of incubation, 4 day old embryos and surrounding CAM were removed and examined. FIG. 1A: embryos treated with EHK-1 RB (rEHK-1 ecto/hIgG1 Fc) were viable and possessed normally developed blood vessels in their surrounding CAM. FIG. 1B: all embryos treated with TIE-2 RB (r TIE-2 ecto/h IgG1 Fc) were dead, diminished in size and were almost completely devoid of surrounding blood vessels.

[0036]
FIG. 2—Vector pJFE14.

[0037]
FIG. 3—Restriction map of λgt10.

[0038] FIGS. 4A-4B—Nucleic acid and deduced amino acid (single letter code) sequences of human TIE-2 ligand 1 from clone λgt10 encoding htie-2 ligand 1.

[0039] FIGS. 5A-5B—Nucleic acid and deduced amino acid (single letter code) sequences of human TIE-2 ligand 1 from T98G clone.

[0040] FIGS. 6A-6B—Nucleic acid and deduced amino acid (single letter code) sequences of human TIE-2 ligand 2 from clone pBluescript KS encoding human TIE 2 ligand 2.

[0041]
FIG. 7—Western blot showing activation of TIE-2 receptor by TIE-2 ligand 1 (Lane L1) but not by TIE-2 ligand 2 (Lane L2) or control (Mock).

[0042]
FIG. 8—Western blot showing that prior treatment of HAEC cells with excess TIE-2 ligand 2 (Lane 2) antagonizes the subsequent ability of dilute TIE-2 ligand 1 to activate the TIE-2 receptor (TIE2-R) as compared with prior treatment of HAEC cells with MOCK medium (Lane 1).

[0043]
FIG. 9—Western blot demonstrating the ability of TL2 to competitively inhibit TL1 activation of the TIE-2 receptor using the human cell hybrid line, EA.hy926.

[0044] FIGS. 10A-10D—Histogram representation of binding to rat TIE-2 IgG immobilized surface by TIE-2 ligand in C2C12 ras, Rat2 ras, SHEP, and T98G concentrated (10×) conditioned medium. Rat TIE-2 (rTIE2) specific binding is demonstrated by the significant reduction in the binding activity in the presence of 25 μg/ml soluble rat TIE-2 RB as compared to a minor reduction in the presence of soluble trkB RB.

[0045] FIGS. 11A-11B—Binding of recombinant human TIE-2 ligand 1 (hTL1) and human TIE-2 ligand 2 (hTL2), in COS cell supernatants, to a human TIE-2 receptorbody (RB) immobilized surface. Human TIE-2-specific binding was determined by incubating the samples with 25 μg/ml of either soluble human TIE-2 RB or trkB RB; significant reduction in the binding activity is observed only for the samples incubated with human TIE-2 RB.

[0046]
FIG. 12—Western blot showing that TIE-2 receptorbody (denoted TIE-2 RB or, as here, TIE2-Fc) blocks the activation of TIE-2 receptors by TIE-2 ligand 1 (TL1) in HUVEC cells, whereas an unrelated receptorbody (TRKB-Fc) does not block this activation.

[0047]
FIG. 13—Agarose gels showing serial dilutions [undiluted (1) to 10−4]of the TL1 and TL2 RT-PCR products obtained from E14.5 mouse fetal liver (Lanes 1—total, Lanes 3—stromal enriched, and Lanes 4—c-kit+TER119 hematopoietic precursor cells) and E14.5 mouse fetal thymus (Lanes 2—total).

[0048]
FIG. 14—Agarose gels showing serial dilutions [undiluted (1) to 10−3]of the TL1 and TL2 RT-PCR products obtained from E17.5 mouse fetal thymus cortical stromal cells (Lanes 1—CDR1+/A2B5−) and medullary stromal cells (Lane CDR1−/A2B5+).

[0049]
FIG. 15—A schematic representation of the hypothesized role of the TIE-2/TIE ligands in angiogenesis. TL1 is represented by (•), TL2 is represented by (*), TIE-2 is represented by (T), VEGF is represented by ([ ]), and flk-1 (a VEGF receptor) is represented by (Y).

[0050]
FIG. 16—In situ hybridization slides showing the temporal expression pattern of TIE-2, TL1, TL2, and VEGF during angiogenesis associated with follicular development and corpus luteum formation in the ovary of a rat that was treated with pregnant mare serum. Column 1: Early pre-ovulatory follicle; Column 2: pre-ovulatory follicle; Column 3: early corpus luteum; and Column 4: atretic follicle; Row A: bright field; Row B: VEGF; Row C: TL2; Row D: TL1 and Row E: TIE-2 receptor.

[0051]
FIG. 17—Diagrammatic representation of the TIE-2 ligands, showing the “coiled coil” and fibrinogen-like domains and the engineering of multimers of the fibrinogen-like domains using antibodies to myc-tags as well as Fc tagging.

[0052]
FIG. 18—Western blot of the covalent multimeric structure of TL1 and TL2 (Panel A) and the interconversion of TL1 and TL2 by the mutation of one cysteine (Panel B).

[0053]
FIG. 19—A typical curve of TIE-2-IgG binding to immobilized TL1 in a quantitative cell-free binding assay.

[0054]
FIG. 20—A typical curve showing TIE-2 ligand 1 ligandbody comprising the fibrinogen-like domain of the ligand bound to the Fc domain of IgG (TL1-fFc) binding to immobilized TIE-2 ectodomain in a quantitative cell-free binding assay.

[0055] FIGS. 21A-21H: Retinal vascular development in ang2+/− and ang2−/− mice.

[0056]
FIGS. 21A and 21B. Retinas from P10 ang2+/− mice show normal retinal vessels extending all the way to the peripheral edge of the retina.

[0057]
FIGS. 21C and 21D. Retinas from P10 ang2−/− mice show incomplete development of retinal vessels which extend only about half way to the edge of the retina (arrows in c show edge of the retina). Arrows in d point to persistent hyaloid vessels.

[0058]
FIGS. 21E and 21F. Retinas from P18 ang2+/− mice show completed retinal vessel development.

[0059]
FIGS. 21G and 21H. Retinas from P18 ang2−/− mice show incomplete retinal vascular development and continued persistence of the hyaloid vessels (h, arrows). The peripheral edge of the retina is still avascular. Bars=1 mm.

[0060] FIGS. 22A-22P. Persistence of hyaloid vessels and absence of intermediate and deep capillary beds in ang2−/− mice. Ang2+/− and ang2−/− mice stained for lacZ to show sites ang2 promoter activity and GSA which identifies vascular cells. Ang2+/− mice show normal development of the superficial, intermediate, and deep vascular beds and regression of the hyaloid vessels by P10. Ang2−/− mice show a few vessels on the surface of the retina, but rare vessels within the retina indicating failure of angiogenic invasion of the retina. They also show lack of regression of the hyaloid vessels. Bars=100 μm

[0061] FIGS. 23A-23F. Absence of ischemia-induced retinal neovascularization in ang2−/− mice.

[0062] FIGS. 23A-23C. P17 ang2−/− mice with oxygen-induced ischemic retinopathy show marked regression of retinal vessels and no retinal neovascularization. In this section, both the posterior (FIG. 23B) and peripheral (FIG. 23C) retina are lacking in vessels.

[0063] FIGS. 23D-23F. P17 ang2−/− mice that were not exposed to oxygen show a few vessels in the posterior retina (FIG. 23E, arrowheads) and an avascular peripheral retina (FIG. 23F). Bars=100 μm.

[0064] FIGS. 24A-24B. Inhibition of VEGF-induced neovascularization in ang2−/− mice. Mice deficient in Ang2 were crossed with rho/VEGF transgenic mice and F1 ang+/− mice that carried a rho/VEGF transgene were then crossed with ang+/− mice. Retinas were taken at P28, and stained with griffonia simplicifolia lection to visualize the vasculature. Wild-type (ang+/+) (not shown) and ang+/− offspring that carried a rho/VEGF transgene exhibited extensive retinal neovascularization (FIG. 24A). In contrast, ang−/− mice that carried a rho/VEGF transgene had little or no evidence of retinal neovascularization (FIG. 24B).

[0065] FIGS. 25A-25B. The effect of Ang2 on neovascularization in tumors. Lewis lung carcinoma cells were implanted subcutaneously into adult mice as follows: two Ang2 knockout mice, two Ang2 heterozygous mice (Ang2+/−), and two wildtype controls (Ang2+/+). The tumors were allowed to grow for 2½ weeks and then observed qualitatively. The tumors in the Ang2 knockout mice (FIGS. 25B and 25D) appeared to be more than 50% smaller and less well vascularized as compared to the Ang2 heterozygotes (FIG. 25C) and wild type controls (FIG. 25A).

DETAILED DESCRIPTION OF THE INVENTION

[0066] As described in greater detail below, applicants have isolated and identified novel ligands that bind the TIE-2 receptor. The TIE-2 ligands of the present invention, which may be purified from nature, or made recombinantly, are referred to herein as TIE-2 ligand 1 (or TL1 or Ang1) and TIE-2 ligand 2 (or TL2 or Ang2). TIE-2 ligand 1, which has an amino acid sequence which is encoded, inter alia, by the nucleic acid set forth in FIGS. 4A-4B or 5A-5B, is a TIE-2 receptor agonist. TIE-2 ligand 2, which has an amino acid sequence which is encoded, inter alia, by the nucleic acid described in FIGS. 6A-6B, is a TIE-2 receptor antagonist.

[0067] The present invention comprises these TIE-2 ligands, as defined by their amino acid sequences, as well as functionally equivalent variants thereof comprising naturally occurring allelic variations, as well as proteins or peptides comprising substitutions, deletions or insertional mutants of the described sequences, which bind the TIE-2 receptors and act as agonists or antagonists thereof. Such variants include those in which amino acid residues are substituted for residues within the sequence resulting in a silent change. For example, one or more amino acid residues within the sequence can be substituted by another amino acid(s) of a similar polarity which acts as a functional equivalent, resulting in a silent alteration. Substitutes for an amino acid within the sequence may be selected from other members of the class to which the amino acid belongs. For example, the class of nonpolar (hydrophobic) amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. The polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine, and glutamine. The positively charged (basic) amino acids include arginine, lysine and histidine. The negatively charged (acidic) amino acids include aspartic acid and glutamic acid. Also included within the scope of the invention are proteins or fragments or derivatives thereof which exhibit the same or similar biological activity as the TIE-2 ligands described herein, and derivatives which are differentially modified during or after translation, e.g., by glycosylation, proteolytic cleavage, linkage to an antibody molecule or other cellular ligand, etc. Functionally equivalent molecules also include molecules that contain modifications, including N-terminal modifications, which result from expression in a particular recombinant host, such as, for example, N-terminal methylation which occurs in certain bacterial (e.g. E. coli) expression systems. Functional equivalents also include mutants in which amino acid substitutions are made for cysteine molecules to improve stability of the molecules and to prevent unwanted crosslinking. As used herein, the term “TIE-2 ligand” also include fragments of the TIE-2 ligands which are associated with the binding of the ligands to the TIE-2 receptor. In a preferred embodiment, the TIE-2 ligand comprises the fibrinogen-like domain of TIE-2 ligands 1 and 2 as described herein.

[0068] The present invention also encompasses the nucleotide sequence that encodes the proteins described herein as TIE-2 ligands 1 and 2, as well as host cells, including yeast, bacteria, viruses, and mammalian cells, which are genetically engineered to produce the protein, by e.g. transfection, transduction, infection, electroporation, or microinjection of nucleic acid encoding the TIE-2 ligands described herein in a suitable expression vector. The present invention also encompasses introduction of the nucleic acids encoding TIE-2 ligands through gene therapy techniques such as is described, for example, in Finkel and Epstein FASEB J. 9:843-851 (1995); Guzman, et al. PNAS (USA) 91:10732-10736 (1994).

[0069] One skilled in the art will also recognize that the present invention encompasses DNA and RNA sequences that hybridize to a deduced TIE-2 ligand encoding sequence, under conditions of moderate stringency, as defined in, for example, Sambrook, et al. Molecular Cloning: A Laboratory Manual, 2 ed. Vol. 1, pp. 101-104, Cold Spring Harbor Laboratory Press (1989). Thus, a nucleic acid molecule contemplated by the invention includes one having a sequence deduced from an amino acid sequence of a TIE-2 ligand prepared as described herein, as well as a molecule having a sequence of nucleic acids that hybridizes to such a nucleic acid sequence, and also a nucleic acid sequence which is degenerate of the above sequences as a result of the genetic code, but which encodes a ligand that binds the TIE-2 receptor and which has an amino acid sequence and other primary, secondary and tertiary characteristics that are sufficiently duplicative of the ligands described herein so as to confer on the molecule the same biological activity as one of the TIE-2 ligands described herein.

[0070] Accordingly, the present invention encompasses an isolated and purified nucleic acid molecule comprising a nucleic acid sequence encoding a human TIE-2 ligand, wherein the nucleic acid sequence is selected from the group consisting of:

[0071] (a) the nucleic acid sequence comprising the coding region of the human TIE-2 ligand as set forth in FIGS. 4A-4B, FIGS. 5A-5B or FIGS. 6A-6B;

[0072] (b) a nucleic acid sequence that hybridizes under moderately stringent conditions to the nucleic acid sequence of (a) and which encodes a TIE-2 ligand that binds TIE-2 receptor; and

[0073] (c) a nucleic acid sequence that is degenerate as a result of the genetic code to a nucleic acid sequence of (a) or (b), and which encodes a TIE-2 ligand that binds TIE-2 receptor.

[0074] The present invention further provides for an isolated and purified human TIE-2 ligand encoded by an isolated nucleic acid molecule of the invention. The invention also provides a vector which comprises an isolated nucleic acid molecule comprising a nucleic acid sequence encoding a human TIE-2 ligand. In one embodiment, the vector is designated as pBluescript KS encoding human TIE 2 ligand 2.

[0075] Any of the methods known to one skilled in the art for the insertion of DNA fragments into a vector may be used to construct expression vectors encoding TIE-2 ligands using appropriate transcriptional/translational control signals and the protein coding sequences. These methods may include in vitro recombinant DNA and synthetic techniques and in vivo recombinations (genetic recombination). Expression of a nucleic acid sequence encoding a TIE-2 ligand or peptide fragments thereof may be regulated by a second nucleic acid sequence which is operably linked to the TIE-2 ligand encoding sequence such that the TIE-2 ligand protein or peptide is expressed in a host transformed with the recombinant DNA molecule. For example, expression of a TIE-2 ligand described herein may be controlled by any promoter/enhancer element known in the art. Promoters which may be used to control expression of the ligand include, but are not limited to the long terminal repeat as described in Squinto et al., (Cell 65:1-20 (1991)); the SV40 early promoter region (Bernoist and Chambon, Nature 290:304-310), the CMV promoter, the M-MuLV 5′ terminal repeat, the promoter contained in the 3′ long terminal repeat of Rous sarcoma virus (Yamamoto, et al., Cell 22:787-797 (1980)), the herpes thymidine kinase promoter (Wagner et al., Proc. Natl. Acad. Sci. U.S.A. 78:144-1445 (1981)), the adenovirus promoter, the regulatory sequences of the metallothionein gene (Brinster et al., Nature 296:39-42 (1982)); prokaryotic expression vectors such as the β-lactamase promoter (Villa-Kamaroff, et al., Proc. Natl. Acad. Sci. U.S.A. 75:3727-3731 (1978)), or the tac promoter (DeBoer, et al., Proc. Natl. Acad. Sci. U.S.A. 80:21-25 (1983)), see also “Useful proteins from recombinant bacteria” in Scientific American, 242:74-94 (1980); promoter elements from yeast or other fungi such as the Gal 4 promoter, the ADH (alcohol dehydrogenase) promoter, PGK (phosphoglycerol kinase) promoter, alkaline phosphatase promoter, and the following animal transcriptional control regions, which exhibit tissue specificity and have been utilized in transgenic animals; elastase I gene control region which is active in pancreatic acinar cells (Swift et al., Cell 38:639-646 (1984); Ornitz et al., Cold Spring Harbor Symp. Quant. Biol. 50:399-409 (1986); MacDonald, Hepatology 7:425-515 (1987); insulin gene control region which is active in pancreatic beta cells [Hanahan, Nature 315:115-122 (1985)]; immunoglobulin gene control region which is active in lymphoid cells (Grosschedl et al., 1984, Cell 38:647-658; Adames et al., 1985, Nature 318:533-538; Alexander et al., 1987, Mol. Cell. Biol. 7:1436-1444), mouse mammary tumor virus control region which is active in testicular, breast, lymphoid and mast cells (Leder et al., 1986, Cell 45:485-495), albumin gene control region which is active in liver (Pinkert et al., 1987, Genes and Devel. 1:268-276), alpha-fetoprotein gene control region which is active in liver (Krumlauf et al., 1985, Mol. Cell. Biol. 5:1639-1648; Hammer et al., 1987, Science 235:53-58); alpha 1-antitrypsin gene control region which is active in the liver (Kelsey et al, 1987, Genes and Devel. 1:161-171), beta-globin gene control region which is active in myeloid cells (Mogram et al., 1985, Nature 315:338-340; Kollias et al., 1986, Cell 46:89-94); myelin basic protein gene control region which is active in oligodendrocytes in the brain (Readhead et al., 1987, Cell 48:703-712); myosin light chain-2 gene control region which is active in skeletal muscle (Shani, 1985, Nature 314:283-286), and gonadotropic releasing hormone gene control region which is active in the hypothalamus (Mason et al., 1986, Science 234:1372-1378). The invention further encompasses the production of antisense compounds which are capable of specifically hybridizing with a sequence of RNA encoding a TIE-2 ligand to modulate its expression. Ecker, U.S. Pat. No. 5,166,195, issued Nov. 24, 1992.

[0076] Thus, according to the invention, expression vectors capable of being replicated in a bacterial or eukaryotic host comprising a nucleic acid encoding a TIE-2 ligand as described herein, are used to transfect a host and thereby direct expression of such nucleic acid to produce the TIE-2 ligand, which may then be recovered in a biologically active form. As used herein, a biologically active form includes a form capable of binding to the TIE-2 receptor and causing a biological response such as a differentiated function or influencing the phenotype of the cell expressing the receptor. Such biologically active forms would, for example, induce phosphorylation of the tyrosine kinase domain of the TIE-2 receptor. Alternatively, the biological activity may be an effect as an antagonist to the TIE-2 receptor, such as is the biological activity of TIE-2 ligand 2. In alternative embodiments, the active form of a TIE-2 ligand is one that can recognize the TIE-2 receptor and thereby act as a targeting agent for the receptor for use in both diagnostics and therapeutics. In accordance with such embodiments, the active form need not confer upon any TIE-2 expressing cell any change in phenotype.

[0077] In an alternative embodiment, the biologically active form of the TIE-2 ligand is one in which the ligand is capable of binding the TIE-1 receptor. Preliminary data indicates that TL 2 binds TIE-1 receptor (albeit with low affinity) raising the possibility that it may be able to bind and activate the receptor, or, as in the case of the TIE-2 receptor, bind and act as antagonist.

[0078] Expression vectors containing the gene inserts can be identified by four general approaches: (a) DNA-DNA hybridization, (b) presence or absence of “marker” gene functions, (c) expression of inserted sequences and (d) PCR detection. In the first approach, the presence of a foreign gene inserted in an expression vector can be detected by DNA-DNA hybridization using probes comprising sequences that are homologous to an inserted TIE-2 ligand encoding gene. In the second approach, the recombinant vector/host system can be identified and selected based upon the presence or absence of certain “marker” gene functions (e.g., thymidine kinase activity, resistance to antibiotics, transformation phenotype, occlusion body formation in baculovirus, etc.) caused by the insertion of foreign genes in the vector. For example, if a nucleic acid encoding a TIE-2 ligand is inserted within the marker gene sequence of the vector, recombinants containing the insert can be identified by the absence of the marker gene function. In the third approach, recombinant expression vectors can be identified by assaying the foreign gene product expressed by the recombinant. Such assays can be based, for example, on the physical or functional properties of a TIE-2 ligand gene product, for example, by binding of the ligand to the TIE-2 receptor or portion thereof which may be tagged with, for example, a detectable antibody or portion thereof or by binding to antibodies produced against the TIE-2 ligand protein or a portion thereof. Cells of the present invention may transiently or, preferably, constitutively and permanently express TIE-2 ligands as described herein. In the fourth approach, DNA nucleotide primers can be prepared corresponding to a tie-2 specific DNA sequence. These primers could then be used to PCR a tie-2 gene fragment. (PCR Protocols: A Guide To Methods and Applications, Edited by Michael A. Innis et al., Academic Press (1990)).

[0079] The recombinant ligands may be purified by any technique which allows for the subsequent formation of a stable, biologically active protein. Preferably, the ligands are secreted into the culture medium from which they are recovered. Alternatively, the ligand may be recovered from cells either as soluble proteins or as inclusion bodies, from which they may be extracted quantitatively by 8M guanidinium hydrochloride and dialysis in accordance with well known methodology. In order to further purify the ligand, affinity chromatography, conventional ion exchange chromatography, hydrophobic interaction chromatography, reverse phase chromatography or gel filtration may be used.

[0080] In additional embodiments of the invention, as described in greater detail in the Examples, a recombinant TIE-2 ligand encoding gene may be used to inactivate or “knock out” the endogenous gene by homologous recombination, and thereby create a TIE-2 ligand deficient cell, tissue, or animal. For example, and not by way of limitation, the recombinant TIE-2 ligand encoding gene may be engineered to contain an insertional mutation, for example the neo gene, which would inactivate the native TIE-2 ligand encoding gene. Such a construct, under the control of a suitable promoter, may be introduced into a cell, such as an embryonic stem cell, by a technique such as transfection, transduction, or injection. Cells containing the construct may then be selected by G418 resistance. Cells which lack an intact TIE-2 ligand encoding gene may then be identified, by Southern blotting, PCR detection, Northern blotting or assay of expression. Cells lacking an intact TIE-2 ligand encoding gene may then be fused to early embryo cells to generate transgenic animals deficient in such ligand. Such an animal may be used to define specific in vivo processes, normally dependent upon the ligand.

[0081] The present invention also provides for antibodies to the TIE-2 ligands described herein which are useful for detection of the ligands in, for example, diagnostic applications. For preparation of monoclonal antibodies directed toward TIE-2 ligand, any technique which provides for the production of antibody molecules by continuous cell lines in culture may be used. For example, the hybridoma technique originally developed by Kohler and Milstein (1975, Nature 256:495-497), as well as the trioma technique, the human B-cell hybridoma technique (Kozbor et al., 1983, Immunology Today 4:72), and the EBV-hybridoma technique to produce human monoclonal antibodies (Cole et al., 1985, in “Monoclonal Antibodies and Cancer Therapy,” Alan R. Liss, Inc. pp. 77-96) and the like are within the scope of the present invention.

[0082] The monoclonal antibodies may be human monoclonal antibodies or chimeric human-mouse (or other species) monoclonal antibodies. Human monoclonal antibodies may be made by any of numerous techniques known in the art (e.g., Teng et al., 1983, Proc. Natl. Acad. Sci. U.S.A. 80:7308-7312; Kozbor et al., 1983, Immunology Today 4:72-79; Olsson et al., 1982, Meth. Enzymol. 92:3-16). Chimeric antibody molecules may be prepared containing a mouse antigen-binding domain with human constant regions (Morrison et al., 1984, Proc. Natl. Acad. Sci. U.S.A. 81:6851, Takeda et al., 1985, Nature 314:452).

[0083] Various procedures known in the art may be used for the production of polyclonal antibodies to epitopes of the TIE-2 ligands described herein. For the production of antibody, various host animals, including but not limited to rabbits, mice and rats can be immunized by injection with a TIE-2 ligand, or a fragment or derivative thereof. Various adjuvants may be used to increase the immunological response, depending on the host species, and including but not limited to Freund's (complete and incomplete), mineral gels such as aluminum hydroxide, surface active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanins, dinitrophenol, and potentially useful human adjuvants such as BCG (Bacille Calmette-Guerin) and Corynebacterium parvum.

[0084] A molecular clone of an antibody to a selected TIE-2 ligand epitope can be prepared by known techniques. Recombinant DNA methodology (see e.g., Maniatis et al., 1982, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.) may be used to construct nucleic acid sequences which encode a monoclonal antibody molecule, or antigen binding region thereof.

[0085] The present invention provides for antibody molecules as well as fragments of such antibody molecules. Antibody fragments which contain the idiotype of the molecule can be generated by known techniques. For example, such fragments include but are not limited to: the F(ab′)2 fragment which can be produced by pepsin digestion of the antibody molecule; the Fab′ fragments which can be generated by reducing the disulfide bridges of the F(ab′)2 fragment, and the Fab fragments which can be generated by treating the antibody molecule with papain and a reducing agent. Also included are single chain Fvs. A single chain Fv (scFv) is a truncated Fab having only the V region of a heavy chain linked by a stretch of synthetic peptide to a V region of a light chain. See, for example, U.S. Pat. Nos. 5,565,332; 5,733,743; 5,837,242; 5,858,657; and 5,871,907 assigned to Cambridge Antibody Technology Limited incorporated by reference herein. Antibody molecules may be purified by known techniques, e.g., immunoabsorption or immunoaffinity chromatography, chromatographic methods such as HPLC (high performance liquid chromatography), or a combination thereof.

[0086] The present invention further encompasses an immunoassay for measuring the amount of a TIE-2 ligand in a biological sample by

[0087] a) contacting the biological sample with at least one antibody which specifically binds the TIE-2 ligand so that the antibody forms a complex with any TIE-2 ligand present in the sample; and

[0088] b) measuring the amount of the complex and thereby measuring the amount of the TIE-2 ligand in the biological sample. The invention further encompasses an assay for measuring the amount of TIE-2 receptor in a biological sample by

[0089] a) contacting the biological sample with at least one ligand of the invention so that the ligand forms a complex with the TIE-2 receptor; and

[0090] b) measuring the amount of the complex and thereby measuring the amount of the TIE-2 receptor in the biological sample.

[0091] The present invention also provides for the utilization of a TIE-2 ligand to support the survival and/or growth and/or migration and/or differentiation of TIE-2 receptor expressing cells. Thus, the ligand may be used as a supplement to support, for example, endothelial cells in culture.

[0092] Further, the discovery by applicants of a cognate ligand for the TIE-2 receptor enables the utilization of assay systems useful for the identification of agonists or antagonists of the TIE-2 receptor. Such assay systems would be useful in identifying molecules capable of promoting or inhibiting angiogenesis. For example, in one embodiment, antagonists of the TIE-2 receptor may be identified as test molecules that are capable of interfering with the interaction of the TIE-2 receptor with a biologically active TIE-2 ligand. Such antagonists are identified by their ability to 1) block the binding of a biologically active TIE-2 ligand to the receptor as measured, for example, using BIAcore biosensor technology (BIAcore; Pharmacia Biosensor, Piscataway, N.J.); or 2) block the ability of a biologically active TIE-2 ligand to cause a biological response. Such biological responses include, but are not limited to, phosphorylation of the TIE-2 receptor or downstream components of the TIE-2 signal transduction pathway, or survival, growth or differentiation of TIE-2 receptor bearing cells.

[0093] In one embodiment, cells engineered to express the TIE-2 receptor may be dependent for growth on the addition of TIE-2 ligand. Such cells provide useful assay systems for identifying additional agonists of the TIE-2 receptor, or antagonists capable of interfering with the activity of TIE-2 ligand on such cells. Alternatively, autocrine cells, engineered to be capable of co-expressing both TIE-2 ligand and receptor, may provide useful systems for assaying potential agonists or antagonists.

[0094] Therefore, the present invention provides for introduction of the TIE-2 receptor into cells that do not normally express this receptor, thus allowing these cells to exhibit profound and easily distinguishable responses to a ligand which binds this receptor. The type of response elicited depends on the cell utilized, and not the specific receptor introduced into the cell. Appropriate cell lines can be chosen to yield a response of the greatest utility for assaying, as well as discovering, molecules that can act on tyrosine kinase receptors. The molecules may be any type of molecule, including but not limited to peptide and non-peptide molecules, that will act in systems to be described in a receptor specific manner.

[0095] One of the more useful systems to be exploited involves the introduction of the TIE-2 receptor (or a chimeric receptor comprising the extracellular domain of another receptor tyrosine kinase such as, for example, trkC and the intracellular domain of a TIE receptor) into a fibroblast cell line (e.g., NIH3T3 cells) thus such a receptor which does not normally mediate proliferative or other responses can, following introduction into fibroblasts, nonetheless be assayed by a variety of well established methods to quantitate effects of fibroblast growth factors (e.g. thymidine incorporation or other types of proliferation assays; see van Zoelen, 1990, “The Use of Biological Assays For Detection Of Polypeptide Growth Factors” in Progress Factor Research, Vol. 2, pp. 131-152; Zhan and M. Goldfarb, 1986, Mol. Cell. Biol., Vol. 6, pp. 3541-3544). These assays have the added advantage that any preparation can be assayed both on the cell line having the introduced receptor as well as the parental cell line lacking the receptor; only specific effects on the cell line with the receptor would be judged as being mediated through the introduced receptor. Such cells may be further engineered to express the TIE-2 ligands, thus creating an autocrine system useful for assaying for molecules that act as antagonists/agonists of this interaction. Thus, the present invention provides for host cells comprising nucleic acid encoding TIE-2 ligands and nucleic acid encoding TIE-2 receptor.

[0096] The TIE-2 receptor/TIE-2 ligand interaction also provides a useful system for identifying small molecule agonists or antagonists of the TIE-2 receptor. For example, fragments, mutants or derivatives of a TIE-2 ligand may be identified that bind the TIE-2 receptor but do not induce any other biological activity. Alternatively, the characterization of a TIE-2 ligand enables the determination of active portions of the molecule. Further, the identification of a ligand enables the determination of the X-ray crystal structure of the receptor/ligand complex, thus enabling identification of the binding site on the receptor. Knowledge of the binding site will provide useful insight into the rational design of novel agonists and antagonists.

[0097] The specific binding of a test molecule to the TIE-2 receptor may be measured in a number of ways. For example, the actual binding of test molecule to cells expressing TIE-2 may be detected or measured, by detecting or measuring (i) test molecule bound to the surface of intact cells; (ii) test molecule cross-linked to TIE-2 protein in cell lysates; or (iii) test molecule bound to TIE-2 in vitro. The specific interaction between test molecule and TIE-2 may be evaluated by using reagents that demonstrate the unique properties of that interaction.

[0098] As a specific, nonlimiting example, the methods of the invention may be used as follows. Consider a case in which the TIE-2 ligand in a sample is to be measured. Varying dilutions of the sample (the test molecule), in parallel with a negative control (NC) containing no TIE-2 ligand activity, and a positive control (PC) containing a known amount of a TIE-2 ligand, may be exposed to cells that express TIE-2 in the presence of a detectably labeled TIE-2 ligand (in this example, radioiodinated ligand). The amount of TIE-2 ligand in the test sample may be evaluated by determining the amount of 125I-labeled TIE-2 ligand that binds to the controls and in each of the dilutions, and then comparing the sample values to a standard curve. The more TIE-2 ligand in the sample, the less 125I-ligand that will bind to TIE-2.

[0099] The amount of 125I-ligand bound may be determined by measuring the amount of radioactivity per cell, or by cross-linking the TIE-2 ligand to cell surface proteins using DSS, as described in Meakin and Shooter, 1991, Neuron 6:153-163, and detecting the amount of labeled protein in cell extracts using, for example, SDS polyacrylamide gel electrophoresis, which may reveal a labeled protein having a size corresponding to TIE-2 receptor/TIE-2 ligand. The specific test molecule/TIE-2 interaction may further be tested by adding to the assays various dilutions of an unlabeled control ligand that does not bind the TIE-2 receptor and therefore should have no-substantial affect on the competition between labeled TIE-2 ligand and test molecule for TIE-2 binding. Alternatively, a molecule known to be able to disrupt TIE-2 receptor/TIE-2 ligand binding, such as, but not limited to, anti-TIE-2 antibody, or TIE-2 receptorbody as described herein, may be expected to interfere with the competition between 125I-TIE-2 ligand and test molecule for TIE-2 receptor binding.

[0100] Detectably labeled TIE-2 ligand includes, but is not limited to, TIE-2 ligand linked covalently or noncovalently to a radioactive substance, a fluorescent substance, a substance that has enzymatic activity, a substance that may serve as a substrate for an enzyme (enzymes and substrates associated with colorimetrically detectable reactions are preferred) or to a substance that can be recognized by an antibody molecule that is preferably a detectably labeled antibody molecule.

[0101] Alternatively, the specific binding of test molecule to TIE-2 may be measured by evaluating the secondary biological effects of TIE-2 ligand/TIE-2 receptor binding, including, but not limited to, cell growth and/or differentiation or immediate early gene expression or phosphorylation of TIE-2. For example, the ability of the test molecule to induce differentiation can be tested in cells that lack tie-2 and in comparable cells that express tie-2; differentiation in tie-2-expressing cells but not in comparable cells that lack tie-2 would be indicative of a specific test molecule/TIE-2 interaction. A similar analysis could be performed by detecting immediate early gene (e.g. fos and jun) induction in tie-2-minus and tie-2-plus cells, or by detecting phosphorylation of TIE-2 using standard phosphorylation assays known in the art. Such analysis might be useful in identifying agonists or antagonists that do not competitively bind to TIE-2.

[0102] Similarly, the present invention provides for a method of identifying an molecule that has the biological activity of a TIE-2 ligand comprising (i) exposing a cell that expresses tie-2 to a test molecule and (ii) detecting the specific binding of the test molecule to TIE-2 receptor, in which specific binding to TIE-2 positively correlates with TIE-2 like activity. Specific binding may be detected by either assaying for direct binding or the secondary biological effects of binding, as discussed supra. Such a method may be particularly useful in identifying new members of the TIE ligand family or, in the pharmaceutical industry, in screening a large array of peptide and non-peptide molecules (e.g., peptidomimetics) for TIE associated biological activity. In a preferred, specific, nonlimiting embodiment of the invention, a large grid of culture wells may be prepared that contain, in alternate rows, PC12 (or fibroblasts, see infra) cells that are either tie-2-minus or engineered to be tie-2-plus. A variety of test molecules may then be added such that each column of the grid, or a portion thereof, contains a different test molecule. Each well could then be scored for the presence or absence of growth and/or differentiation. An extremely large number of test molecules could be screened for such activity in this manner.

[0103] In additional embodiments, the invention provides for methods of detecting or measuring TIE ligand-like activity or identifying a molecule as having such activity comprising (i) exposing a test molecule to a TIE-2 receptor protein in vitro under conditions that permit binding to occur and (ii) detecting binding of the test molecule to the TIE-2 receptor protein, in which binding of test molecule to TIE-2 receptor correlates with TIE ligand-like activity. According to such methods, the TIE-2 receptor may or may not be substantially purified, may be affixed to a solid support (e.g. as an affinity column or as an ELISA assay), or may be incorporated into an artificial membrane. Binding of test molecule to TIE-2 receptor may be evaluated by any method known in the art. In preferred embodiments, the binding of test molecule may be detected or measured by evaluating its ability to compete with detectably labeled known TIE-2 ligands for TIE-2 receptor binding.

[0104] The present invention also provides for a method of detecting the ability of a test molecule to function as an antagonist of TIE ligand-like activity comprising detecting the ability of the molecule to inhibit an effect of TIE ligand binding to TIE-2 receptor on a cell that expresses the receptor. Such an antagonist may or may not interfere with TIE-2 receptor/TIE-2 ligand binding. Effects of TIE-2 ligand binding to TIE-2 receptor are preferably biological or biochemical effects, including, but not limited to, cell survival or proliferation, cell transformation, immediate early gene induction, or TIE-2 phosphorylation.

[0105] The invention further provides for both a method of identifying antibodies or other molecules capable of neutralizing the ligand or blocking binding to the receptor, as well as the molecules identified by the method. By way of nonlimiting example, the method may be performed via an assay which is conceptually similar to an ELISA assay. For example, TIE-2 receptorbody may be bound to a solid support, such as a plastic multiwell plate. As a control, a known amount of TIE-2 ligand which has been Myc-tagged may then be introduced to the well and any tagged TIE ligand which binds the receptorbody may then be identified by means of a reporter antibody directed against the Myc-tag. This assay system may then be used to screen test samples for molecules which are capable of i) binding to the tagged ligand or ii) binding to the receptorbody and thereby blocking binding to the receptorbody by the tagged ligand. For example, a test sample containing a putative molecule of interest together with a known amount of tagged ligand may be introduced to the well and the amount of tagged ligand which binds to the receptorbody may be measured. By comparing the amount of bound tagged ligand in the test sample to the amount in the control, samples containing molecules which are capable of blocking ligand binding to the receptor may be identified. The molecules of interest thus identified may be isolated using methods well known to one of skill in the art.

[0106] Once a blocker of ligand binding is found, one of skill in the art would know to perform secondary assays to determine whether the blocker is binding to the receptor or to the ligand, as well as assays to determine if the blocker molecule can neutralize the biological activity of the ligand. For example, by using a binding assay which employs BIAcore biosensor technology (or the equivalent), in which either TIE-2 receptorbody or TIE-2 ligand or ligandbody is covalently attached to a solid support (e.g. carboxymethyl dextran on a gold surface), one of skill in the art would be able to determine if the blocker molecule is binding specifically to the ligand, ligandbody or to the receptorbody. To determine if the blocker molecule can neutralize the biological activity of the ligand, one of skill in the art could perform a phosphorylation assay (see Example 5) or alternatively, a functional bioassay, such as a survival assay, by using primary cultures of, for example, endothelial cells. Alternatively, a blocker molecule which binds to the receptorbody could be an agonist and one of skill in the art would know to how to determine this by performing an appropriate assay for identifying additional agonists of the TIE-2 receptor.

[0107] In addition, the invention further contemplates compositions wherein the TIE-2 ligand is the receptor binding domains of the TIE-2 ligands described herein. For example, TIE-2 ligand 1 consists of a “coiled coil” domain (beginning at the 5′ end and extending to the nucleotide at about position 1160 of FIGS. 4A-4B and about position 1157 of FIGS. 5A-5B) and a fibrinogen-like domain (which is encoded by the nucleotide sequence of FIGS. 4A-4B beginning at about position 1161 and about position 1158 of FIGS. 5A-5B). The fibrinogen-like domain of TIE-2 ligand 2 is believed to begin on or around the same amino acid sequence as in ligand 1 (FRDCA) which is encoded by nucleotides beginning around 1197 of FIGS. 6A-6B. Multimerization of the coiled coil domains during production of the ligand hampers purification. As described in Example 19, Applicants have discovered, however, that the fibrinogen-like domain comprises the TIE-2 receptor binding domain. The monomeric forms of the fibrinogen-like domain do not, however, appear to bind the receptor. Studies utilizing myc-tagged fibrinogen-like domain, which has been “clustered” using anti-myc antibodies, do bind the TIE-2 receptor. [Methods of production of “clustered ligands and ligandbodies are described in Davis, et al. Science 266:816-819 (1994)]. Based on these finding, applicants produced “ligandbodies” which comprise the fibrinogen-like domain of the TIE-2 ligands coupled to the Fc domain of IgG (“fFc's”). These ligandbodies, which form dimers, efficiently bind the TIE-2 receptor. Accordingly, the present invention contemplates the production of TIE-2 ligandbodies which may be used as targeting agents, in diagnostics or in therapeutic applications, such as targeting agents for tumors and/or associated vasculature wherein a TIE-2 antagonist is indicated.

[0108] The invention herein further provides for the development of the ligand, a fragment or derivative thereof, or another molecule which is a receptor agonist or antagonist, as a therapeutic for the treatment of patients suffering from disorders involving cells, tissues or organs which express the TIE-2 receptor. Such molecules may be used in a method of treatment of the human or animal body, or in a method of diagnosis. Because TIE-2 receptor has been identified in association with endothelial cells and, as demonstrated herein, blocking of TIE-2 ligand 1 appears to prevent vascularization, applicants have demonstrated that the TIE-2 ligand 1 will be useful for the induction of vascularization in diseases or disorders where such vascularization is indicated. Such diseases or disorders would include wound healing, ischaemia and diabetes. The ligands may be tested in animal models and used therapeutically as described for other agents, such as vascular endothelial growth factor (VEGF), another endothelial cell-specific factor that is angiogenic. Ferrara, et al. U.S. Pat. No. 5,332,671 issued Jul. 26, 1994. The Ferrara reference, as well as other studies, describe in vitro and in vivo studies that may be used to demonstrate the effect of an angiogenic factor in enhancing blood flow to ischemic myocardium, enhancing wound healing, and in other therapeutic settings wherein neoangiogenesis is desired. [see Sudo, et al. European Patent Application 0 550 296 A2 published Jul. 7, 1993; Banai, et al. Circulation 89:2183-2189 (1994); Unger, et al. Am. J. Physiol. 266:H1588-H1595 (1994); Lazarous, et al. Circulation 91:145-153 (1995)]. According to the invention, the TIE-2 ligands may be used alone or in combination with one or more additional pharmaceutically active compounds such as, for example, VEGF or basic fibroblast growth factor (bFGF), as well as cytokines, neurotrophins, etc.

[0109] Conversely, antagonists of the TIE-2 receptor, such as receptorbodies as described herein in Examples 2 and 3, and TIE-2 ligand 2 as described in Example 9, would be useful to prevent or attenuate vascularization, thus preventing or attenuating, for example, tumor growth. These agents may be used alone or in combination with other compositions, such as anti-VEGF antibodies, that have been shown to be useful in treating conditions in which the therapeutic intent is to block angiogenesis. The TIE-2 ligands described herein may also be used in combination with agents, such as cytokine antagonists such as IL-6 antagonists, that are known to block inflammation.

[0110] For example, applicants have determined that the TIE-2 ligands are expressed in cells within, or closely associated with, tumors. TIE-2 ligand 2 appears to be tightly associated with tumor endothelial cells. Accordingly, TIE-2 antagonists, such as TIE-2 ligand 2 or TIE-2 ligandbodies may also be useful in preventing or attenuating, for example, tumor growth. In addition, TIE-2 ligands or ligandbodies may be useful for the delivery of toxins to a receptor bearing cell. Alternatively, other molecules, such as growth factors, cytokines or nutrients, may be delivered to a TIE-2 receptor bearing cell via TIE-2 ligands or ligandbodies. TIE-2 ligands or ligandbodies could also be used as a diagnostic reagent for the TIE-2 receptor, to detect the receptor in vivo or in vitro. Where the TIE-2 receptor is associated with a disease state, TIE-2 ligands or ligandbodies may be useful as diagnostic reagents for detecting the disease by, for example, tissue staining or whole body imaging. Such reagents include radioisotopes, flurochromes, dyes, enzymes and biotin. Such diagnostics or targeting agents may be prepared as described in Alitalo, et al. WO 95/26364 published Oct. 5, 1995 and Burrows, F. and P. Thorpe, PNAS (USA) 90:8996-9000 (1993) which is incorporated herein in its entirety.

[0111] In other embodiments, the TIE-2 ligands described herein are used as hematopoietic factors. A variety of hematopoietic factors and their receptors are involved in the proliferation and/or differentiation and/or migration of the various cells types contained within blood. Because the TIE-2 receptors are expressed in early hematopoietic cells, the TIE-2 ligands are expected to play a comparable role in the proliferation or differentiation or migration of these cells. Thus, for example, TIE-2 containing compositions may be prepared, assayed, examined in in vitro and in vivo biological systems and used therapeutically as described in any of the following: Sousa, U.S. Pat. No. 4,810,643, Lee, et al., Proc. Natl. Acad. Sci. USA 82:4360-4364 (1985) Wong, et al. Science, 228:810-814 (1985); Yokota, et al. Proc. Natl. Acad. Sci (USA) 81:1070 (1984); Bosselman, et al. WO 9105795 published May 2, 1991 entitled “Stem Cell Factor” and Kirkness, et al. WO 95/19985 published Jul. 27, 1995 entitled “Haemopoietic Maturation Factor”. Accordingly, the ligands may be used to diagnose or treat conditions in which normal hematopoiesis is suppressed, including, but not limited to anemia, thrombocytopenia, leukopenia and granulocytopenia. In a preferred embodiment, the TIE-2 ligands may be used to stimulate differentiation of blood cell precursors in situations where a patient has a disease, such as acquired immune deficiency syndrome (AIDS) which has caused a reduction in normal blood cell levels, or in clinical settings in which enhancement of hematopoietic populations is desired, such as in conjunction with bone marrow transplant, or in the treatment of aplasia or myelosuppression caused by radiation, chemical treatment or chemotherapy.

[0112] The TIE-2 ligands of the present invention may be used alone, or in combination with another pharmaceutically active agent such as, for example, cytokines, neurotrophins, interleukins, etc. In a preferred embodiment, the ligands may be used in conjunction with any of a number of the above referenced factors which are known to induce stem cell or other hematopoietic precursor proliferation, or factors acting on later cells in the hematopoietic pathway, including, but not limited to, hemopoietic maturation factor, thrombopoietin, stem cell factor, erythropoietin, G-CSF, GM-CSF, etc.

[0113] In an alternative embodiment, TIE-2 receptor antagonists, such as TL2, are used to diagnose or treat patients in which the desired result is inhibition of a hematopoietic pathway, such as for the treatment of myeloproliferative or other proliferative disorders of blood forming organs such as thrombocythemias, polycythemias and leukemias. In such embodiments, treatment may comprise use of a therapeutically effective amount of the TIE-2 ligand 2, TIE-2 antibody, TIE-2 receptorbody, a conjugate of a TIE-2 ligand or a ligandbody or fFC as described herein.

[0114] The present invention also provides for pharmaceutical compositions comprising the TIE-2 ligands or ligandbodies described herein, peptide fragments thereof, or derivatives in a pharmacologically acceptable vehicle. The TIE-2 ligand proteins, peptide fragments, or derivatives may be administered systemically or locally. Any appropriate mode of administration known in the art may be used, including, but not limited to, intravenous, intrathecal, intraarterial, intranasal, oral, subcutaneous, intraperitoneal, or by local injection or surgical implant. Sustained release formulations are also provided for.

[0115] The present invention also provides for an antibody which specifically binds such a therapeutic molecule. The antibody may be monoclonal or polyclonal. The invention also provides for a method of using such a monoclonal or polyclonal antibody to measure the amount of the therapeutic molecule in a sample taken from a patient for purposes of monitoring the course of therapy.

[0116] The invention further provides for a therapeutic composition comprising a human TIE-2 ligand or ligandbody and a cytotoxic agent conjugated thereto. In one embodiment, the cytotoxic agent may be a radioisotope or toxin.

[0117] The invention also provides for an antibody which specifically binds a human TIE-2 ligand. The antibody may be monoclonal or polyclonal. The invention further provides for a method of purifying a human TIE-2 ligand comprising:

[0118] a) coupling at least one TIE-2 binding substrate to a solid matrix;

[0119] b) incubating the substrate of a) with a cell lysate so that the substrate forms a complex with any human TIE-2 ligand in the cell lysate;

[0120] c) washing the solid matrix; and

[0121] d) eluting the human TIE-2 ligand from the coupled substrate.

[0122] The substrate may be any substance that specifically binds the human TIE-2 ligand. In one embodiment, the substrate is selected from the group consisting of anti-TIE-2 ligand antibody, TIE-2 receptor and TIE-2 receptorbody. The invention further provides for a receptorbody which specifically binds a human TIE-2 ligand, as well as a therapeutic composition comprising the receptorbody in a pharmaceutically acceptable vehicle, and a method of blocking blood vessel growth in a human comprising administering an effective amount of the therapeutic composition.

[0123] The invention also provides for a therapeutic composition comprising a human TIE-2 ligand or ligandbody in a pharmaceutically acceptable vehicle, as well as a method of promoting neovascularization in a patient comprising administering to the patient an effective amount of the therapeutic composition.

[0124] In a specific embodiment of the present invention, methods and compositions are provided for the treatment of ocular neovascularization including proliferative retinopathies characterized by pre-retinal or optic disc neovascularization, such as diabetic retinopathy, venous obstructive disease, ocular ischemic syndrome, retinopathy of prematurity, hemoglobinopathies (e.g. sickle cell anemia), radiation retinopathy, ocular inflammation or infection, ocular neoplasias, tumors and metastases, ocular trauma or injury, or ocular neovascularization occurring idiopathically. In another specific embodiment of the present invention, methods and compositions are provided for the treatment of choroidal neovascularization disorders and diseases characterized by abnormal growth of new vessels, that originate in the choroid, penetrate Bruch's membrane and proliferate beneath the retinal pigment epithelium and/or neural retina, including, age related macular degeneration, angiod streaks, radiation retinopathy, ocular inflammation or infection, ocular neoplasias, tumors and metastases, ocular trauma or injury, or those occurring idiopathically, and other pathological conditions, in a subject in need of such treatment. The subject is typically a mammal, and most preferably a human. Diagnosis of ocular neovascularization is known by those skilled in the art.

[0125] The methods of the invention comprise administering a therapeutically effective amount of pharmaceutical compositions comprising agents capable of interfering with Ang2-mediated biological effects, in an acceptable pharmaceutical carrier to a subject in need, i.e., a subject afflicted with ocular neovascularization. Suitable agents capable of interfering with Ang2-mediated biological effects include, but are not limited to, any agent capable of preventing Ang2 expression, such as anti-sense molecules, or any agent capable of blocking Ang2 interaction with the Tie1 or Tie2 receptor, or any agent capable of interfering with Ang2-mediated biological effects including but not limited to cell signaling. Suitable agents capable of blocking such interaction between Ang2 and Tie 1 or Tie2 include antibodies, including monoclonal antibodies, directed against either Tie 1, Tie2, or Ang2. Skilled artisans will readily recognize that suitable anti-Tie2 or anti-Tie 1 antibodies include blocking antibodies, whereas suitable anti-Ang2 antibodies include neutralizing antibodies. In addition, fragments of antibodies, including scFvs and molecules containing scFvs, are also suitable for use in interfering with Ang2-mediated biological effects. Such scFvs or other antibody fragments may be derived from either anti-Ang2, anti-Tie1, or anti-Tie2 antibodies, and may be used alone or in combination with each other or other agents. Other agents capable of interfering with Ang2 mediated biological effects include receptor bodies that contain the extracellular domain of either Tie1 or Tie2 fused to the constant or Fc domain of an IgG. Specifically, Tie1 or Tie2 receptor bodies are suitable agents for blocking and inhibiting Ang2-mediated biological effects. Other suitable agents are small molecules, lipids, proteins, carbohydrates, or other organic or inorganic molecules capable of interfering with Ang2 mediated biological effects. Such molecules may act to bind to the Tie2 receptor, thus preventing Ang2 binding, or they may prevent or limit or otherwise alter Tie1 or Tie2 receptor signaling subsequent to Ang2 binding. Also suitable are small molecules, lipids, proteins, carbohydrates, or other organic or inorganic molecules that bind to or interact with Ang2 directly, thus preventing, inhibiting or limiting Ang2's ability to bind to the Tie2 receptor.

[0126] The agents may be administered to a subject per se or in the form of a pharmaceutical composition. Pharmaceutical compositions comprising the compounds of the invention may be manufactured by means of conventional mixing, dissolving, granulating, dragee-making, levigating, emulsifying, encapsulating, entrapping or lyophilizing processes. Pharmaceutical compositions may be formulated in conventional manner using one or more physiologically acceptable carriers, diluents, excipients or auxiliaries which facilitate processing of the active agents into preparations which can be used pharmaceutically. Proper formulation is dependent upon the route of administration chosen.

[0127] For topical administration the compounds may be formulated as solutions, gels, ointments, creams, suspensions, etc. as are well-known in the art.

[0128] Systemic formulations include those designed for administration by injection, e.g. subcutaneous, intraocular, intravenous, intramuscular, intrathecal or intraperitoneal injection, as well as those designed for transdermal, transmucosal, oral or pulmonary administration. For injection, the compounds of the invention may be formulated in aqueous solutions, preferably in physiologically compatible buffers such as Hanks's solution, Ringer's solution, or physiological saline buffer. The solution may contain formulatory agents such as suspending, stabilizing and/or dispersing agents. Alternatively, the compounds may be in powder form for constitution with a suitable vehicle, e.g., sterile pyrogen-free water, before use.

[0129] For transmucosal administration, penetrants appropriate to the barrier to be permeated are used in the formulation. Such penetrants are generally known in the art.

[0130] For oral administration, the compounds can be readily formulated by combining the active agents with pharmaceutically acceptable carriers well known in the art. Such carriers enable the compounds of the invention to be formulated as tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, suspensions and the like, for oral ingestion by a patient to be treated. For oral solid formulations such as, for example, powders, capsules and tablets, suitable excipients include fillers such as sugars, such as lactose, sucrose, mannitol and sorbitol; cellulose preparations such as maize starch, wheat starch, rice starch, potato starch, gelatin, gum tragacanth, methyl cellulose, hydroxypropylmethyl-cellulose, sodium carboxymethylcellulose, and/or polyvinylpyrrolidone (PVP); granulating agents; and binding agents. If desired, disintegrating agents may be added, such as the cross-linked polyvinylpyrrolidone, agar, or alginic acid or a salt thereof such as sodium alginate. If desired, solid dosage forms may be sugar-coated or enteric-coated using standard techniques.

[0131] For oral liquid preparations such as, for example, suspensions, elixirs and solutions, suitable carriers, excipients or diluents include water, glycols, oils, alcohols, etc. Additionally, flavoring agents, preservatives, coloring agents and the like may be added.

[0132] For buccal administration, the compounds may take the form of tablets, lozenges, etc. formulated in conventional manner.

[0133] For administration by inhalation, the compounds for use according to the present invention are conveniently delivered in the form of an aerosol spray from pressurized packs or a nebulizer, with the use of a suitable propellant, e.g., dichlorodifluoromethane, trichlorofluoromethane, dichlorotetrafluoroethane, carbon dioxide or other suitable gas. In the case of a pressurized aerosol the dosage unit may be determined by providing a valve to deliver a metered amount. Capsules and cartridges of e.g. gelatin for use in an inhaler or insufflator may be formulated containing a powder mix of the compound and a suitable powder base such as lactose or starch.

[0134] The compounds may also be formulated in rectal or vaginal compositions such as suppositories or retention enemas, e.g, containing conventional suppository bases such as cocoa butter or other glycerides.

[0135] In addition to the formulations described previously, the compounds may also be formulated as a depot preparation. Such long acting formulations may be administered by implantation (for example subcutaneously or intramuscularly) or by intramuscular injection. Thus, for example, the compounds may be formulated with suitable polymeric or hydrophobic materials (for example as an emulsion in an acceptable oil) or ion exchange resins, or as sparingly soluble derivatives, for example, as a sparingly soluble salt.

[0136] Alternatively, other pharmaceutical delivery systems may be employed. Liposomes and emulsions are well known examples of delivery vehicles that may be used to deliver the agents of the invention. Certain organic solvents such as dimethylsulfoxide also may be employed, although usually at the cost of greater toxicity. Additionally, the compounds may be delivered using a sustained-release system, such as semipermeable matrices of solid polymers containing the therapeutic agent. Various of sustained-release materials have been established and are well known by those skilled in the art. Sustained-release capsules may, depending on their chemical nature, release the compounds for a few weeks up to over 100 days. Depending on the chemical nature and the biological stability of the therapeutic reagent, additional strategies for protein stabilization may be employed.

[0137] As the compounds of the invention may contain charged side chains or termini, they may be included in any of the above-described formulations as the free acids or bases or as pharmaceutically acceptable salts. Pharmaceutically acceptable salts are those salts which substantially retain the antimicrobial activity of the free bases and which are prepared by reaction with inorganic acids. Pharmaceutical salts tend to be more soluble in aqueous and other protic solvents than are the corresponding free base forms.

[0138] In preferred embodiments, the agents are prepared as a lyophilized powder that is to be reconstituted with sterile water for injection at appropriate concentrations. The reconstituted solution is buffered at an appropriate pH, typically neutral. In the preferred embodiment, the agents contain no preservative.

[0139] The agents will generally be used in an amount effective to achieve the intended purpose, i.e., treating ocular neovascularization. For use to treat ocular neovascularization, the agents, or pharmaceutical compositions thereof, are administered or applied in a therapeutically effective amount. By therapeutically effective amount is meant an amount effective to ameliorate or prevent the symptoms or conditions of the patient being treated. Determination of a therapeutically effective amount is well within the capabilities of those skilled in the art, especially in light of the detailed disclosure provided herein.

[0140] For systemic administration, a therapeutically effective dose can be estimated initially from in vitro assays. For example, a dose can be formulated in animal models to achieve a circulating concentration range that includes the IC50 as determined in cell culture. Such information can be used to more accurately determine useful doses in humans. Initial dosages can also be estimated from in vivo data, e.g., animal models, using techniques that are well known in the art. One having ordinary skill in the art could readily optimize administration to humans based on animal data.

[0141] Dosage amount and interval may be adjusted individually to provide plasma levels of the compounds which are sufficient to maintain therapeutic effect. In cases of local administration or selective uptake, the effective local concentration of the compounds may not be related to plasma concentration. One having skill in the art will be able to optimize therapeutically effective local dosages without undue experimentation.

[0142] The amount of compound administered will, of course, be dependent on the subject being treated, on the subject's weight, the severity of the affliction, the manner of administration and the judgment of the prescribing physician.

[0143] The therapy may be repeated intermittently while symptoms are detectable or even when they are not detectable. The therapy may be provided alone or in combination with other drugs.

[0144] Preferably, a therapeutically effective dose of the compounds described herein will provide therapeutic benefit without causing substantial toxicity.

[0145] Toxicity of the compounds described herein can be determined by standard pharmaceutical procedures in cell cultures or experimental animals, e.g., by determining the LD50 (the dose lethal to 50% of the population) or the LD100 (the dose lethal to 100% of the population). The dose ratio between toxic and therapeutic effect is the therapeutic index. Compounds which exhibit high therapeutic indices are preferred. The data obtained from these cell culture assays and animal studies can be used in formulating a dosage range that is not toxic for use in human. The dosage of the compounds described herein lies preferably within a range of circulating concentrations that include the effective dose with little or no toxicity. The dosage may vary within this range depending upon the dosage form employed and the route of administration utilized. The exact formulation, route of administration and dosage can be chosen by the individual physician in view of the patient's condition. (See, e.g., Fingl et al., 1975, In: The Pharmacological Basis of Therapeutics, Ch.1, p.1).

[0146] In addition, the present invention provides for a method for identifying a cell which expresses TIE-2 receptor which comprises contacting a cell with a detectably labeled TIE-2 ligand or ligandbody, under conditions permitting binding of the detectably labeled ligand to the TIE-2 receptor and determining whether the detectably labeled ligand is bound to the TIE-2 receptor, thereby identifying the cell as one which expresses TIE-2 receptor. The present invention also provides for a therapeutic composition comprising a TIE-2 ligand or ligandbody and a cytotoxic agent conjugated thereto. The cytotoxic agent may be a radioisotope or toxin.

[0147] The invention also provides a method of detecting expression of TIE-2 ligand by a cell which comprises obtaining mRNA from the cell, contacting the mRNA so obtained with a labeled nucleic acid molecule encoding a TIE-2 ligand, under hybridizing conditions, determining the presence of mRNA hybridized to the labeled molecule, and thereby detecting the expression of the TIE-2 ligand in the cell.

[0148] The invention further provides a method of detecting expression of a TIE-2 ligand in tissue sections which comprises contacting the tissue sections with a labeled nucleic acid molecule encoding a TIE-2 ligand, under hybridizing conditions, determining the presence of mRNA hybridized to the labelled molecule, and thereby detecting the expression of the TIE-2 ligand in tissue sections.

EXAMPLES

Example 1

Identification of the ABAE Cell Line as Reporter Cells for the TIE-2 Receptor

[0149] Adult BAE cells are registered in the European Cell Culture Repository, under ECACC#92010601. (See PNAS 75:2621 (1978)). Northern (RNA) analyses revealed moderate levels of tie-2 transcripts in the ABAE (Adult Bovine Arterial Endothelial) cell line, consistent with in situ hybridization results that demonstrated almost exclusive localization of tie-2 RNAs to vascular endothelial cells. We therefore examined ABAE cell lysates for the presence of TIE-2 protein, as well as the extent to which this TIE-2 protein is tyrosine-phosphorylated under normal versus serum-deprived growth conditions. ABAE cell lysates were harvested and subjected to immunoprecipitation, followed by Western blot analyses of immunoprecipitated proteins with TIE-2 specific and phosphotyrosine-specific antisera. Omission or inclusion of TIE-2 peptides as specific blocking molecules during TIE-2 immunoprecipitation allowed unambiguous identification of TIE-2 as a moderately detectable protein of ˜150 kD whose steady-state phosphotyrosine levels diminish to near undetectable levels by prior serum-starvation of the cells.

[0150] Culture of ABAE cells and harvest of cell lysates was done as follows. Low-passage-number ABAE cells were plated as a monolayer at a density of 2×106 cells/150 mm plastic petri plate (Falcon) and cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% bovine calf serum (10% BCS), 2 mM L-glutamine (Q) and 1% each of penicillin and streptomycin (P-S) in an atmosphere of 5% CO2. Prior to harvest of cell lysates, cells were serum-starved for 24 hours in DMEM/Q/P-S, followed by aspiration of the medium and rinsing of the plates with ice-cold phosphate buffered saline (PBS) supplemented with sodium orthovanadate, sodium fluoride and sodium benzamidine. Cells were lysed in a small volume of this rinse buffer that had been supplemented with 1% NP40 detergent and the protease inhibitors PMSF and aprotinin. Insoluble debris was removed from the cell lysates by centrifugation at 14,000×G for 10 minutes, at 4° C. and the supernatants were subjected to immunoprecipitation with antisera specific for TIE-2 receptor, with or without the presence of blocking peptides added to ˜20 μg/ml lysate. Immunoprecipitated proteins were resolved by PAGE (7.5% Laemmli gel), and then electro-transferred to PVDF membrane and incubated either with various TIE-2- or phosphotyrosine-specific antisera. TIE-2 protein was visualized by incubation of the membrane with HRP-linked secondary antisera followed by treatment with ECL reagent (Amersham).

Example 2

Cloning and Expression of TIE-2 Receptorbody for Affinity-Based Study of TIE-2 Ligand Interactions

[0151] An expression construct was created that would yield a secreted protein consisting of the entire extracellular portion of the rat TIE-2 receptor fused to the human immunoglobulin gamma-1 constant region (IgG1 Fc). This fusion protein is called a TIE-2 “receptorbody” (RB), and would be normally expected to exist as a dimer in solution based on formation of disulfide linkages between individual IgG1 Fc tails. The Fc portion of the TIE-2 RB was prepared as follows. A DNA fragment encoding the Fc portion of human IgG1 that spans from the hinge region to the carboxy-terminus of the protein, was amplified from human placental cDNA by PCR with oligonucleotides corresponding to the published sequence of human IgG1; the resulting DNA fragment was cloned in a plasmid vector. Appropriate DNA restriction fragments from a plasmid encoding the full-length TIE-2 receptor and from the human IgG1 Fc plasmid were ligated on either side of a short PCR-derived fragment that was designed so as to fuse, in-frame, the TIE-2 and human IgG1 Fc protein-coding sequences. Thus, the resulting TIE-2 ectodomain-Fc fusion protein precisely substituted the IgG1 Fc in place of the region spanning the TIE-2 transmembrane and cytoplasmic domains. An alternative method of preparing RBs is described in Goodwin, et. al. Cell 73:447-456 (1993).

[0152] Milligram quantities of TIE-2 RB were obtained by cloning the TIE-2 RB DNA fragment into the pVL1393 baculovirus vector and subsequently infecting the Spodoptera frugiperda SF-21AE insect cell line. Alternatively, the cell line SF-9 (ATCC Accession No. CRL-1711) or the cell line BTI-TN-5b1-4 may be used. DNA encoding the TIE-2 RB was cloned as an Eco RI-NotI fragment into the baculovirus transfer plasmid pVL1393. Plasmid DNA purified by cesium chloride density gradient centrifugation was recombined into viral DNA by mixing 3 μg of plasmid DNA with 0.5 μg of Baculo-Gold DNA (Pharminigen), followed by introduction into liposomes using 30 μg Lipofectin (GIBCO-BRL). DNA-liposome mixtures were added to SF-21AE cells (2×106 cells/60 mm dish) in TMN-FH medium (Modified Grace's Insect Cell Medium (GIBCO-BRL) for 5 hours at 27° C., followed by incubation at 27° C. for 5 days in TMN-FH medium supplemented with 5% fetal calf serum. Tissue culture medium was harvested for plaque purification of recombinant viruses, which was carried out using methods previously described (O'Reilly, D. R., L. K. Miller, and V. A. Luckow, Baculovirus Expression Vectors—A Laboratory Manual. 1992, New York: W. H. Freeman) except that the agarose overlay contained 125 μg/mL X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside; GIBCO-BRL). After 5 days of incubation at 27° C., non-recombinant plaques were scored by positive chromogenic reaction to the X-gal substrate, and their positions marked. Recombinant plaques were then visualized by addition of a second overlay containing 100 μg/mL MTT (3-[4,5-dimethylthiazol-2yl]2,5,diphenyltetrazolium bromide; Sigma). Putative recombinant virus plaques were picked by plug aspiration, and purified by multiple rounds of plaque isolation to assure homogeneity. Virus stocks were generated by serial, low-multiplicity passage of plaque-purified virus. Low passage stocks of one virus clone (vTIE-2 receptorbody) were produced.

[0153] SF-21AE cells were cultured in serum free medium (SF-900 II, Gibco BRL) containing 1×antibiotic/antimycotic solution (Gibco BRL) and 25 mg/L Gentamycin (Gibco BRL). Pluronic F-68 was added as a surfactant to a final concentration of 1 g/L. Cultures (4L) were raised in a bioreactor (Artisan Cell Station System) for at least three days prior to infection. Cells were grown at 27° C., with gassing to 50% dissolved oxygen, at a gas flow rate of 80 mL/min (aeration at a sparge ring). Agitation was by means of a marine impeller at a rate of 100 rpm. Cells were harvested in mid-logarithmic growth phase (˜2×106 cells/mL), concentrated by centrifugation, and infected with 5 plaque forming units of vTIE-2 receptorbody per cell. Cells and inoculum were brought to 400 mL with fresh medium, and virus was adsorbed for 2 hours at 27° C. in a spinner flask. The culture was then resuspended in a final volume of 8L with fresh serum-free medium, and the cells incubated in the bioreactor using the previously described conditions.

[0154] Culture medium from vTIE-2 receptorbody-infected SF21AE cells were collected by centrifugation (500×g, 10 minutes) at 72 hours post-infection. Cell supernatants were brought to pH 8 with NaOH. EDTA was added to a final concentration of 10 mM and the supernatant pH was readjusted to 8. Supernatants were filtered (0.45 μm, Millipore) and loaded on a protein A column (protein A sepharose 4 fast flow or HiTrap protein A, both from Pharmacia). The column was washed with PBS containing 0.5 M NaCl until the absorbance at 280 nm decreased to baseline. The column was washed in PBS and eluted with 0.5 M acetic acid. Column fractions were immediately neutralized by eluting into tubes containing 1 M Tris pH 9. The peak fractions containing the TIE-2 receptorbody were pooled and dialyzed versus PBS.

Example 3

Demonstration That TIE-2 has a Critical Role in Development of the Vasculature

[0155] Insight into the function of TIE-2 was gained by introduction of “excess” soluble TIE-2 receptorbody (TIE-2 RB) into a developing system. The potential ability of TIE-2 RB to bind, and thereby neutralize, available TIE-2 ligand could result in an observable disruption of normal vascular development and characterization of the ligand. To examine whether TIE-2 RB could be used to disrupt vascular development in early chick embryos, small pieces of a biologically resorbable foam were soaked with TIE-2 RB and inserted immediately beneath the chorioallantoic membrane at positions just lateral to the primitive embryo.

[0156] Early chicken embryos develop atop the yolk from a small disk of cells that is covered by the chorioallantoic membrane (CAM). The endothelial cells that will come to line the vasculature in the embryo arise from both extra- and intra-embryonic cell sources. Extra-embryonically-derived endothelial cells, which provide the major source of endothelial cells in the embryo, originate from accretions of mesenchyme that are situated laterally around the embryo-proper, just underneath the CAM. As these mesenchyme cells mature, they give rise to a common progenitor of both the endothelial and hematopoietic cell lineages, termed the hemangioblast. In turn, the hemangioblast gives rise to a mixed population of angioblasts (the endothelial cell progenitor) and hematoblasts (the pluripotential hematopoietic precursor). Formation of rudiments of the circulatory system begins when endothelial cell progeny segregate to form a one-cell-thick vesicle that surrounds the primitive blood cells. Proliferation and migration of these cellular components eventually produces a vast network of blood-filled microvessels under the CAM that will ultimately invade the embryo to join with limited, intra-embryonically-derived vascular elements.

[0157] Newly fertilized chicken eggs obtained from Spafas, Inc. (Boston, Mass.) were incubated at 99.5° F., 55% relative humidity. At about 24 hrs. of development, the egg shell was wiped down with 70% ethanol and a dentist's drill was used to make a 1.5 cm. hole in the blunt apex of each egg. The shell membrane was removed to reveal an air space directly above the embryo. Small rectangular pieces of sterile Gelfoam (Upjohn) were cut with a scalpel and soaked in equal concentrations of either TIE-2 or EHK-1 receptorbody. EHK-1 receptorbody was made as set forth in Example 2 using the EHK-1 extracellular domain instead of the TIE-2 extracellular domain (Maisonpierre et al., Oncogene 8:3277-3288 (1993). Each Gelfoam piece absorbed approximately 6 μg of protein in 30 μl.

[0158] Sterile watchmakers forceps were used to make a small tear in the CAM at a position several millimeters lateral to the primitive embryo. The majority of the piece of RB-soaked Gelfoam was inserted under the CAM and the egg shell was sealed over with a piece of adhesive tape. Other similarly-staged eggs were treated in parallel with RB of the unrelated, neuronally expressed receptor tyrosine kinase, EHK-1 (Maisonpierre et al., Oncogene 8:3277-3288 (1993). Development was allowed to proceed for 4 days and then the embryos were examined by visual inspection. Embryos were removed by carefully breaking the shells in dishes of warmed PBS and carefully cutting away the embryo with surrounding CAM. Of 12 eggs treated with each RB, 6 TIE-2 RB and 5 EHK-1 RB treated embryos had developed beyond the stage observed at the start of the experiment. A dramatic difference was seen between these developed embryos, as shown in FIGS. 1A and 1B. Those treated with EHK-1 RB appeared to have developed relatively normally. Four out of five EHK-1 embryos were viable as judged by the presence of a beating heart. Furthermore, the extra-embryonic vasculature, which is visually obvious due to the presence of red blood cells, was profuse and extended several centimeters laterally under the CAM. By contrast, those treated with TIE-2 RB were severely stunted, ranging from 2-5 mm. in diameter, as compared with more than 10 mm in diameter for the EHK-1 RB embryos. All of the TIE-2 RB treated embryos were dead and their CAMs were devoid of blood vessels. The ability of TIE-2 RB to block vascular development in the chicken demonstrates that TIE-2 ligand is necessary for development of the vasculature.

Example 4

Identification of a TIE-2-Specific Binding Activity in Conditioned Medium From the ras Oncogene-Transformed C2C12 Mouse Myoblast Cell Line

[0159] Screening of ten-fold-concentrated cell-conditioned media (10×CCM) from various cell lines for the presence of soluble, TIE-2-specific binding activity (BIAcore; Pharmacia Biosensor, Piscataway, N.J.) revealed binding activity in serum-free medium from oncogenic-ras-transformed C2C12 cells (C2C12-ras-), RAT 2-ras (which is a ras transformed fibroblast cell line), human glioblastoma T98G and the human neuroblastoma cell line known as SHEP-1.

[0160] The C2C12-ras- 10×CCM originated from a stably transfected line of C2C12 myoblasts that was oncogenically transformed by transfection with the T-24 mutant of H-ras by standard calcium phosphate-based methods. An SV40 based neomycin-resistance expression plasmid was physically linked with the ras expression plasmid in order to permit selection of transfected clones. Resulting G418-resistant ras-C2C12 cells were routinely maintained as a monolayer on plastic dishes in DMEM/glutamine/penicillin/streptomycin supplemented with 10% fetal calf serum (FCS). Serum-free C2C12-ras—10×CCM was made by plating the cells at 60% confluence in a serum free defined media for 12 hours. [Zhan and Goldfarb, Mol. Cell. Biol. 6: 3541-3544 (1986)); Zhan, et al. Oncogene 1: 369-376 (1987)]. The medium was discarded and replaced with fresh DMEM/Q/P-S for 24 hours. This medium was harvested and cells were re-fed fresh DMEM/Q/P-S, which was also harvested after a further 24 hours. These CCM were supplemented with the protease inhibitors PMSF (1 mM) and aprotinin (10 g/ml), and ten-fold concentrated on sterile size-exclusion membranes (Amicon). TIE-2-binding activity could be neutralized by incubation of the medium with an excess of TIE-2 RB, but not by incubation with EHK-1 RB, prior to BIAcore analysis.

[0161] Binding activity of the 10×CCM was measured using biosensor technology (BIAcore; Pharmacia Biosensor, Piscataway, N.J.) which monitors biomolecular interactions in real-time via surface plasmon resonance. Purified TIE-2 RB was covalently coupled through primary amines to the carboxymethyl dextran layer of a CM5 research grade sensor chip (Pharmacia Biosensor; Piscataway, N.J.). The sensor chip surface was activated using a mixture of N-hydroxysuccinimide (NHS) and N-ethyl-N′-(3-dimethylaminopropyl)carbodiimide (EDC), followed by immobilization of TIE-2 RB (25 μg/mL, pH 4.5) and deactivation of unreacted sites with 1.0 M ethanolamine (pH 8.5). A negative control surface of the EHK-1 receptorbody was prepared in a similar manner.

[0162] The running buffer used in the system was HBS (10 mM Hepes, 3.4 mM EDTA, 150 mM NaCl, 0.005% P20 surfactant, pH 7.4). The 10×CCM samples were centrifuged for 15 min at 4° C. and further clarified using a sterile, low protein-binding 0.45 μm filter (Millipore; Bedford, Mass.). Dextran (2 mg/ml) and P20 surfactant (0.005%) were added to each CCM sample. Aliquots of 40 μL were injected across the immobilized surface (either TIE-2 or EHK-1) at a flow rate of 5 μL/min and the receptor binding was monitored for 8 min. The binding activity (resonance units, RU) was measured as the difference between a baseline value determined 30 s prior to the sample injection and a measurement taken at 30 s post-injection. Regeneration of the surface was accomplished with one 12-μL pulse of 3 M MgCl2.

[0163] The instrument noise level is 20 RU; therefore, any binding activity with a signal above 20 RU may be interpreted as a real interaction with the receptor. For C2C12-ras—conditioned media, the binding activities were in the range 60-90 RU for the TIE-2 RB immobilized surface. For the same samples assayed on a EHK-1 RB immobilized surface, the measured activities were less than 35 RU. Specific binding to the TIE-2 receptorbody was evaluated by incubating the samples with an excess of either soluble TIE-2 or EHK-1 RB prior to assaying the binding activity. The addition of soluble EHK-1 RB had no effect on the TIE-2 binding activity of any of the samples, while in the presence of soluble TIE-2 binding to the surface is two-thirds less than that measured in the absence of TIE-2. A repeat assay using >50×concentrated C2C12-ras-CCM resulted in a four-fold enhancement over background of the TIE-2 specific binding signal.

Example 5

C2C12-ras—CCM Contains an Activity That Induces Tyrosine Phosphorylation of TIE-2 Receptor

[0164] C2C12-ras—10×CCM was examined for its ability to induce tyrosine phosphorylation of TIE-2 in ABAE cells. Serum-starved ABAE cells were briefly incubated with C2C12-ras CCM, lysed and subjected to immunoprecipitation and Western analyses as described above. Stimulation of serum-starved ABAE cells with serum-free C2C12-ras—10×CCM was done as follows. The medium of ABAE cells starved as described above was removed and replaced with either defined medium or 10×CCM that had been pre-warmed to 37° C. After 10 minutes, the media were removed and the cells were twice rinsed on ice with an excess of chilled PBS supplemented with orthovanadate/NaF/benzamidine. Cell lysis and TIE-2-specific immunoprecipitation was done as described above.

[0165] ABAE cells incubated for 10 minutes with defined medium showed no induction of TIE-2 tyrosine phosphorylation, whereas incubation with C2C12-ras—CCM stimulated at least a 100×increase in TIE-2 phosphorylation. This activity was almost totally depleted by preincubation of the C2C12-ras—10×CCM for 90 minutes at room temperature with 13 μg of TIE-2 RB coupled to protein G-Sepharose beads. Medium incubated with protein G Sepharose alone was not depleted of this phosphorylating activity.

Example 6

Expression Cloning of TIE-2 Ligand

[0166] COS-7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 1% each of penicillin and streptomycin (P/S) and 2 mM glutamine in an atmosphere of 5% CO2. The mouse myoblast C2C12 ras cell line was cultured in Eagle's minimal essential medium (EMEM) with 10% FBS, (P/S) and 2 mM glutamine. Full length mouse TIE-2 ligand cDNA clones were obtained by screening a C2C12 ras cDNA library in the pJFE14 vector expressed in COS cells. This vector, as shown in FIG. 2, is a modified version of the vector pSRα (Takebe, et al. 1988, Mol. Cell. Biol. 8:466-472). The library was created using the two BSTX1 restriction sites in the pJFE14 vector.

[0167] COS-7 cells were transiently transfected with either the pJFE14 library or control vector by the DEAE-dextran transfection protocol. Briefly, COS-7 cells were plated at a density of 1.0×106 cells/100 mm plate 24 hours prior to transfection. For transfection, the cells were cultured in serum-free DMEM containing 400 μg/ml of DEAE-dextran, 1 μM chloroquine, and 2 mM glutamine, and 1 μg of the appropriate DNA for 3-4 hours at 37° C. in an atmosphere of 5% CO2. The transfection media was aspirated and replaced with PBS with 10% DMSO for 2-3 min. Following this DMSO “shock”, the COS-7 cells were placed into DMEM with 10% FBS, 1% each of penicillin and streptomycin, and 2 mM glutamine for 48 hours.

[0168] Because the TIE-2 ligand is secreted it was necessary to permeabilize the cells to detect binding of the receptorbody probe to the ligand. Two days after transfection the cells were rinsed with PBS and then incubated with PBS containing 1.8% formaldehyde for 15-30 min. at room temperature. Cells were then washed with PBS and incubated for 15 min. with PBS containing 0.1% Triton X-100 and 10% Bovine Calf Serum to permeabilize the cells and block non-specific binding sites.

[0169] The screening was conducted by direct localization of staining using a TIE-2 receptorbody (RB), which consisted of the extracellular domain of TIE-2 fused to the IgG1 constant region. This receptorbody was prepared as set forth in Example 2. A 100 mm dish of transfected, fixed and permeabilized COS cells was probed by incubating them for 30 min with TIE-2 RB. The cells were then washed twice with PBS and incubated for an additional 30 min with PBS/10% Bovine Calf Serum/anti-human IgG-alkaline phosphatase conjugate. After three PBS washes, cells were incubated in alkaline-phosphatase substrate for 30-60 min. The dish was then inspected microscopically for the presence of stained cells. For each stained cell, a small area of cells including the stained cell was scraped from the dish using a plastic pipette tip and plasmid DNA was then rescued and used to electroporate bacterial cells. Single bacterial colonies resulting from the electroporation were picked and plasmid DNA prepared from these colonies was used to transfect COS-7 cells which were probed for TIE-2 ligand expression as evidenced by binding to TIE-2 receptorbodies. This allowed identification of single clones coding for TIE-2 ligand. Confirmation of TIE-2 ligand expression was obtained by phosphorylation of the TIE-2 receptor using the method set forth in Example 5. A plasmid clone encoding the TIE-2 ligand was deposited with the ATCC on Oct. 7, 1994 and designated as “pJFE14 encoding TIE-2 ligand” under ATCC Accession No. 75910.

Example 7

Isolation and Sequencing of Full Length cDNA Clone Encoding Human TIE-2 Ligand

[0170] A human fetal lung cDNA library in lambda gt-10 (see FIG. 3) was obtained from Clontech Laboratories, Inc. (Palo Alto, Calif.). Plaques were plated at a density of 1.25×106/20×20 cm plate, and replica filters taken following standard procedures (Sambrook, et al., Molecular Cloning: A Laboratory Manual, 2nd Ed., page 8.46, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.).

[0171] Isolation of human tie-2 ligand clones was carried out as follows. A 2.2 kb XhoI fragment from the deposited tie-2 ligand clone (ATCC NO. 75910—see Example 6 above) was labeled by random priming to a specific activity of approximately 5×108 cpm/ng. Hybridization was carried out at 65° C. in hybridization solution containing 0.5 mg/ml salmon sperm DNA. The filters were washed at 65° C. in 2×SSC, 0.1% SDS and exposed to Kodak XAR-5 film overnight at −70° C. Positive phage were plaque purified. High titer phage lysates of pure phage were used for isolation of DNA via a Qiagen column using standard techniques (Qiagen, Inc., Chatsworth, Calif., 1995 catalog, page 36). Phage DNA was digested with EcoRI to release the cloned cDNA fragment for subsequent subcloning. A lambda phage vector containing human tie-2 ligand DNA was deposited with the ATCC on Oct. 26, 1994 under the designation λgt10 encoding htie-2 ligand 1 (ATCC Accession No. 75928). Phage DNA may be subjected directly to DNA sequence analysis by the dideoxy chain termination method (Sanger, et al., 1977, Proc. Natl. Acad. Sci. U.S.A. 74: 5463-5467).

[0172] Subcloning of the human tie-2 ligand DNA into a mammalian expression vector may be accomplished as follows. The clone λgt10 encoding htie-2 ligand 1 contains an EcoRI site located 490 base pairs downstream from the start of the coding sequence for the human TIE-2 ligand. The coding region may be excised using unique restriction sites upstream and downstream of the initiator and stop codons respectively. For example, an SpeI site, located 70 bp 5′ to the initiator codon, and a Bpu1102i (also known as Blpl) site, located 265 bp 3′ to the stop codon, may be used to excise the complete coding region. This may then be subcloned into the pJFE14 cloning vector, using the XbaI (compatible to the SpeI overhang) and the PstI sites (the PstI and Bpu1102i sites are both made blunt ended).

[0173] The coding region from the clone λgt10 encoding htie-2 ligand 1 was sequenced using the ABI 373A DNA sequencer and Taq Dideoxy Terminator Cycle Sequencing Kit (Applied Biosystems, Inc., Foster City, Calif.). The nucleotide and deduced amino acid sequence of human TIE-2 ligand from the clone λgt10 encoding htie-2 ligand 1 is shown in FIGS. 4A-4B.

[0174] In addition, full length human tie-2 ligand cDNA clones were obtained by screening a human glioblastoma T98G cDNA library in the pJFE14 vector. Clones encoding human TIE-2 ligand were identified by DNA hybridization using a 2.2 kb XhoI fragment from the deposited tie-2 ligand clone (ATCC NO. 75910) as a probe (see Example 6 above). The coding region was sequenced using the ABI 373A DNA sequencer and Taq Dideoxy Terminator Cycle Sequencing Kit (Applied Biosystems, Inc., Foster City, Calif.). This sequence was nearly identical to that of clone λgt10 encoding htie-2 ligand 1. As shown in FIGS. 4A-4B, the clone λgt10 encoding htie-2 ligand 1 contains an additional glycine residue which is encoded by nucleotides 1114-1116. The coding sequence of the T98G clone does not contain this glycine residue but otherwise is identical to the coding sequence of the clone λgt10 encoding htie-2 ligand 1. FIGS. 5A-5B sets forth the nucleotide and deduced amino acid sequence of human TIE-2 ligand from the T98G clone.

Example 8

Isolation and Sequencing of Second Full Length cDNA Clone a Encoding Human TIE-2 Ligand

[0175] A human fetal lung cDNA library in lambda gt-10 (see FIG. 3) was obtained from Clontech Laboratories, Inc. (Palo Alto, Calif.). Plaques were plated at a density of 1.25×106/20×20 cm plate, and replica filters taken following standard procedures (Sambrook, et al., Molecular Cloning: A Laboratory Manual, 2nd Ed., page 8.46, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). Duplicate filters were screened at low stringency (2×SSC, 55° C.) with probes made to the human TIE-2 ligand 1 sequence. One of the duplicate filters was probed with a 5′ probe, encoding amino acids 25-265 of human TIE-2 ligand 1 as set forth in FIGS. 4A-4B. The second duplicate filter was probed with a 3′ probe, encoding amino acids 282-498 of human TIE-2 ligand 1 sequence (see FIGS. 4A-4B). Both probes were hybridized at 55° C. in hybridization solution containing 0.5 mg/ml salmon sperm DNA. Filters were washed in 2×SSC at 55° C. and exposed overnight to X-ray film. In addition, duplicate filters were also hybridized at normal stringency (2×SSC, 65° C.) to the full length coding probe of mouse TIE-2 ligand 1 (F3-15, XhoI insert). Three positive clones were picked that fulfilled the following criteria: i. hybridization had not been seen to the full length (mouse) probe at normal stringency, and ii. hybridization was seen at low stringency to both 5′ and 3′ probes. EcoRI digestion of phage DNA obtained from these clones indicated two independent clones with insert sizes of approximately 2.2 kb and approximately 1.8 kb. The 2.2 kb EcoRI insert was subcloned into the EcoRI sites of both pBluescript KS (Stratagene) and a mammalian expression vector suitable for use in COS cells. Two orientations were identified for the mammalian expression vector. The 2.2 kb insert in pBluescript KS was deposited with the ATCC on Dec. 9, 1994 and designated as pBluescript KS encoding human TIE 2 ligand 2. The start site of the TIE-2 ligand 2 coding sequence is approximately 355 base pairs downstream of the pBluescript EcoRI site.

[0176] COS-7 cells were transiently transfected with either the expression vector or control vector by the DEAE-dextran transfection protocol. Briefly, COS-7 cells were plated at a density of 1.0×106 cells/100 mm plate 24 hours prior to transfection. For transfection, the cells were cultured in serum-free DMEM containing 400 /g/ml of DEAE-dextran, 1 μM chloroquine, and 2 mM glutamine, and 1 μg of the appropriate DNA for 3-4 hours at 37° C. in an atmosphere of 5% CO2. The transfection media was aspirated and replaced with phosphate-buffered saline with 10% DMSO for 2-3 min. Following this DMSO “shock”, the COS-7 cells were placed into DMEM with 10% FBS, 1% each of penicillin and streptomycin, and 2 mM glutamine for 48 hours.

[0177] Because the TIE-2 ligand is secreted it was necessary to permeabilize the cells to detect binding of the receptorbody probe to the ligand. Transfected COS-7 cells were plated at a density of 1.0×106 cells/100 mm plate. The cells were rinsed with PBS and then incubated with PBS containing 1.8% formaldehyde for 15-30 min. at room temperature. Cells were then washed with PBS and incubated for 15 min. with PBS containing 0.1% Triton X-100 and 10% Bovine Calf Serum to permeabilize the cells and block non-specific binding sites.

[0178] The screening was conducted by direct localization of staining using a TIE-2 receptorbody, which consisted of the extracellular domain of TIE-2 fused to the IgG1 constant region. This receptorbody was prepared as set forth in Example 2. Transfected COS cells were probed by incubating them for 30 min with TIE-2 receptorbody. The cells were then washed twice with PBS, fixed with methanol, and then incubated for an additional 30 min with PBS/10% Bovine Calf Serum/anti-human IgG-alkaline phosphatase conjugate. After three PBS washes, cells were incubated in alkaline-phosphatase substrate for 30-60 min. The dish was then inspected microscopically for the presence of stained cells. Cells expressing one orientation of the clone, but not the other orientation, were seen to bind the TIE-2 receptorbody.

[0179] One of skill in the art will readily see that the described methods may be used to further identify other related members of the TIE ligand family.

[0180] The coding region from the clone pBluescript KS encoding human TIE-2 ligand 2 was sequenced using the ABI 373A DNA sequencer and Taq Dideoxy Terminator Cycle Sequencing Kit (Applied Biosystems, Inc., Foster City, Calif.). The nucleotide and deduced amino acid sequence of human TIE-2 ligand from the clone pBluescript KS encoding human TIE-2 ligand 2 is shown in FIGS. 6A-6B.

Example 9

TIE-2 Ligand 2 is a Receptor Antagonist

[0181] Conditioned media from COS cells expressing either TIE-2 ligand 2 (TL2) or TIE-2 ligand 1 (TL1) were compared for their ability to activate TIE-2 receptors naturally present in human endothelial cell lines.

[0182] Lipofectamine reagent (GIBCO-BRL, Inc.) and recommended protocols were used to transfect COS-7 cells with either the pJFE14 expression vector alone, pJFE14 vector containing the human TIE-2 ligand 1 cDNA, or with a pMT21 expression vector (Kaufman, R. J., 1985, Proc. Natl. Acad. Sci. USA 82: 689-693) containing the human TIE-2 ligand 2 cDNA. COS media containing secreted ligands were harvested after three days and concentrated 20-fold by diafiltration (DIAFLO ultrafiltration membranes, Amicon, Inc.). The quantity of active TIE-2 ligand 1 and TIE-2 ligand 2 present in these media was determined and expressed as the amount (in resonance units, R.U.) of TIE-2 receptor specific binding activity measured by a BIAcore binding assay.

[0183] Northern (RNA) analyses revealed significant levels of TIE-2 transcripts in HAEC (Human Aortic Endothelial Cell) human primary endothelial cells (Clonetics, Inc.). Therefore, these cells were used to examine whether TIE-2 receptor is tyrosine-phosphorylated when exposed to COS media containing the TIE-2 ligands. HAEC cells were maintained in a complete endothelial cell growth medium (Clonetics, Inc.) that contained 5% fetal bovine serum, soluble bovine brain extract, 10 ng/ml human EGF, 1 mg/ml hydrocortisone, 50 mg/ml gentamicin and 50 ng/ml amphotericin-B. Assessment of whether TL1 and TL2 could activate TIE-2 receptor in the HAEC cells was done as follows. Semi-confluent HAEC cells were serum-starved for two hours in high-glucose Dulbecco's MEM with added L-glutamine and penicillin-streptomycin at 37° C. followed by replacement of the starvation medium with ligand-containing conditioned COS media for 7 minutes at 37° C. in a 5% CO2 incubator. The cells were subsequently lysed and TIE-2 receptor protein was recovered by immunoprecipitation of the lysates with TIE-2 peptide antiserum, followed by Western blotting with antiphosphotyrosine antiserum, exactly as described in example 1. The results are shown in FIG. 7. Phosphotyrosine levels on the TIE-2 receptor (TIE-2-R) were induced by treatment of HEAC cells with TIE-2 ligand 1 (Lane L1) but not by TIE-2 ligand 2 (Lane L2) conditioned COS media. MOCK is conditioned media from COS transfected with JFE14 empty vector.

[0184] Evidence that both TL1 and TL2 specifically bind to the TIE-2 receptor was demonstrated by using a BIAcore to assay the TIE-2 receptor specific binding activities in transfected COS media and by immunostaining of TL1 and TL2-expressing COS cells with TIE-2 receptorbodies.

[0185] Because TL2 did not activate the TIE-2 receptor, applicants set out to determine whether TL2 might be capable of serving as an antagonist of TL1 activity. HAEC phosphorylation assays were performed in which cells were first incubated with an “excess” of TL2, followed by addition of dilute TL1. It was reasoned that prior occupancy of TIE-2 receptor due to high levels of TL2 might prevent subsequent stimulation of the receptor following exposure to TL1 present at a limiting concentration.

[0186] Semi-confluent HAEC cells were serum-starved as described above and then incubated for 3 min., at 37° C. with 1-2 ml. of 20×COS/JFE14-TL2 conditioned medium. Control plates were treated with 20×COS/JFE14 only medium (MOCK). The plates were removed from the incubator and various dilutions of COS/JFE14-TL1 medium were then added, followed by further incubation of the plates for 5-7 min. at 37° C. Cells were subsequently rinsed, lysed and TIE-2-specific tyrosine phosphorylation in the lysates was examined by receptor immunoprecipitation and Western blotting, as described above. TL1 dilutions were made using 20×COS/JFE14-TL1 medium diluted to 2×, 0.5×, 0.1×, or 0.02×by addition of 20×COS/JFE14-alone medium. An assay of the initial 20×TL1 and 20×TL2 COS media using BIAcore biosensor technology indicated that they contained similar amounts of TIE-2-specific binding activities, i.e., 445 R.U. and 511 R.U. for TL1 and TL2, respectively. The results of the antiphosphotyrosine Western blot, shown in FIG. 8, indicate that when compared to prior treatment of HAEC cells with MOCK medium (lane 1), prior treatment of HAEC cells with excess TIE-2 ligand 2 (lane 2) antagonizes the subsequent ability of dilute TIE-2 ligand 1 to activate the TIE-2 receptor (TIE-2-R).

[0187] The ability of TL2 to competitively inhibit TL1 activation of the TIE-2-R was further demonstrated using the human cell hybrid line, EA.hy926 (see Example 21 for detailed description of this cell line and its maintenance). Experiments were performed in which unconcentrated COS cell media containing TL1 were mixed at varying dilutions with either MOCK- or TL2conditioned media and placed on serum-starved EA.hy926 cell monolayers for 5 minutes at 37° C. The media were then removed, the cells were harvested by lysis and TIE-2-specific tyrosine phosphorylation was examined by Western blots, as described above. FIG. 9 shows an experiment which contains three groups of treatments, as viewed from left to right. As shown in the four lanes at the left, treatment of the EA.hy926 cells with 1×COS-TL1 alone robustly activated the endogenous TIE-2-R in these cells, whereas 1×TL2 COS medium was inactive.

[0188] However, mixture of TL1 with either MOCK or TL2 demonstrated that TL2 can block the activity of TL1 in a dose-dependent fashion. In the central three pairs of lanes the ratio of TL2 (or MOCK) was decreased while the amount of TL1 in the mixture was correspondingly increased from 0.1×to 0.3×. At any of these mixture ratios the TL1:TL2 lanes showed a reduced level of TIE-2-R phosphorylation compared to that of the corresponding TL1:MOCK lanes. When the amount TL1 was held steady and the amount of TL2 (or MOCK) was decreased, however (shown in the three pairs of lanes at the right), a point was reached at which the TL2 in the sample was too dilute to effectively inhibit TL1 activity. The relative amount of each ligand present in these conditioned COS media could be estimated from their binding units as measured by the BIAcore assay and from Western blots of the COS media with ligand-specific antibodies. Consequently, we can infer that only a few-fold molar excess of TL2 is required to effectively block the activity of TL1 in vitro. This is significant because we have observed distinct examples in vivo (see Example 17 and FIG. 16) where TL2 mRNAs achieve considerable abundance relative to those of TL1. Thus, TL2 may be serving an important physiological role in effectively blocking signaling by the TIE-2-R at these sites.

[0189] Taken together these data confirm that, unlike TL1, TL2 is unable to stimulate endogenously expressed TIE-2-R on endothelial cells. Furthermore, at a few fold molar excess TL2 can block TL1 stimulation of the TIE-2 receptor, indicating that TL2 is a naturally occurring TIE-2 receptor antagonist.

Example 10

Identification of TIE-2-Specific Binding Activity in Conditioned Medium and COS Cell Supernatants

[0190] Binding activity of 10×CCM from the cell lines C2C12-ras, Rat2 ras, SHEP, and T98G, or COS cell supernatants after transfection with either human TIE-2 ligand 1 (hTL1) or human TIE-2 ligand 2 (hTL2) was measured using biosensor technology (BIAcore; Pharmacia Biosensor, Piscataway, N.J.) which monitors biomolecular interactions in real-time via surface plasmon resonance (SPR). Purified rat or human TIE-2 RB was covalently coupled through primary amines to the carboxymethyl dextran layer of a CM5 research grade sensor chip (Pharmacia Biosensor; Piscataway, N.J.). The sensor chip surface was activated using a mixture of N-hydroxysuccinimide (NHS) and N-ethyl-N′-(3-dimethylaminopropyl)carbodiimide (EDC), followed by immobilization of TIE-2 RB (25 μg/mL, pH 4.5) and deactivation of unreacted sites with 1.0 M ethanolamine (pH 8.5). In general, 9000-10000 RU of each receptorbody was coupled to the sensor chip.

[0191] The running buffer used in the system was HBS (10 mM Hepes, 150 mM NaCl, 0.005% P20 surfactant, pH 7.4). The samples were centrifuged for 15 min at 4° C. and further clarified using a sterile, low protein-binding 0.45 μm filter (Millipore; Bedford, Mass.). Dextran (2 mg/ml) and P20 surfactant (0.005%) were added to each sample. Aliquots of 40 μL were injected across the immobilized surface (either rat or human TIE-2) at a flow rate of 5 μL/min and the receptor binding was monitored for 8 min. The binding activity (resonance units, RU) was measured as the difference between a baseline value determined 30 s prior to the sample injection and a measurement taken at 30 s post-injection. Regeneration of the surface was accomplished with one 15-μL pulse of 3 M MgCl2.

[0192] The CCM samples (C2C12-ras, Rat2-ras, SHEP, T98G) were tested on the rat TIE-2 RB immobilized surface, while the recombinant hTL1 and hTL2 were tested on the human TIE-2 RB immobilized surface. In each case, specific binding to the TIE-2 receptorbody was evaluated by incubating the samples with 25 μg/ml of either soluble TIE-2 (rat or human) RB or trkB RB prior to assaying the binding activity. As shown in FIGS. 10A-10D and FIG. 11, the addition of soluble trkB RB causes a slight decrease in the TIE-2 binding activity, while the addition of soluble TIE-2 RB significantly reduces the binding activity as compared to that measured in the absence of TIE-2 RB.

Example 11

TIE-2 RB Specifically Blocks Activation of the TIE-2 Receptor by TIE-2 Ligand 1

[0193] The applicants sought to determine whether soluble TIE-2 RB can serve as a competitive inhibitor to block activation of TIE-2 receptor by TIE-2 ligand 1 (TL1). To do this, TL1-containing COS media were preincubated with either TIE-2- or TrkB-RB and then compared for their ability to activate TIE-2 receptors naturally present in a human endothelial cell line.

[0194] Conditioned COS media were generated from COS-7 cells transfected with either the pJFE14 expression vector alone (MOCK), or pJFE14 vector containing the human TIE-2 ligand 1 cDNA (TL1) and harvested as described in Example 9 hereinabove, with the exception that the media were sterile filtered but not concentrated. The quantity of TL1 was determined and expressed as the amount (in resonance units, R.U.) of TIE-2 receptor-specific binding activity measured by BIAcore binding assay.

[0195] Northern (RNA) analyses revealed significant levels of tie-2 transcripts in HUVEC (Human Umbilical Vein Endothelial Cell) human primary endothelial cells (Clonetics, Inc.). Therefore, these cells were used to examine whether TIE-2 receptor can be tyrosine-phosphorylated when exposed in the presence of TIE-2- or TrkB-RBs to COS media containing TL1. HUVEC cells were maintained at 37° C., 5% CO2 in a complete endothelial cell growth medium (Clonetics, Inc.) that contained 5% fetal bovine serum, soluble bovine brain extract with 10 μg/ml heparin, 10 ng/ml human EGF, 1 ug/ml hydrocortisone, 50 μg/ml gentamicin and 50 ng/ml amphotericin-B. Assessment of whether TL1 could activate TIE-2 receptor in the HUVEC cells was done as follows. Confluent dishes of HUVEC cells were serum-starved for two-to-four hours in low-glucose Dulbecco's MEM at 37° C., 5% CO2, followed by 10 minute incubation in starvation medium that included 0.1 mM sodium orthovanadate, a potent inhibitor of phosphotyrosine phosphatases. Meanwhile, conditioned COS media were preincubated 30 min. at room temperature with either TIE-2- or TrkB-RB added to 50 μg/ml. The starvation medium was then removed from the HUVEC dishes and incubated with the RB-containing COS media for 7 minutes at 37° C. HUVEC cells were subsequently lysed and TIE-2 receptor protein was recovered by immunoprecipitation with TIE-2 peptide antiserum, followed by Western blotting with an anti-phosphotyrosine antibody, as described in Example 1. The results are shown in FIG. 12. Phosphotyrosine levels on the TIE-2 receptor were induced by treatment of HUVEC cells with TIE-2 ligand 1 (TL1) relative to that seen with control medium (MOCK) and this induction is specifically blocked by prior incubation with TIE-2-RB (TIE-2Fc) but not by incubation with TrkB-RB (TrkB-Fc). These data indicate that soluble TIE-2 RB can serve as a selective inhibitor to block activation of TIE-2 receptor by TIE-2 ligand 1.

Example 12

Construction of TIE-2 Ligandbodies

[0196] An expression construct was created that would yield a secreted protein consisting of the entire coding sequence of human TIE-2 ligand 1 (TL1) or TIE-2 ligand 2 (TL2) fused to the human immunoglobulin gamma-1 constant region (IgG1 Fc). These fusion proteins are called TIE-2 “ligandbodies” (TL1-Fc or TL2-Fc). The Fc portion of TL1-Fc and TL2-Fc was prepared as follows. A DNA fragment encoding the Fc portion of human IgG1 that spans from the hinge region to the carboxy-terminus of the protein, was amplified from human placental cDNA by PCR with oligonucleotides corresponding to the published sequence of human IgG1; the resulting DNA fragment was cloned in a plasmid vector. Appropriate DNA restriction fragments from a plasmid encoding full-length TL1 or TL2 and from the human IgG1 Fc plasmid were ligated on either side of a short PCR-derived fragment that was designed so as to fuse, in-frame, TL1 or TL2 with human IgG1 Fc protein-coding sequences.

[0197] Milligram quantities of TL2-Fc were obtained by cloning the TL2-Fc DNA fragment into the pVL1393 baculovirus vector and subsequently infecting the Spodoptera frugiperda SF-21AE insect cell line. Alternatively, the cell line SF-9 (ATCC Accession No. CRL-1711) or the cell line BTI-TN-5b1-4 may be used. DNA encoding the TL2-Fc was cloned as an Eco RI-NotI fragment into the baculovirus transfer plasmid pVL1393. Plasmid DNA was recombined into viral DNA by mixing 3 μg of plasmid DNA with 0.5 μg of Baculo-Gold DNA (Pharminigen), followed by introduction into liposomes using 30 μg Lipofectin (GIBCO-BRL). DNA-liposome mixtures were added to SF-21AE cells (2×106 cells/60 mm dish) in TMN-FH medium (Modified Grace's Insect Cell Medium (GIBCO-BRL) for 5 hours at 27° C., followed by incubation at 27° C. for 5 days in TMN-FH medium supplemented with 5% fetal calf serum. Tissue culture medium was harvested for plaque purification of recombinant viruses, which was carried out using methods previously described (O'Reilly, D. R., L. K. Miller, and V. A. Luckow, Baculovirus Expression Vectors—A Laboratory Manual. 1992, New York: W. H. Freeman) except that the agarose overlay contained 125 mg/mL X-gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside; GIBCO-BRL). After 5 days of incubation at 27° C., non-recombinant plaques were scored by positive chromogenic reaction to the X-gal substrate, and their positions marked. Recombinant plaques were then visualized by addition of a second overlay containing 100 mg/mL MTT (3-[4,5-dimethylthiazol-2yl]2,5,diphenyltetrazolium bromide; Sigma). Putative recombinant virus plaques were picked by plug aspiration, and purified by multiple rounds of plaque isolation to assure homogeneity. Virus stocks were generated by serial, low-multiplicity passage of plaque-purified virus. Low passage stocks of one virus clone (vTL2-Fc Clone #7) were produced.

[0198] SF-21AE cells were cultured in serum-free medium (SF-900 II, Gibco BRL) containing 1×antibiotic/antimycotic solution (Gibco BRL) and 25 mg/L Gentamycin (Gibco BRL). Pluronic F-68 was added as a surfactant to a final concentration of 1 g/L. Cultures (4L) were raised in a bioreactor (Artisan Cell Station System) for at least three days prior to infection. Cells were grown at 27° C., with gassing to 50% dissolved oxygen, at a gas flow rate of 80 mL/min (aeration at a sparge ring). Agitation was by means of a marine impeller at a rate of 100 rpm. Cells were harvested in mid-logarithmic growth phase (˜2×10 6 cells/mL), concentrated by centrifugation, and infected with 5 plaque forming units of vTL2-Fc per cell. Cells and inoculum were brought to 400 mL with fresh medium, and virus was adsorbed for 2 hours at 27° C. in a spinner flask. The culture was then resuspended in a final volume of 8L with fresh serum-free medium, and the cells incubated in the bioreactor using the previously described conditions.

[0199] Culture medium from vTL2-Fc-infected SF21AE cells were collected by centrifugation (500×g, 10 minutes) at 72 hours post-infection. Cell supernatants were brought to pH 8 with NaOH. EDTA was added to a final concentration of 10 mM and the supernatant pH was readjusted to 8. Supernatants were filtered (0.45 μm, Millipore) and loaded on a protein A column (protein A sepharose 4 fast flow or HiTrap protein A, both from Pharmacia). The column was washed with PBS containing 0.5 M NaCl until the absorbance at 280 nm decreased to baseline. The column was washed in PBS and eluted with 0.5 M acetic acid. Column fractions were immediately neutralized by eluting into tubes containing 1 M Tris pH 9. The peak fractions containing the TL2-Fc were pooled and dialyzed versus PBS.

Example 13

Expression of TIE-1, TIE-2, TL1, and TL2 in Renal Cell Carcinoma

[0200] In situ hybridization experiments were performed on human renal cell carcinoma tumor tissue using TIE-1, TIE-2, TL1, and TL2 cDNA probes. TIE-2, TIE-1, TL1, and TL2 expression were all up-regulated in the tumor vasculature. Ligand expression appeared to be localized to either the vascular endothelial cells (TL2) or very near the vascular endothelial cells in the mesenchyme (TL1). VEGF has been shown to be dramatically up-regulated in this tumor tissue. Brown, et al. Am. J. Pathol. 143:1255-1262 (1993).

Example 14

Expression of TIE-1, TIE-2, TL1, and TL2 in wound Healing

[0201] In situ hybridization experiments were performed on cross-sectional tissue slices obtained from a rat cutaneous wound model using TIE-1, TIE-2, TL1, and TL2 cDNA probes. The wound healing model involves pressing a small cork bore against the skin of a rat and removing a small, cylindrical plug of skin. As healing begins at the base of the wound, a vertical slice of tissue is taken and used for in situ hybridization. In the tested tissue sample, TL1 and TL2 appeared to be slightly up-regulated by four days post-injury. In contrast to the slightly up-regulated expression of TL1 and TL2 in this tissue, VEGF expression, which may precede TL1 and TL2 expression, is dramatically up-regulated.

Example 15

Expression of TIE Ligands in Fetal Liver and Thymus

[0202] Reverse transcription-PCR (RT-PCR) was performed on mouse E14.5 fetal liver and mouse E17.5 fetal thymus. Agarose gel electrophoresis of the RT-PCR products revealed that in the mouse fetal liver, TIE-2 ligand 1 (TL1) RNA is enriched in the stromal region, but is absent in c-kit+TER119 hematopoietic precursor cells. In this same tissue, TIE-2 ligand 2 (TL2) RNA is enriched in the stromal cells, but absent in the hematopoietic precursor cells (FIG. 13). In the mouse fetal thymus, TL2 is enriched in the stromal cells (FIG. 14).

Example 16

The TIE Receptor/Ligand System in Angiogenesis

[0203] Although the TIE-2/TIE ligand system appears to play an important role in endothelial cell biology, it has not been shown to play a significant, active role in the early to intermediate stages of vascularization (e.g. angioblast or endothelial cell proliferation and migration, tubule formation, and other early stage events in vascular modeling). In contrast to the receptors and factors known to mediate these aspects of vascular development, the temporally late pattern of expression of TIE-2 and TL1 in the course of vascularization suggests that this system plays a distinct role in the latter stages vascular development, including the structural and functional differentiation and stabilization of new blood vessels. The pattern of expression of TIE-2/TL1 also is consistent with a continuing role in the maintenance of the structural integrity and/or physiological characteristics of an established vasculature.

[0204] TIE Ligand 2 (TL2) appears to be a competitive inhibitor of TL1. The spatiotemporal characteristics of TL2 expression suggest that this single inhibitory molecule may play multiple, context-dependent roles essential to appropriate vascular development or remodeling (e.g. de-stabilization/de-differentiation of mature endothelial cells allowing the formation of new vessels from existing vasculature, inhibition of inappropriate blood vessel formation, and regression/involution of mature blood vessels). FIG. 15 is a schematic representation of the hypothesized role of the TIE-2/TIE ligands in angiogenesis. In this figure TL1 is represented by (•), TL2 is represented by (*), TIE-2 is represented by (T), VEGF is represented by ([ ]), and flk-1 (a VEGF receptor) is represented by (Y).

Example 17

Expression of TIE Ligands in the Female Reproductive System: Expression in the Ovary

[0205] Preliminary observations made in experiments examining the expression of the TIE receptors and ligands in the female reproductive system are consistent with the hypothesis the TL1 plays a role in neovascularization which temporally follows that of VEGF. The pattern of TL2 expression is also consistent with an antagonism of the action of TL1, and a specific role in vascular regression. To verify this, expression of relevant mRNAs can be examined following experimental induction of follicular and luteal development so that their temporal relation to various aspects of neovascularization/vascular regression can be more clearly defined (e.g. in conjunction with endothelial cell staining, vascular fills). Angiogenesis associated with follicular development and corpus luteum formation in staged ovaries of mature, female rats or following induced ovulation in pre-pubertal animals was followed using in situ hybridization. FIG. 16 contains photographs of in situ hybridization slides showing the temporal expression pattern of TIE-2, TL1, TL2, and VEGF during the ovarian cycle [Column 1: Early pre-ovulatory follicle; Column 2: pre-ovulatory follicle; Column 3: early corpus luteum; and Column 4: atretic follicle; Row A: bright field; Row B: VEGF; Row C: TL2; Row D: TL1 and Row E: TIE-2 receptor]. These studies revealed that VEGF, TL1 and TL2 are expressed in a temporally and spatially coordinate fashion with respect to the development and regression of vasculature in the ovary, -specifically with respect to the establishment of the vascular system which is generated in the course of the conversion of an ovarian follicle to a corpus luteum (CL).

[0206] Briefly, VEGF expression increases in the follicular granule layer prior to its vascularization during the process of luteinization. During the process of CL formation, highest levels of VEGF expression are apparent in the center of the developing CL in the vicinity of luteinizing cells which are not yet vascularized. VEGF levels remain moderately high and are diffusely distributed in the developed CL. In contrast, noticeably enhanced expression of TIE-2 ligand 1 occurs only late in process of CL formation, after a primary vascular plexus has been established. Later, TL1 expression is apparent throughout the CL at which time the definitive capillary network of the CL has been established.

[0207] TL2 exhibits a more complex pattern of expression than either VEGF or TL1. In the developing CL, TL2 is expressed at highest levels at the front of the developing capillary plexus—between the central avascular region of the CL where VEGF expression is highest, and the most peripheral portion of the CL where TL1 expression is dominant and where the luteinization process is complete and the vascular system is most mature. TL2 also appears to be expressed at high levels in the follicular layer of large follicles which are undergoing atresia. While TL1 is also apparent in atretic follicles, VEGF is not expressed.

[0208] The pattern of expression described above is most consistent with a role for VEGF in the initiation of angiogenesis, with TL1 acting late in this process-for example in modeling and/or stabilization of the definitive vascular network. In contrast, TL2 is present both in areas of active expansion of a newly forming vascular network (during CL formation), and in regions which fail to establish a new vasculature and vascular regression is in progress (atretic follicles). This suggests a more dynamic and complex role for TL2, possibly involving destabilization of existing vasculature (necessary for regression) or developing vasculature (necessary for the dynamic modeling of newly forming vessels).

Example 18

Construction and Characterization of the CYS-TL1 Mutant

[0209] The TIE-2 ligands have two major structural domains, one described as a “coiled-coil” domain comprising the approximate C-terminal third of the protein and the other a “fibrinogen-like” domain comprising the approximate N-terminal two-thirds of the protein. Although the TIE-2 ligands, designated TL1 and TL2, share similar structural homology, they exhibit different physical and biological properties. Under non-reducing electrophoretic conditions, both proteins exhibit covalent, multimeric structures, with TL1 existing primarily as a trimer and TL2 existing primarily as a dimer. FIG. 17 is a schematic representation of how the TIE-2 ligands may be interacting to form multimers. In terms of biological activity, TL1 has been shown to be an agonist of the TIE-2 receptor, as demonstrated by induction of phosphorylation in TIE-2 expressing cells. TL2, on the other hand, appears to be a competitive inhibitor of TL1. Investigations into what factors might be contributing to the different physical and biological properties of the two molecules revealed the presence of a cysteine residue (CYS265) preceding the fibrinogen-like domain in TL1 but absent in TL2. This CYS265 residue in TL1 is encoded by TGC and is located at about nucleotides 1102-1104 at the approximate junction between the coiled-coil and fibrinogen-like domains. Because cysteine residues are generally involved in disulfide bond formation, the presence of which can contribute to both the tertiary structure and biological properties of a molecule, it was thought the perhaps the presence of the CYS265 in TL1 might be at least partially responsible for the different properties of the two molecules. To test this hypothesis, an expression plasmid was constructed which contained a mutation in TL1 in which the CYS was replaced with an amino acid which does not form disulfide bonds. In addition to this TL1/CYS− mutant, a second expression plasmid was constructed which mutated the corresponding position in TL2 so that this residue was now a cysteine. Both non-mutated and mutated expression plasmids of TL1 and TL2 were transiently transfected into COS cells. Cell supernatants containing the recombinant proteins were harvested and samples subjected to both reducing and non-reducing SDS/PAGE electrophoresis and subsequent western blotting. FIG. 18 contains western blots of both non-mutated and mutated TL1 and TL2 proteins under revealing that the TL1/CYS− mutant behaves more TL2-like in that it runs as a dimer and that the TL2/CYS+ mutant behaves more TL1-like in that it is able to form a trimer as well as higher-order multimers. Interestingly, when the two mutant proteins were tested for their ability to induce phosphorylation in TIE-2 expressing cells, the TL1/CYS− mutant was able to activate the TIE-2 receptor, whereas the TL2/CYS+ mutant did not gain any activating activity.

Example 19

Construction and Characterization of Fibrinogen-Like Domain Only Mutants

[0210] In order to test whether the fibrinogen-like domain (F-domain) of the TIE-2 ligands contained TIE-2 activating activity, expression plasmids were constructed which deleted the coiled-coil domain, leaving only that portion of the DNA sequence encoding the F-domain (beginning at about nucleotide 1159, amino acid residue ARG284). This mutant construct was transiently transfected into COS cells. The supernatant containing the recombinant protein was harvested. The TL1/F-domain mutant was tested for it's ability to bind the TIE-2 receptor. The results showed that, as a monomer, the TL1/F-domain mutant was not able to bind TIE-2 at a detectable level. However, when the TL1/F-domain monomer was myc-tagged and subsequently clustered with an antibody directed against the myc tag, it did exhibit detectable binding to TIE-2. However, the antibody-clustered TL1/F-domain mutant was not able to induce phosphorylation in a TIE-2 expressing cell line. FIG. 17 shows a schematic representation of the F-domain construct and its binding ability plus and minus antibody clustering.

Example 20

Receptorbody Binding Assay and a Ligand Binding and Competition Assay

[0211] A quantitative cell-free binding assay with two alternate formats has been developed for detecting either TIE-2 receptorbody binding or ligand binding and competition. In the receptorbody binding version of the assay, TIE-2 ligands (purified or partially purified; either TL1 or TL2) are coated onto an ELISA plate. Receptorbody at varying concentrations is then added, which binds to the immobilized ligand in a dose-dependent manner. At the end of 2 hours, excess receptorbody is washed away, then the amount bound to the plate is reported using a specific anti-human Fc antibody which is alkaline phosphatase tagged. Excess reporter antibody is washed away, then the AP reaction is developed using a colored substrate. The assay is quantitated using a spectrophotometer. FIG. 19 shows a typical TIE-2-IgG binding curve. This assay has been used to evaluate the integrity of TIE-2-IgG after injection into rats and mice. The assay can also be used in this format as a ligand competition assay, in which purified or partially-purified TIE ligands compete with immobilized ligand for receptorbody. In the ligand binding and competition version of the binding assay, TIE-2 ectodomain is coated onto the ELISA plate. The Fc-tagged fibrinogen-like domain fragments of the TIE ligands (TL1-fFc and TL2-fFc) then bind to the ectodomain, and can be detected using the same anti-human Fc antibody as described above. FIG. 20 shows an example of TL1-fFc binding to TIE-2 ectodomain. This version of the assay can also be used to quantitate levels of TL1-fFc in serum or other samples. If untagged ligand (again, either purified or unpurified) is added at the same time as the TL1-fFc, then a competition is set up between tagged ligand fragment and full-length ligand. The full-length ligand can displace the Fc-tagged fragment, and a competition curve is generated.

Example 21

EA.hy926 Cell Line can be Used as a Reporter Cell Line for TIE Ligand Activity

[0212] EA.hy926 is a cell hybrid line that was established by fusion of HUVEC with the human lung carcinoma-derived line, A549 [Edgell, et al. Proc. Natl. Acad. Sci. (USA) 80, 3734-3737 (1983). EA.hy926 cells have been found to express significant levels of TIE-2 receptor protein with low basal phosphotyrosine levels. The density at which EA.hy926 cells are passaged prior to their use for receptor assays, as well as their degree of confluency at the time of assay, can affect TIE-2 receptor abundance and relative inducibility in response to treatment with ligand. By adopting the following regimen for growing these cells the EA.hy926 cell line can be used as a dependable system for assay of TIE-2 ligand activities.

[0213] EA.hy926 cells are seeded at 1.5×106 cells in T-75 flasks (Falconware) and re-fed every other day with high-glucose Dulbecco's MEM, 10% fetal bovine serum, L-glutamine, penicillin-streptomycin, and 1×hypoxanthine-aminopterin-thymidine (HAT, Gibco/BRL). After three to four days of growth, the cells are passaged once again at 1.5×106 cells per T-75 flask and cultured an additional three to four days. For phosphorylation assays, cells prepared as described above were serum-starved by replacement of the culture medium with high-glucose DMEM and incubation for 2-3 hours at 37° C. This medium was aspirated from the flask and samples of conditioned media or purified ligand were added to the flask in a total volume of 1.5 ml followed by incubation at 37° C. for 5 minutes. Flasks were removed from the incubator and placed on a bed of ice. The medium was removed and replaced with 1.25 ml Lysis Buffer containing 1% nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS in 20 mM Tris, pH 7.6, 150 mM NaCl, 50 mM NaF, 1 mM sodium orthovanadate, 5 mM benzamidine, and 1 mM EDTA containing the protease inhibitors PMSF, aprotinin, and leupeptin. After 10 minutes on ice to allow membrane solubilization, plates were scraped and cell lysates were clarified by microcentrifugation at top speed for 10 minutes at 4° C. TIE-2 receptor was immunoprecipitated from the clarified supernatant by incubation in the cold with an anti-TIE-2 polyclonal antiserum and Protein G-conjugated Sepharose beads. The beads were washed three times with cold cell lysis buffer and boiled 5 minutes in Laemmli sample buffer, which was then loaded on 7.5% SDS-polyacrylamide gels. Resolved proteins were electrotransferred to PVDF (Lamblia-P) membrane and then subjected to Western blot analysis using anti-phosphotyrosine antibody and the ECL reagent. Subsequent comparison of total TIE-2 protein levels on the same blots was done by stripping the anti-phosphotyrosine antibody and reincubating with a polyclonal antiserum specific to the ectodomain of TIE-2.

Example 22

ANG2 Plays an Important Role in Retinal Angiogenesis

[0214] Introduction

[0215] Several studies suggest that embryonic vascular development is a coordinated collaboration between vascular endothelial cell growth factor (VEGF), VEGF receptor-1 (VEGFR1), VEGFR2, the Tie receptors, and their binding partners, Angiopoietin-1 (Ang1) and Ang2 (Carmeliet et al., 1996; Davis et al., 1996; Dumont et al., 1994; Ferrara et al., 1996; Fong et al., 1995; Puri et al., 1995; Sato et al., 1995; Shalaby et al., 1995; Suri et al., 1996). A model has been proposed to explain observations made in mice with targeted disruption of genes for each of the collaborators (Breier et al., 1997; Hanahan, 1997; Maisonpierre et al., 1997). Acting through VEGFR2 (Flk1, KDR), VEGF stimulates differentiation of angioblasts into endothelial cells, and by activating VEGFR1 (Flt1), VEGF promotes assembly of endothelial cells into tubes. Ang1-induced activation of Tie2 results in remodeling, differentiation and stabilization of newly formed blood vessels, in part by promoting the appropriate association of endothelial cells with matrix and periendothelial support cells.

[0216] In addition to their effects on the embryonic vasculature, there is mounting evidence suggesting that Ang1 and Ang2 collaborate with VEGF during neovascularization at later stages of development and in adults. A survey of Ang 2 expression in adult tissues showed undetectable Ang2 mRNA levels in most tissues, but substantial levels in ovary, uterus, and placenta, three tissues in which physiologic angiogenesis occurs (Maisonpierre et al., 1997). There is a recurring cycle of vascular quiescence, followed by angiogenesis, and then vascular regression in corpus luteum and Ang2 levels were found to be low during quiescence and high during angiogenesis and regression, suggesting that Ang2 may play a role in both positive and negative vascular remodeling (Maisonpierre et al., 1997). Ang2 mRNA is also expressed at high levels in blood vessels associated with experimental and human glioblastoma (Holash et al., 1999; Zagzag et al., 1999; Zagzag et al., 2000). Interestingly, induction of Ang2 expression occurs rapidly following association of tumor cells with extant vessels. In the absence of appreciable VEGF expression, many of these blood vessels regress. As VEGF expression is increased as the tumor grows, angiogenesis is initiated in vessels that continue to express Ang2. The correlation between the initiation of angiogenesis and the coincident expression of Ang2 and VEGF has also been noted in other types of tumors, and in other conditions characterized by pathological neovascularization. These observations suggest that Ang2 participates with VEGF in vascular remodeling in situations other than embryonic vascular development.

[0217] Materials and Methods

[0218] Mice

[0219] Mice with a promoterless β-galactosidase gene replacing the translation initiation site and the signal peptide encoding sequences of the ang2 gene, resulting in disruption of the ang2 gene and expression of β-galactosidase under control of the ang2 promoter were generated using standard techniques familiar to the skilled artisan.

[0220] Murine Model of Oxygen-Induced Ischemic Retinopathy

[0221] Ischemic retinopathy was produced as described by Smith, L. E. H., et al., (1994) Invest. Ophthalmol. Vis. Sci., 35:101-111. Litters containing seven-day-old ang2+/− (heterozygotes) and ang2−/− (homozygous) mice were placed in an airtight incubator and exposed to an atmosphere of 75±3% oxygen for 5 days. Incubator temperature was maintained at 23±2° C., and oxygen was measured every 8 hours with an oxygen analyzer. After 5 days, the mice were removed from the incubator and sacrificed after various amounts of time in room air. Controls comprised ang2+/− and ang2−/− mice of the same age raised in room air.

[0222] Results ps Effects of Ang2 Gene Deletion on the Development of Retinal Vasculature.

[0223] At P10 and P18, ang2−/− mice and littermate control ang2+/− mice were perfused with fluorescein-labeled dextran and retinal flat mounts were prepared. In control mice and ang2+/− mice, by P10, a dense vascular bed extended almost to the peripheral edge of the retina and there was no remaining hyaloid vasculature (FIG. 21A and FIG. 21B). In contrast, in P10 ang2−/− mice, a substantial part of the peripheral retina was still avascular (FIG. 21C, arrows) and in the posterior retina, the retinal capillaries were less dense than those in ang+/− mice (FIG. 21D vs. FIG. 21B). Also, the hyaloid vasculature was still present (FIG. 21D, arrows).

[0224] In both heterozygous and homozygous mice, retinal cross sections were double-labeled with biotinylated Griffonia simplicifolia lectin (GSA), which selectively stains vascular cells, and for LacZ, which identifies sites where the ang2 promoter is turned on (FIG. 22A-FIG. 22P). At P7, there were some remaining hyaloid vessels in ang2+/− mice (FIG. 22A, arrows), but many more hyaloid vessels in ang2−/− mice, and they stained intensely for LacZ (FIG. 22C and FIG. 22D, arrows). Ang2−/− mice showed less GSA staining on the surface of the retina (FIG. 22D vs FIG. 22B, arrowheads) indicating fewer blood vessels, and the vessels that were present also showed LacZ labeling. At P10, there was complete regression of the hyaloid vasculature in ang2+/− mice (FIG. 22E), but ang2−/− mice still showed numerous hyaloid vessels that stained intensely for LacZ (FIG. 22G, arrows). Ang2+/− mice showed a well-developed superficial capillary bed with penetrating vessels extending into the retina (FIG. 22F, arrowhead), while ang2−/− mice showed no penetrating vessels (FIG. 22H, arrowheads). At P18 and P28, there were hyaloid vessels in ang2−/− (FIGS. 22K, 22L, 22O, 22P, arrows), but not in ang2+/− mice (FIGS. 22I, 22J, 22M, 22N), and the latter had a completely developed retinal vasculature with superficial, intermediate, and deep capillary beds, but the former had few vessels on the retinal surface and rare penetrating vessels (FIG. 22L, arrowheads) with no intermediate or deep capillary beds. These data confirm the findings in retinal whole mounts and demonstrate that in the absence of Ang2, the hyaloid vessels fail to regress and there is incomplete development of the retinal vessels, particularly the deep capillary beds. The few vessels that were present within the retinas of some ang2−/− mice indicate that while physiologic angiogenesis was markedly perturbed, it was not completely eliminated.

[0225] Effects of Ang2 Gene Deletion in Oxygen-Induced Ischemic Retinopathy

[0226] To investigate the effect of Ang2 deficiency on ischemia-induced retinal neovascularization, Applicants utilized the oxygen-induced ischemic retinopathy model (Smith, L. E. H., et al., (1994), 35:101-111). Briefly described, when wild type mice are placed in high oxygen at P7, some newly formed retinal vessels regress and then when the mice are placed in room air at P12, the retina is ischemic resulting in neovascularization by P17. When Ang2-deficient mice were placed in oxygen at P7, there was regression of the already sparse retinal vessels so that only a few retinal vessels remained. When the oxygen-treated Ang2-deficient mice were subsequently placed in room air, there was no retinal neovascularization. This is demonstrated in FIGS. 23A-23F which shows a retina from a P17 Ang2-deficient mouse that was exposed to high oxygen from P7 through P12 (FIGS. 23A, 23B, and 23C) compared to the retina of an Ang2 deficient mouse that was never exposed to high oxygen (FIGS. 23D, 23E, and 23F). A cross-section through the midperipheral retina of the oxygen-exposed mouse shows little GSA staining indicating few retinal vessels and no neovascularization (FIG. 23B). A corresponding section from the unexposed Ang2-deficient mouse shows more retinal vessels within the retina (FIG. 23E, arrowheads). In both types of mice, the far periphery of the retina was avascular (FIGS. 23C and 23F). A total of 3 oxygen-exposed Ang2-deficient mice were examined and they all showed no retinal neovascularization despite what must have been very ischemic retina due to few retinal blood vessels. They also showed persistence of the hyaloid vasculature.

[0227] Discussion

[0228] Mice deficient in Ang2 show abnormal postnatal vascular remodeling, which in the eye is manifested by lack of regression of the hyaloid vasculature and incomplete development of the retinal vasculature. In this study, Applicants have shown that the abnormal retinal vascular development consists of some formation of the superficial capillary bed, but almost complete lack of vessel penetration into the retina resulting in absence of the intermediate and deep capillary beds. In addition, Ang2-deficient mice fail to develop ischemia-induced retinal neovascularization.

[0229] Applicants' previous expression data are consistent with this interpretation. Expression of ang2 mRNA in the retina is low between P0 and P7 when the superficial capillary bed is developing and peaks on P8 when sprouting from the superficial capillaries occurs to form the deep capillary bed (Hackett, S. F., et al., (2000) J. Cell. Physiol., 184:275-284). Also, ischemia results in enhanced retinal expression of ang2 mRNA.

[0230] Ang2 binds to Tie2, but does not typically stimulate phosphorylation and thereby acts as a natural, context-specific antagonist of Tie2. Transgenic mice that overexpress Ang2 have a phenotype similar to both Tie2-deficient and Ang1-deficient mice in which blood vessels form, but fail to mature properly (Sato, T. N., et al., (1995) 376:70-74; Suri, C., et al., (1996) Cell, 87:1171-1180.). In the corpus luteum of the ovary there is a cycle of quiescence, vascular growth, and vascular regression. During quiescence Ang2 and VEGF levels are low, during vascular growth Ang2 and VEGF levels are high, and during vascular regression, Ang2 levels are high and VEGF levels are low (Maisonpierre, P. C., et al., (1997) Science, 277:55-60). These findings led to the hypothesis, set forth above in Examples 16 and 17, and FIGS. 15 and 16, that Ang2 collaborates with VEGF to help regulate vascular growth and regression. In the absence of Ang2, Tie2 signaling is intact and endothelial cells associate normally with their supporting cells and matrix, which makes them less responsive to soluble signals such as VEGF and thereby promotes quiescence. In the presence of Ang2, Tie2 signaling is blocked which perturbs these interactions making endothelial cells more responsive to and dependent upon VEGF or other soluble survival- and growth-promoting signals; in the presence of VEGF, vascular growth occurs, but in the absence of VEGF both soluble and matrix-derived survival signals are lacking and vascular regression occurs. This hypothesis is further supported by the present study, which demonstrates that normal regression of the hyaloid vasculature, and both physiologic and pathologic angiogenesis in the retina, require Ang2.

[0231] Conclusion

[0232] It has been demonstrated that the temporal and spatial expression of Ang2 in the retina is consistent with the hypothesis that it plays a role in physiologic and pathologic neovascularization in the retina (Hackett, S. F., et al., (2000) J. Cell. Physiol., 184:275-284). Targeted disruption of the ang2 gene provides a powerful way to explore the functions of Ang2, and an initial study has demonstrated that Ang2-deficient mice show lack of regression of the hyaloid vasculature and very abnormal retinal vascular development. In this study, Applicants have extended those findings to demonstrate that in the absence of Ang2, retinal vasculogenesis remains at least partially in tact, but both physiologic and pathologic angiogenesis are severely perturbed.

Example 23

Ang2 Plays an Important Role in two Forms of Ocular Neovascularization

[0233] Introduction

[0234] Mice deficient in Ang2 show lack of regression of the hyaloid vasculature and abnormal retinal vascular development. In this study, Applicants used Ang2-deficient mice to more broadly explore the role of Ang2 in ocular neovascularization.

[0235] Materials and Methods

[0236] Ang2-deficient mice and littermate controls were compared in 2 models of ocular neovascularization: oxygen-induced ischemic retinopathy (OIR) and ectopic expression of VEGF in photoreceptors (rho/VEGF transgenic mice).

[0237] Results

[0238] Oxygen-induced ischemic retinopathy. As described in detail above, in the OIR model, exposure of Ang2-deficient mice to high levels of oxygen resulted in partial regression of the already somewhat sparse retinal vessels. When these oxygen-exposed mice with few retinal vessels were moved to room air, there was no ischemia-induced retinal neovascularization, while littermate controls developed extensive retinal neovascularization.

[0239] Ectopic expression of VEGF in photoreceptors. Otherwise normal mice that carry a transgene expressing human VEGF regulated by the bovine rhodopsin promoter (V6) develop neovascular sprouts arising from the deep capillary bed which grow down into the subjacent photoreceptor layer, toward the subretinal space. Mice deficient in Ang2 were crossed with rho/VEGF transgenic mice. Subsequently, F1 ang+/− mice that carried a rho/VEGF transgene were crossed with ang+/− mice. At P21, wild type and ang+/−offspring that carried a rho/VEGF transgene exhibited extensive retinal neovascularization, while ang−/− mice that carried a rho/VEGF transgene had little evidence of neovascularization. Like wild-type mice that carry the rho/VEGF transgene, mice that are heterozygous for the ang2 mutation and rho/VEGF develop sprouts that appear as hyperfluorescent spots under low-power magnification of fluorescein-perfused flat-mounted retinas (FIG. 24A), arrows). The fine microvascular structure of the sprouts can be seen under higher power (inset). In contrast, mice homozygous for the ang2 mutation and heterozygous for rho/VEGF develop only a limited deep capillary bed in the central retina, and as shown in FIG. 24B, the number of sprouts that develop is far less than in heterozygotes, even considering the decreased area of the deep capillary bed (arrows and inset).

[0240] Conclusions

[0241] Mice deficient in Ang2 develop little or no ocular neovascularization in two different models of ocular neovascularization. These data suggest that Ang2 plays an important role in the development of pathologic neovascularization.

[0242] Clinical Relevance

[0243] Ocular neovascularization is an extremely common clinical problem. Retinal neovascularization in patients with diabetic retinopathy accounts for the most common cause of new blindness in individuals younger than 60 in developed countries and choroidal neovascularization in patients with age-related macular degeneration accounts for the majority of severe vision loss in patients over 60 (The Macular Photocoagulation Study Group. (1991) Arch. Ophthalmol., 109:1109-1114.; Klein, R., et al., (1984) Arch. Ophthalmol., 102:520-526). Applicants' data suggest that Ang2 and other components of the Tie2 signaling pathway provide molecular targets for development of new treatments for ocular neovascularization. As demonstrated, Ang2 collaborates with VEGF to promote neovascularization. Therefore, inhibiting or blocking Ang2 should be a viable treatment option for disorders characterized by ocular neovascularization. Non-limiting examples of ocular neovascularization disorders include proliferative retinopathies characterized by pre-retinal or optic disc neovascularization, such as diabetic retinopathy, venous obstructive disease, ocular ischemic syndrome, retinopathy of prematurity, hemoglobinopathies (e.g. sickle cell anemia), radiation retinopathy, ocular inflammation or infection, ocular neoplasias, tumors and metastases, ocular trauma or injury, or ocular neovascularization occurring idiopathically. Choroidal neovascularization disorders and diseases characterized by abnormal growth of new vessels, that originate in the choroids, penetrate Bruch's membrane and proliferate beneath the retinal pigment epithelium and/or neural retina, include, age related macular degeneration, angiod streaks, radiation retinopathy, ocular inflammation or infection, ocular neoplasias, tumors and metastases, ocular trauma or injury, or those occurring idiopathically.

Example 24

The Effect of Ang2 on Neovascularization in Tumors

[0244] An experiment was performed in which Lewis lung carcinoma cells were implanted subcutaneously in adult Ang2 knockout (Ang2−/−) mice. The experimental groups were as follows: two Ang2 knockout mice with tumors, two Ang2 heterozygous mice (Ang2+/−), and two wildtype controls (Ang2+/+). The tumors were allowed to grow for 2½ weeks and then observed qualitatively. The tumors in the Ang2 knockout mice (FIGS. 25B and 25D) appeared to be more than 50% smaller and less well vascularized as compared to the Ang2 heterozygotes (FIG. 25C) and wild type controls (FIG. 25A). These data demonstrate that mice deficient in Ang2 exhibit significantly decreased neovascularization in tumors.

[0245] Conclusions

[0246] These findings support previous expression data suggesting that Ang2 is important in pathological neovascularization and that inhibiting Ang2 mediated biological effects such as those leading to pathological neovascularization could ameliorate tumor angiogenesis.

[0247] The apparent role of Ang2 in such diverse types of pathological neovascularization clearly suggests the broad clinical applicability of treating patients with agents capable of interfering with Ang2-mediated biological effects. In addition to treating tumors and pathological ocular neovascularization disorders such as those described herein, any disorder characterized by pathological neovascularization would be amenable to such intervention, including but not limited to disorders and diseases of the female reproductive tract such as endometriosis and polycystic ovary syndrome, would also be.

Deposits

[0248] The following have been deposited with the American Type Culture Collection, 10801 University Boulevard, Manassas, Va. 20110-2209 in accordance with the Budapest Treaty. A plasmid clone encoding a TIE-2 ligand was deposited with the ATCC on Oct. 7, 1994 and designated as “pJFE14 encoding TIE-2 ligand” under ATCC Accession No. 75910. Recombinant Autographa californica baculovirus encoding TIE-2 receptorbody was deposited with the ATCC on Oct. 7, 1994 and designated as “vTIE-2 receptorbody” under ATCC Accession No. VR2484. A lambda phage vector containing human tie-2 ligand DNA was deposited with the ATCC on Oct. 26, 1994 and designated as λgt10 encoding htie-2 ligand 1 under ATCC Accession No. 75928. A plasmid clone encoding a second TIE-2 ligand was deposited with the ATCC on Dec. 9, 1994 and designated as “pBluescript KS encoding human TIE 2 ligand 2” under ATCC Accession No. 75963.

[0249] The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will become apparent to those skilled in the art from the foregoing description and accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

	Number	Date	Country
Parent	08930721	Mar 1998	US
Child	09442717	Nov 1999	US
Parent	PCT/US96/04806	Apr 1996	US
Child	09442717	Nov 1999	US

	Number	Date	Country
Parent	09442717	Nov 1999	US
Child	10186817	Jul 2002	US

TIE-2 ligands, methods of making and uses thereof

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

US Classifications

International Classifications

Abstract

Description

Claims

Parent Case Info

Continuations (2)

Continuation in Parts (1)