Fibroblast growth factor

FIELD OF THE INVENTION

[0002] The invention relates to compositions and methods for inhibiting cell proliferation, especially angiogenesis. The invention specifically relates to fusions of the extracellular domain of a fibroblast growth factor receptor (FGFR) with an immunoglobulin (Ig), especially an Ig Fc region, as an FGFR antagonist. The invention also relates to novel FGFR-Ig fusion proteins.

BACKGROUND OF THE INVENTION

[0003] Angiogenesis, or development of new blood vessels, is implicated in a host of diseases including tumorigenesis, metastasis and tumor growth, retinopathies, neovascular ocular disorders, and postangioplasty or postatherectomy restenosis (Bicknell et al. (1996) Curr. Opin. Oncol. 8:60-65; Gariano et al. (1996) Survey Ophthalmol. 40:481-490; and Wilcox (1993) Am. J. Cardiol. 72:88E-95E).

[0004] Expression and secretion of angiogenic factors by tumors has been investigated. It has been suggested that because tumors express multiple angiogenic factors, broad spectrum antagonists of angiogenesis can provide effective means of tumor stabilization. Anti-angiogenic approaches to tumor therapy have been defined to involve interference with growth, migration, and differentiation of blood vessels associated with tumor growth. Anti-angiogenic agents have been categorized to include protease inhibitors, modulators of cytokines, heparin-like molecules, and antagonists of vascular growth factors. Growth factor antagonists have been categorized to include heparin-like molecules, angiogenin antagonists, antisense fibroblast growth factor, DS 4152, suramin analogs, and protein-bound saccharide-K (Bicknell et al. (1996) Curr. Opin. Oncol. 8:60-65).

[0005] Various growth factors and growth factor receptors are known to be associated with particular types of tumors. At the molecular level, growth factors and growth factor receptors belong to multi-member families categorized based on structural and functional characteristics. Fibroblast growth factor (FGF) is involved in growth and differentiation of a number of cell types, and can contribute significantly to tumorigenicity. The FGF family includes FGF-1 or acidic FGF (aFGF), FGF-2 or basic FGF (bFGF), FGF-7 or KGF, oncogene products FGF-3 or int-2, hsp/Kaposi-FGF (K-FGF or FGF-4), FGF-5, and FGF-6. These members of the FGF family bind heparin, may exhibit mitogenic activity toward various cells, and may be potent mediators of angiogenesis (Pontaliano et al. (1994) Biochemistry 33:10229-10248; Kiefer et al. (1991) Growth Factors 5:115-127).

[0006] FGF receptor (FGFR) includes FGFR1 or flg, FGFR2 or bek, FGFR3 or cek2, and FGFR4 (Kiefer et al. (1991) Growth Factors 5:115-127). FGFR belongs to the tyrosine kinase family of receptors and to the immunoglobulin (Ig) supergene family. mRNA splicing variants of FGFR exist that produce secreted and transmembrane forms of the receptors with various ligand binding affinities and specificities. In transmembrane forms of the receptor, the tyrosine kinase domain is intracellular and the (Ig)-like domains are extracellular. Both transmembrane and secreted forms bind FGF. Heparin and related compounds promote the interaction between FGF and FGFR by acting as cofactors in dimerization or higher oligomerization of FGFR. The dimerization process is thought to be necessary for activation of FGFR.

[0007] FGFR fusion proteins present the possibility of constructing preoligomerized, particularly predimerized forms of FGFR. Such preoligomerized forms would be useful as potent and therapeutically effective inhibitors of FGF-mediated cell proliferation. FGFR antagonists would be especially useful to treat diseases involving angiogenesis.

[0008] Monomeric forms of the FGFR extracellular domain have been used to inhibit FGF-mediated events (Kiefer et al. (1991) Growth Factors 5:115-127). However, preoligomerized forms of the FGFR extracellular domain have not been used as FGFR antagonists. Thus, there is a need for providing preoligomerized forms of FGFR extracellular domain as antagonists of FGFR. Given the implicated role of this ligand/receptor system in angiogenesis, and the breadth of involvement of angiogenesis in several malignancies and other disorders, the approach promises a useful tool in providing an effective therapy for such disorders.

[0009] Receptor-immunoglobulin (Ig) fusion proteins have been used in the art. For example, an Ig fusion protein with a human tumor necrosis factor receptor has been applied to treatment of rheumatoid arthritis and septic shock (Moreland et al. (1997) New Engl. J. Med. 337:141-147; Fisher et al. (1996) New Engl. J. Med. 334:1697-1702). An Ig fusion protein with urokinase plasminogen activator (uPA) has been used as a uPA receptor antagonist to inhibit angiogenesis and tumor growth (Min et al. (1996) Cancer Res. 56:2428-2433). WO 95/21258 describes using FGFR-Ig fusion proteins in a method of identifying agonists and antagonists of FGFR. However, construction and use of the specific FGFR-Ig fusion proteins as antagonists of FGFR has not been suggested. Other examples of receptor-Ig fusion proteins include those described in U.S. Pat. Nos. 5,726,044; 5,707,632; and 5,750,375.

SUMMARY OF THE INVENTION

[0010] The invention is directed at providing oligomerized forms of FGFR as FGFR antagonists, constructed by fusing extracellular domains of FGFR with heterologous oligomerization domains. Compositions comprising polypeptides and polynucleotides encoding the fusion polypeptides are provided, as well as methods of using the compositions for treating disorders mediated by FGF, FGFR or angiogenesis, such as cancer and other hyperproliferative diseases.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011]
FIG. 1 depicts in vitro binding of 125I-bFGF to FGFR. FGFR extracellular domain monomer, FGFR extracellular domain-Ig Fc fusion dimer, and bFGF are compared as competitors of 125I-bFGF binding to immobilized FGFR receptors. Bound 125I-bFGF is plotted against a range of concentrations of each competitor.

[0012]
FIG. 2 depicts binding of 125I-bFGF to FGFR: whole cells. FGFR extracellular domain monomer and FGFR extracellular domain-Ig Fc fusion dimer are compared as competitors of 125I-bFGF for binding to stable HEK293 cell lines overexpressing FGFR1. Bound 125I-bFGF is plotted against a range of concentrations of each competitor.

[0013]
FIG. 3 depicts bFGF/serum stimulated HUVEC proliferation. FGFR extracellular domain-Ig Fc fusion dimer is tested for its ability to inhibit proliferation of human umbilical vein endothelial cells (HUVEC) in serum- and bFGF-containing media. Optical density (OD) indicating growth is plotted against time after dosing.

[0014]
FIG. 4 depicts a comparison of FGFR extracellular domain monomer and FGFR extracellular domain-Ig Fc dimmer on HUVEC proliferation. FGFR extracellular domain monomer and FGFR extracellular domain-Ig Fc fusion dimer are compared for their ability to inhibit HUVEC proliferation at indicated doses.

[0015]
FIG. 5 depicts a comparison of FGFR extracellular domain-Ig Fc to anti-bFGF Mab. FGFR extracellular domain-Ig Fc fusion protein and an anti-bFGF monoclonal antibody (Mab) are compared as competitors of 125I-bFGF binding to immobilized FGF receptors. Bound 125I-bFGF is plotted against a range of concentrations of each competitor.

DETAILED DESCRIPTION OF THE INVENTION

[0016] Compositions

[0017] The invention is based on means to provide potent FGF receptor (FGFR) antagonists. Dimeric or higher-order oligomeric forms of FGF are required for activation of FGFR, which proceeds through receptor dimerization (Moy et al. (1997) Biochemistry 36:4782-4791; Pontaliano et al. (1994) Biochemistry 33:10229-10248). Accordingly, the invention provides polypeptides capable of forming oligomers of FGFR, particularly dimers. The polypeptides comprise FGFR extracellular domains fused to heterologous oligomerization domains. Such oligomerized FGF receptors are provided as FGFR antagonists with higher potency relative to the monomeric FGFR antagonist. More particularly, the invention provides polypeptides comprising specific FGFR extracellular domains fused to regions of immunoglobulin (Ig) molecules which are known to be capable of forming oligomers with other Ig regions.

[0018] “FGFR extracellular domain” as used herein includes that portion of FGFR that is extracellular in native transmembrane forms of the receptor, or is of such extracellular origin, or consists of all or part of the naturally secreted forms of the receptor. It is understood, however, that the extracellular domain could contain other regions of the FGFR receptor (i.e., non-extracellular portions) as long as these portions do not interfere with or-significantly alter the function of the extracellular domain that is relevant to the methods described herein.

[0019] The FGFR extracellular domain shares homology with the immunoglobulin supergene family. The FGFR extracellular domain contains Ig-like segments. Furthermore, extracellular domains of different members of the FGFR family contain different numbers of Ig-like segments. The Ig-like segments are classified by their position relative to the amino-terminus of the FGF receptor, and by sequence homology to known Ig-like domains. Such classifications are known by a person of ordinary skill in the art. See, for example, Pontaliano et al. (1993) Biochemistry 33:10229-10248; Note 1, 10229.

[0020] The Ig-like segments are generally designated by numbering according to the relative positions of the segments from the amino terminus of FGFR with three Ig-like domains. Such designations are known by a person of ordinary skill in the art and are used for the purposes of this application, unless indicated otherwise. For example, see Pontaliano et al. (1994) Biochemistry 33:10229-10248. Accordingly, in forms of FGFR having three Ig-like segments, the Ig I segment is classified as the first Ig-like segment of an FGFR extracellular domain from the amino terminus of the molecule, the Ig II segment as the second and the Ig III segment as the third.

[0021] In naturally or artificially truncated forms of FGFR which have less than three Ig-like segments by virtue of the truncation of the extracellular domain, the numbering of the Ig-like segments are retained as they were prior to the truncation. The numbers designating the Ig-like segments are not reassigned according to relative positions of the segments from the amino-terminus subsequent to the truncation. Such designations are known in the art and are used for the purposes of this application, unless indicated otherwise. For example, a truncated form of FGFR containing two Ig-like domains due to deletion of a region encompassing the Ig I segment, is referred to as containing the Ig II and Ig III segments, although in this truncated form, the Ig II segment is the first Ig-like segment from the amino-terminus and the Ig III segment is the second.

[0022] Extracellular domains of FGFR containing different numbers and types of such Ig-like segments are capable of binding various forms of FGF. The affinity and specificity of this binding is at least partially attributable to the type of Ig-like segment contained within the extracellular domain.

[0023] Thus, the invention relates to providing variants of the extracellular domain of FGFR such that FGFR antagonists can be created based on FGF binding capability, affinity, and specificity.

[0024] By “variants” is intended substantially similar sequences. Thus, for nucleotide sequences, variants include those sequences that encode corresponding parts of the fusion polypeptides of the invention, but that differ conservatively because of the degeneracy of the genetic code. These naturally occurring allelic variants can be identified with the use of well-known molecular biology techniques, such as polymerase chain reaction (PCR) and hybridization techniques as outlined below.

[0025] Variant nucleotide sequences also include synthetically derived nucleotide sequences that have been generated, for example, by using site-directed mutagenesis but which still encode corresponding parts of the fusion polypeptide sequences provided in the present invention as discussed below. Generally, nucleotide sequence variants of the invention will have at least 70%, preferably at least 80%, more preferably about 90 to 95% or more, and most preferably about 98% or more sequence identity to the provided nucleotide sequence.

[0026] A variant polypeptide can differ in amino acid sequence by one or more substitutions, deletions, insertions, inversions, fusions, and truncations, or a combination thereof. With respect to the amino acid sequences for various domains of the fusion polypeptides, variants include those domains that are derived from corresponding native domains by deletion (so-called truncation) or addition of one or more amino acids to the N-terminal and/or C-terminal end of the polypeptide; deletion or addition of one or more amino acids at one or more sites in the polypeptide; or substitution of one or more amino acids at one or more sites in the polypeptide. Such variants may result from, for example, genetic polymorphism or from human manipulation. Methods for such manipulations are generally known in the art.

[0027] Variants of the FGFR extracellular domain include deletion variants. In one embodiment, amino acid segments that do not contribute to the desired capability, affinity, and specificity of binding of FGF are deleted, while those segments that contribute favorably to these functional properties are retained.

[0028] Deletion variants also include those in which deletion of particular amino acid segments positively affects the desired affinity and binding of FGF.

[0029] Deletion could comprise any segment of the extracellular domain including but not limited to the Ig I segment. In this aspect, the invention provides FGFR extracellular domain fusion polypeptides which lack the Ig I segment. Natural variants of FGFR containing the Ig I segment and those which lack the Ig I segment are capable of binding acidic FGF (aFGF) and basic FGF (bFGF) (Kiefer et al. (1991) Growth Factors 5:115-127; Johnson and Williams (1993) Adv. Cancer. Res. 60:1-41). Thus, the Ig I segment is not necessary for binding of aFGF and bFGF. The Ig I deletion further increases the affinity for aFGF and heparin, protects the core of the molecule from proteolysis, and abrogates the heparin requirement for aFGF binding.

[0030] Deletions can range from portions of a segment to deletion of an entire segment. Further, deletions can include one or more deletions in one or more of the Ig-like segments of the FGFR extracellular domain.

[0031] The invention further provides fusion polypeptides in which the FGFR extracellular domain lacks the acid box segment. The acid box segment is a known common feature of the FGFR extracellular domain and is characterized by multiple acidic amino acid motifs. For example, see Kiefer et al. (1991) Growth Factors 5:115-127.

[0032] The Ig II segment is typically defined as the second Ig-like segment from the amino-terminus of an FGFR with three Ig-like segments. The invention encompasses oligomerized FGFR extracellular domains comprising polypeptides which are variants with respect to the Ig II segment. For example, it is recognized that fusion polypeptides lacking both Ig I and Ig II extracellular segments may have favorable FGF binding characteristics with respect to affinity and specificity of binding and be useful as an FGFR antagonist.

[0033] The Ig III segment is typically defined as the third Ig-like segment from the amino-terminus of an FGFR with three Ig-like segments. Sequence variants of the C-terminal half of the Ig III segment are associated with differential FGF binding affinities and specificities. The IIIc variant of the Ig III segment binds aFGF and bFGF with an equal affinity, higher than that for FGF-7. The IIIb variant binds aFGF and FGF-7 with an equal affinity, higher than that for bFGF. The IIIa variant binds bFGF with a higher affinity than that for aFGF. The IIIc variant is the most widely expressed natural variant of the Ig III segment, and its deletion decreases the affinity for all ligands of the receptor. Further descriptions of the Ig IIIa, Ig IIIb, and Ig IIIc sequence variants are provided in Werner et al. (1992) Mol. and Cell Biol. 12:82-88, herein incorporated by reference.

[0034] The invention provides fusion polypeptides comprising sequence variants of the Ig III segment. The selection of the particular Ig III variants is based on the desired FGF binding affinities and specificities.

[0035] More preferred is a fusion polypeptide in which the FGFR extracellular domain comprises the IIIc sequence variant of the Ig III segment. The invention encompasses fusion polypeptides in which the FGFR extracellular domain comprises the IIIa or the IIIb sequence variants of the Ig III segment.

[0036] The invention further relates to providing a fusion polypeptide in which the FGFR extracellular domain comprises combinations of the above-described variants, including those with deletions of the Ig I segment, those with deletions of the acid box segment, and those comprising sequence variants of the Ig III segment.

[0037] More specifically, the invention provides human FGFR I extracellular domain fusion polypeptides having the amino acid sequences set forth in SEQ ID NO:2, SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:8, SEQ ID NO:10, and SEQ ID NO:12. The polypeptides having the sequences set forth in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:8, SEQ ID NO:10, and SEQ ID NO:12 comprise FGFR extracellular domain deletion variants of human FGFR1; lacking one or more segments as described above. Polynucleotide sequences encoding the above polypeptides are also provided and set forth in SEQ ID NO:1, SEQ ID NO:3, SEQ ID NO:5, SEQ ID NO:7, SEQ ID NO:9, and SEQ ID NO:11.

[0038] The FGF receptors of the invention also encompass polypeptides and polypeptide fragments that comprise FGFR extracellular domains of other members of the FGFR family, including but not limited to FGFR2, FGFR3, and FGFR4 (Kiefer et al. (1991) Growth Factors 5:115-127).

[0039] The invention also encompasses FGFR fusion polypeptides comprising other FGFR variants. These variants include substantially homologous FGFR proteins encoded by the same genetic locus, i.e., an allelic variant. The variants also include splicing variants of FGFR. The variants also encompass proteins derived from other genetic loci, but having substantial homology to the provided FGFR. The variants also include proteins substantially homologous to the provided FGFR but derived from another organism (i.e., non-human), i.e., an ortholog. The variants also include proteins substantially homologous that are produced by chemical synthesis. The variants also include proteins substantially homologous that are produced by recombinant methods.

[0040] “Homology” refers to the degree of similarity between x and y. The correspondence between the sequence from one form to another can be determined by techniques known in the art. For example, they can be determined by a direct comparison of the sequence information of the polynucleotide. Typically, two sequences, either polynucleotide or polypeptide, are homologous if the sequences exhibit at least 45% sequence identity; more typically, 50% sequence identity; more typically, 55% sequence identity; more typically, 60% sequence identity; more typically, 65% sequence identity; even more typically, 70% sequence identity. Usually, two sequences are homologous if the sequences exhibit at least 75% sequence identity; more usually, 80% sequence identity; even more usually, 85% sequence identity; even more usually, 90% sequence identity; and even more usually, 95% sequence identity.

[0041] Thus, the invention encompasses polynucleotides having 75%, preferably 80%, more preferably 85%, even more preferably 90%, and most preferably 95% or greater sequence identity to the polynucleotide sequences set forth in SEQ ID NOS:1, 3, 5, 7, 9, and 11. The invention further encompasses polypeptides having 75%, preferably 80%, more preferably 85%, even more preferably 90%, and most preferably 95% or greater sequence identity to the polypeptide sequences set forth in SEQ ID NOS:2, 4, 6, 8, 10, and 12).

[0042] Alternatively, homology can be determined by hybridization of the polynucleotides under conditions which form stable duplexes between homologous regions. Stable duplexes are those, for example, which would withstand digestion with a single-stranded specific nuclease(s), such as S1. Such duplexes can be analyzed by various methods, such as size determination of digested fragments.

[0043] “Hybridization” refers to the association of two nucleic acid sequences to one another by hydrogen bonding. Typically, one sequence will be fixed to a solid support and the other will be free in solution. Then, the two sequences will be placed in contact with one another under conditions that favor hydrogen bonding. Factors that affect this bonding include: the type and volume of solvent; reaction temperature; time of hybridization; agitation; agents to block the non-specific attachment of the liquid phase sequence to the solid support (Denhardt's reagent or BLOTTO); concentration of the sequences; use of compounds to increase the rate of association of sequences (dextran sulfate or polyethylene glycol); and the stringency of the washing conditions following hybridization. See Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Volume 2, chapter 9, pages 9.47 to 9.57.

[0044] “Stringency” refers to conditions in a hybridization reaction that favor association of very similar sequences over sequences that differ. For example, the combination of temperature and salt concentration should be chosen that is approximately 12° C. to 20° C. below the calculated Tm of the hybrid under study. The temperature and salt conditions can often be determined empirically in preliminary experiments in which samples of genomic DNA immobilized on filters are hybridized to the sequence of interest and then washed under conditions of different stringencies. See Sambrook et al., above, at page 9.50.

[0045] Variables to consider when performing, for example, a Southern blot are (1) the complexity of the DNA being blotted and (2) the homology between the probe and the sequences being detected. The total amount of the fragment(s) to be studied can vary a magnitude of 10, from 0.1 to 1 μg for a plasmid or phage digest to 10−9 to 10−8 μg for a single copy gene in a highly complex eukaryotic genome. For lower complexity polynucleotides, substantially shorter blotting, hybridization, and exposure times, a smaller amount of starting polynucleotides, and lower specific activity of probes can be used. For example, a single-copy yeast gene can be detected with an exposure time of only 1 hour starting with 1 μg of yeast DNA, blotting for two hours, and hybridizing for 4-8 hours with a probe of 108 cpm/μg. For a single-copy mammalian gene a conservative approach would start with 10 μg of DNA, blot overnight, and hybridize overnight in the presence of 10% dextran sulfate using a probe of greater than 108 cpm/μg, resulting in an exposure time of ˜24 hours.

[0046] Several factors can affect the melting temperature (Tm) of a DNA-DNA hybrid between the probe and the fragment of interest, and consequently, the appropriate conditions for hybridization and washing. In many cases the probe is not 100% homologous to the fragment. Other commonly encountered variables include the length and total G+C content of the hybridizing sequences and the ionic strength and formamide content of the hybridization buffer. The effects of all of these factors can be approximated by a single equation:

T

m
=81+16.6(log 10Ci)+0.4[% G+C)]−0.6(% formamide)−600/n−1.5 (% mismatch)

[0047] where Ci is the salt concentration (monovalent ions) and n is the length of the hybrid in base pairs (slightly modified from Meinkoth and Wahl (1984) Anal. Biochem. 138:267-284).

[0048] In designing a hybridization experiment, some factors affecting nucleic acid hybridization can be conveniently altered. The temperature of the hybridization and washes and the salt concentration during the washes are the simplest to adjust. As the temperature of the hybridization increases (i.e., stringency), it becomes less likely for hybridization to occur between strands that are nonhomologous, and as a result, background decreases. If the radiolabeled probe is not completely homologous with the immobilized fragment (as is frequently the case in gene family and interspecies hybridization experiments), the hybridization temperature must be reduced, and background will increase. The temperature of the washes affects the intensity of the hybridizing band and the degree of background in a similar manner. The stringency of the washes is also increased with decreasing salt concentrations.

[0049] In general, convenient hybridization temperatures in the presence of 50% formamide are 42° C. for a probe which is 95% to 100% homologous to the target fragment, 37° C. for 90% to 95% homology, and 32° C. for 85% to 90% homology. For lower homologies, formamide content should be lowered and temperature adjusted accordingly, using the equation above. If the homology and between the probe and the target fragment are not known, the simplest approach is to start with both hybridization and wash conditions which are nonstringent. If nonspecific bands or high background are observed after autoradiography, the filter can be washed at high stringency and reexposed. If the time required for exposure makes this approach impractical, several hybridization and/or washing stringencies should be tested in parallel.

[0050] To determine the percent homology of two amino acid sequences, or of two nucleic acids, the sequences are aligned for optimal comparison purposes (e.g., gaps can be introduced in the sequence of one protein or nucleic acid for optimal alignment with the other protein or nucleic acid). The amino acid residues or nucleotides at corresponding amino acid positions or nucleotide positions are then compared. When a position in one sequence is occupied by the same amino acid residue or nucleotide as the corresponding position in the other sequence, then the molecules are homologous at that position. As used herein, amino acid or nucleic acid “homology” is equivalent to amino acid or nucleic acid “identity.” The percent homology between the two sequences is a function of the number of identical positions shared by the sequences (i.e., percent homology equals the number of identical positions/total number of positions times 100).

[0051] The invention also encompasses fusion polypeptides having a lower degree of identity than those described above, but having sufficient similarity so as to perform one or more of the same functions performed by the fusion polypeptides of the invention. Similarity is determined by conserved amino acid substitution. Such substitutions are those that substitute a given amino acid in a polypeptide by another amino acid of like characteristics. Conservative substitutions are likely to be phenotypically silent. Typically seen as conservative substitutions are the replacements, one for another, among the aliphatic amino acids Ala, Val, Leu, and Ile; interchange of the hydroxyl residues Ser and Thr; exchange of the acidic residues Asp and Glu; substitution between the amide residues Asn and Gln; exchange of the basic residues Lys and Arg; and replacements among the aromatic residues Phe, Tyr. Guidance concerning which amino acid changes are likely to be phenotypically silent are found in Bowie et al. (1990) Science 247:1306-1310.

1TABLE 1Conservative Amino Acid Substitutions.AromaticPhenylalanineTryptophanTyrosineHydrophobicLeucineIsoleucineValinePolarGlutamineAsparagineBasicArginineLysineHistidineAcidicAspartic AcidGlutamic AcidSmallAlanineSerineThreonineMethionineGlycine

[0052] Both identity and similarity can be readily calculated (Computational Molecular Biology, ed. Lesk (Oxford University Press, New York, 1988); Biocomputing: Informatics and Genome Projects, ed. Smith (Academic Press, New York, 1993); Computer Analysis of Sequence Data, Part 1, ed. Griffin and Griffin (Humana Press, New Jersey, 1994); von Heinje (1987) Sequence Analysis in Molecular Biology (Academic Press); and Sequence Analysis Primer, ed. Gribskov and Devereux (M. Stockton Press, New York, 1991).

[0053] Preferred computer program methods to determine identity and similarity between two sequences include, but are not limited to, GCG program package (Devereux et al. (1984) Nucleic Acids Res. 12(1):387), BLASTP, BLASTN, FASTA (Atschul et al. (1990) J. Molec. Biol. 215:403).

[0054] A variant can differ in amino acid sequence by one or more substitutions, deletions, insertions, inversions, fusions, and truncations, or a combination of any of these.

[0055] Variant polypeptides can be fully functional or can lack function in one or more activities. Thus, in the present case, functions affected by variations include FGF binding capability, FGF binding affinity, FGF binding specificity, heparin binding, inhibition of FGF-stimulated cell proliferation, inhibition of FGF-mediated disorders, inhibition of FGFR-mediated disorders, inhibition of angiogenesis-mediated disorders, and inhibition of cancer and other hyperproliferative disorders.

[0056] Fully functional variants typically contain only conservative variation or variation in non-critical residues or in non-critical regions. Functional variants can also contain substitution of similar amino acids which result in no change or an insignificant change in function. Alternatively, such substitutions may positively or negatively effect function to some degree.

[0057] Non-functional variants typically contain one or more non-conservative amino acid substitutions, deletions, insertions, inversions, or truncation or a substitution, insertion, inversion, or deletion in a critical residue or critical region.

[0058] As indicated, variants can be naturally occurring or can be made by recombinant means or chemical synthesis to provide useful and novel characteristics for the receptor polypeptide. This includes preventing immunogenicity from pharmaceutical formulations by preventing protein aggregation.

[0059] Useful variations include alteration of ligand binding characteristics. For example, one specific embodiment encompasses fusion polypeptide dimers that bind aFGF and bFGF with equal affinity. A further embodiment encompasses fusion polypeptide dimers that bind aFGF and bFGF with equal affinity, but with higher than that for FGF-7. Another embodiment encompasses fusion polypeptide dimers that bind aFGF and bFGF with high affinity.

[0060] Another useful variation is one that provides for a protease cleavage site between the extracellular domain of FGFR and the Ig portion of the fusion polypeptide. While constructs containing the cleavage sites are not suited for in vivo use due to the presence of the cleavage site, they are useful as experimental controls. One resulting product from utilizing this cleavage site is FGFR extracellular domain monomer, which is useful as a control in assessing the functional characteristics of the corresponding FGFR extracellular domain-Ig dimer without the cleavage site.

[0061] The invention provides polypeptides comprising FGFR extracellular domains fused to heterologous oligomerization domains. By “heterologous oligomerization domain” is intended that domain of a polypeptide of the invention which is not naturally associated with the extracellular domain of the polypeptide, and is capable of forming oligomers, which are at least dimers, with other polypeptides. Specific examples of such heterologous oligomerization domains include, but are not limited to, the Fc region of an immunoglobulin molecule; the hinge region of an immunoglobulin molecule; the CH1 region of an immunoglobulin molecule; the CH2 region of an immunoglobulin molecule; the CH3 region of an immunoglobulin molecule; the CH4 region of an immunoglobulin molecule; the light chain of an immunoglobulin molecule; and a peptide comprising a leucine zipper motif.

[0062] Leucine zipper motifs are known in the art, and are typically found in some of the DNA-binding proteins. Leucine zippers are formed from a region of α-helix that contain at least four leucines, each leucine separated by six amino acids from one another (i.e., Leu-X6-Leu-X6-Leu-X6-Leu, where X is any common amino acid). See Devlin (1997) Textbook of Biochemistry with Clinical Correlations (4th ed., John Wiley and Sons, Inc.), pp. 110-111.

[0063] Examples of utilizing various immunoglobulin domains for constructing oligomeric fusion proteins are known in the art. See, for example, EP-A 0464 533 and U.S. Pat. No. 5,726,044, which describe fusion proteins comprising immunoglobulin Fc regions. See also U.S. Pat. No. 5,750,375, which describes fusion proteins comprising various heavy chain domains, as well as light chain kappa.

[0064] The heavy chain Fc region of an immunoglobulin molecule which contains the hinge region, but not the CH1 region of the heavy chain, is useful in therapy and diagnosis; and its incorporation into a protein results, for example, in improved pharmacokinetic properties of the protein. For example, see EP-A 0232 262. In drug discovery applications, human proteins have been fused with Fc portions for the purpose of high-throughput screening assays to identify antagonists (Bennett et al. (1995) Journal of Molecular Recognition 8:52-58 and Johanson et al. (1995) The Journal of Biological Chemistry 270, 16:9459-9471). This invention encompasses soluble fusion proteins containing a receptor polypeptide and various portions of the constant regions of heavy or light chains of immunoglobulins of various subclass (IgG, IgD, IgM, IgA, IgE). Preferred as immunoglobulin is the constant part of the heavy chain of human IgG, particularly IgG1, where dimerization between two heavy chains takes place at the hinge region. It is recognized that inclusion of the CH2 and CH3 domains of the Fc region as part of the fusion polypeptide increases the in vivo circulation half-life of the polypeptide comprising the Fc region, and that of the oligomer or dimer comprising the polypeptide.

[0065] Furthermore, the proteins of the invention are generally designed to eliminate or at least minimize the contribution of the Ig Fc fusion protein to immunogenic responses. To this end, native or mutated Ig Fc portions are preferred which have low or diminished affinity for Fc receptors, and have diminished capacities for interaction with complement; Duncan and Winter (1988) Nature 332:738-740; Xu et al. (1994) J. Biol. Chem. 269:3469-3474. For example, mutations of amino acids corresponding to Leu 235 and Pro 331 of human IgG1 to Glu and Ser respectively, are provided. More specifically, these mutations are provided as set forth in SEQ ID NOS:7 and 8 (Construct #4) and SEQ ID NOS:9 and 10 (Construct #5), and described in more detail in Example 6 below.

[0066] In order to express any of the fusion proteins of the invention in a secreted form, a signal peptide is typically contained at the N-terminus of the fusion protein. Generally, the signal peptide that is native to the FGFR extracellular domain is comprised by the fusion proteins of the invention. Alternatively, signal peptides that are heterologous with respect to the extracellular domain may be used.

[0067] Heparin is known to be required for optimal FGF binding to FGFR. Thus, in constructing the fusion proteins, the heparin binding site is generally retained as part of the FGFR extracellular domain.

[0068] Methods for testing the function of the fusion proteins of the invention include, but are not limited to, the following methods, herein incorporated by reference: in vitro and in situ growth factor binding assays (Pontaliano et al. (1994) Biochemistry 33:10229-10248; Kiefer et al. (1991) Growth Factors 5:115-127; U.S. Pat. No. 5,229,501); cell proliferation assays (U.S. Pat. No. 5,229,501; WST cell proliferation assay, Boehringer Mannheim); in vivo and ex vivo assessments of angiogenesis (Min et al. (1996) Cancer Res. 56:2428-2433; Bickness et al. (1996) Curr. Opin. Oncol. 8:60-65); and assessments of tumor growth (Kim et al. (1993) Nature 362:841-844; Millauer et al. (1993) Nature 367:576-579).

[0069] It is understood that other methodologies associated with particular pathologies, biological conditions, or processes may be employed when applicable to testing the fusion proteins of the invention. Examples of such methodologies include molecular biological, immunochemical, histochemical, and morphological assessments relevant to cell proliferation in restenosis (Wilcox (1993) Am. J. Cardiol. 72:88E-95E) and ocular diseases (Gariano et al. (1996) Survey Ophthalm. 40:481-490). Additional methodologies for assessing vascular density and tumor growth; biochemical assays utilizing markers of angiogenesis; and in vivo methods of assessing therapeutic effects of antiangiogenic agents disclosed by Bickness et al. (1996) Curr. Opin. Oncol. 8:60-65, may also be employed in testing the fusion proteins of the invention, and are herein incorporated by reference.

[0070] Polynucleotide Constructs and Host Cells

[0071] DNA constructs encoding the FGFR fusion polypeptides of the invention, DNA constructs capable of expressing the FGFR fusion polypeptides, and host cells containing or capable of expressing such constructs are also provided by the invention.

[0072] Routine techniques for the construction of the vectors comprising the fusion proteins of the present invention are well-known to those of ordinary skill in the art and can be found in such references as Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual (2nd ed., Cold Spring Harbor, N.Y.). A variety of strategies are available for ligating fragments of DNA, the choice of which depends on the nature of the termini of the DNA fragments and which choices can be readily made by those of skill in the art. Generally, and unless otherwise specified, the 3′ end of the DNA segment encoding the desired FGFR extracellular domains will be ligated in frame to the 5′ end of a DNA segment encoding the desired oligomerization domain, such that a contiguous fusion protein is produced upon expression of the ligated DNA. These strategies may encompass PCR techniques in obtaining or modifying pertinent DNA segments. Also available to one of ordinary skill in the art is a variety of host cells for containing and expressing the desired constructs.

[0073] Constructs include native sequences and variants, as well as sequences that hybridize under stringent conditions.

[0074] Methods of Using FGFR-IG Fusions

[0075] Working examples of the invention provide methods of inhibiting FGF-stimulated cell proliferation by administering an FGFR antagonist comprising a fusion of the extracellular domain of FGFR with the Fc region of an Ig molecule as the heterologous oligomerization domain in an amount effective to inhibit the proliferation. The invention also provides methods of inhibiting angiogenesis by administering to cells capable of undergoing angiogenesis the fusion protein in an amount effective to inhibit angiogenesis. The inhibition could be in vitro or in vivo.

[0076] “Amount effective to inhibit” is intended to mean that amount of the fusion protein which prevents or induces a measurable inhibition of FGF-stimulated cell proliferation. Methods of measuring such inhibition are known to those skilled in the art and include available commercial kits which are based on measuring numbers of viable cells.

[0077] Where the inhibition is in vivo, the amount effective to inhibit can induce a concentration of the fusion protein in the target organ or tissue needed to inhibit cell proliferation in the cells of the target organ or tissue.

[0078] The invention provides methods of treating disorders which are FGF-mediated, FGFR-mediated, or angiogenesis-mediated. The methods encompass administering therapeutically effective amounts of the polypeptides of the invention or vectors comprising polynucleotides encoding the polypeptides of the invention to a mammal. The administration can be alone or in conjunction with other agents, including other inhibitors of angiogenesis, tumorigenesis, anticancer agents, and the like.

[0079] Where the treating of a disorder involves administering polynucleotides comprising coding regions comprising the polypeptides of the invention, the polynucleotides are provided in expression vectors capable of expressing the polynucleotides in a particular organism, organ, tissue, or cell type; such that the coding region is operably linked to the promoter of the expression vector.

[0080] The therapeutic polynucleotides and polypeptides of the present invention may be utilized in gene delivery vehicles. The gene delivery vehicle may be of viral or non-viral origin (see generally, Jolly (1994) Cancer Gene Therapy 1:51-64; Kimura (1994) Human Gene Therapy 5:845-852; Connelly (1995) Human Gene Therapy 1:185-193; and Kaplitt (1994) Nature Genetics 6:148-153). Gene therapy vehicles for delivery of constructs including a coding sequence of a therapeutic of the invention can be administered either locally or systemically. These constructs can utilize viral or non-viral vector approaches. Expression of such coding sequences can be induced using endogenous mammalian or heterologous promoters. Expression of the coding sequence can be either constitutive or regulated.

[0081] The present invention can employ recombinant retroviruses which are constructed to carry or express a selected nucleic acid molecule of interest. Retrovirus vectors that can be employed include those described in EP 0 415 731; WO 90/07936; WO 94/03622; WO 93/25698; WO 93/25234; U.S. Pat. No. 5,219,740; WO 93/11230; WO 93/10218; Vile and Hart (1993) Cancer Res. 53:3860-3864; Vile and Hart (1993) Cancer Res. 53:962-967; Ram et al. (1993) Cancer Res. 53:83-88; Takamiya et al. (1993) J. Neurosci. Res. 33:493-503; Baba et al. (1993) J. Neurosurg. 79:729-735; U.S. Pat. No. 4,777,127; GB Patent No. 2,200,651; and EP 0 345 242. Preferred recombinant retroviruses include those described in WO 91/02805.

[0082] Packaging cell lines suitable for use with the above-described retroviral vector constructs may be readily prepared (see PCT publications WO 95/30763 and WO 92/05266), and used to create producer cell lines (also termed vector cell lines) for the production of recombinant vector particles. Within particularly preferred embodiments of the invention, packaging cell lines are made from human (such as HT1080 cells) or mink parent cell lines, thereby allowing production of recombinant retroviruses that can survive inactivation in human serum.

[0083] The present invention also employs alphavirus-based vectors that can function as gene delivery vehicles. Such vectors can be constructed from a wide variety of alphaviruses, including, for example, Sindbis virus vectors, Semliki forest virus (ATCC VR-67; ATCC VR-1247), Ross River virus (ATCC VR-373; ATCC VR-1246) and Venezuelan equine encephalitis virus (ATCC VR-923; ATCC VR-1250; ATCC VR 1249; ATCC VR-532). Representative examples of such vector systems include those described in U.S. Pat. Nos. 5,091,309; 5,217,879; and 5,185,440; and PCT Nos. WO 92/10578; WO 94/21792; WO 95/27069; WO 95/27044; and WO 95/07994.

[0084] Gene delivery vehicles of the present invention can also employ parvovirus such as adeno-associated virus (AAV) vectors. Representative examples include the AAV vectors disclosed by Srivastava in WO 93/09239, Samulski et al. (1989) J. Vir. 63:3822-3828; Mendelson et al. (1988) Virol. 166:154-165; and Flotte et al. (1993) PNAS 90:10613-10617.

[0085] Representative examples of adenoviral vectors include those described by Berkner (1988) Biotechniques 6:616-627; Rosenfeld et al. (1991) Science 252:431-434; WO 93/19191; Kolls et al. (1994) PNAS 91:215-219; Kass-Eisler et al. (1993) PNAS 90:11498-11502; Guzman et al. (1993) Circulation 88:2838-2848; Guzman et al. (1993) Cir. Res. 73:1202-1207; Zabner et al. (1993) Cell 75:207-216; Li et al. (1993) Hum. Gene Ther. 4:403-409; Cailaud et al. (1993) Eur. J. Neurosci. 5:1287-1291; Vincent et al. (1993) Nat. Genet. 5:130-134; Jaffe et al. (1992) Nat. Genet. 1:372-378; and Levrero et al. (1991) Gene 101:195-202. Exemplary adenoviral gene therapy vectors employable in this invention also include those described in WO 94/12649; WO 93/03769; WO 93/19191; WO 94/28938; WO 95/11984; and WO 95/00655. Administration of DNA linked to killed adenovirus as described in Curiel (1992) Human Gene Therapy 3:147-154 maybe employed.

[0086] Other gene delivery vehicles and methods may be employed, including polycationic condensed DNA linked or unlinked to killed adenovirus alone, for example Curiel (1992) Human Gene Therapy 3:147-154; ligand linked DNA, for example see Wu (1989) Biol. Chem. 264:16985-16987; eukaryotic cell delivery vehicles cells, for example see U.S. Ser. No. 08/240,030, filed May 9, 1994, and U.S. Ser. No. 08/404,796; deposition of photopolymerized hydrogel materials; hand-held gene transfer particle gun, as described in U.S. Pat. No. 5,149,655; ionizing radiation as described in U.S. Pat. No. 5,206,152 and in WO 92/11033; nucleic charge neutralization or fusion with cell membranes. Additional approaches are described in Philip (1994) Mol. Cell Biol. 14:2411-2418, and in Woffendin (1994) Proc. Natl. Acad. Sci. USA 91:1581-1585.

[0087] Naked DNA may also be employed. Exemplary naked DNA introduction methods are described in WO 90/11092 and U.S. Pat. No. 5,580,859. Uptake efficiency may be improved using biodegradable latex beads. DNA coated latex beads are efficiently transported into cells after endocytosis initiation by the beads. The method may be improved further by treatment of the beads to increase hydrophobicity and thereby facilitate disruption of the endosome and release of the DNA into the cytoplasm. Liposomes that can act as gene delivery vehicles are described in U.S. Pat. No. 5,422,120; PCT Nos. WO 95/13796, WO 94/23697, and WO 91/14445; and EP No. 0 524 968.

[0088] Further non-viral delivery suitable for use includes mechanical delivery systems such as the approach described in Woffendin et al. (1994) Proc. Natl. Acad. Sci. USA 91(24):11581-11585. Moreover, the coding sequence and the product of expression of such can be delivered through deposition of photopolymerized hydrogel materials. Other conventional methods for gene delivery that can be used for delivery of the coding sequence include, for example, use of hand-held gene transfer particle gun, as described in U.S. Pat. No. 5,149,655; use of ionizing radiation for activating transferred gene, as described in U.S. Pat. No. 5,206,152 and PCT No. WO 92/11033.

[0089] The invention also provides a method of diagnosing disorders involving FGF-stimulated cell proliferation, angiogenesis and tumorigenesis.

[0090] In one embodiment, the method encompasses administering to test cells an effective amount of the fusion proteins of the invention and assessing inhibition compared to a control.

[0091] In another embodiment of the invention, the above-described methods of diagnosis encompass in vitro administration of any of the described fusion protein antagonists to cells in culture.

[0092] In another embodiment of the invention, the above-described methods of diagnosis involve administering any of the described fusion proteins in vivo.

[0093] In one embodiment, therapeutically effective amounts of pharmacological compositions containing fusion protein antagonists as described herein are administered to a patient or an animal model in need of such administration. The methods of treatment or prevention encompass administering effective amounts of a pharmacological composition containing a fusion protein of the invention as described herein.

[0094] Methods of constructing therapeutically effective fusion protein antagonists are provided. In one embodiment of the invention, binding affinities and specificities of the fusion proteins are first characterized in vitro. Next, proteins with desired affinities and specificities against various forms of FGF are selected and further assessed for inhibition of FGF-stimulated cell proliferation. Next, new fusion proteins are constructed by deletion of segments determined not to be necessary for desired affinities and specificities of binding to FGF or high potency of inhibition of FGF-stimulated cell proliferation. The above-described assessments and selections are repeated with the smaller deletion constructs; until a minimal protein structure having the desired affinities, specificities and potencies is constructed. The therapeutic effectiveness of selected minimal constructs are then assessed in vivo.

[0095] The administration includes, but is not limited to, administration to animal models and patients manifesting the following disorders: restenosis after angioplasty or atherectomy (Wilcox (1993) Am. J. Cardiol. 72:88E-95E), ophthalmological disorders involving excessive vasoproliferation (Gariano et al. (1996) Survey Ophthalm. 40:481-490), various tumors and cancers (Kim et al. (1993) Nature 334:841-844; Kim et al. (1993) Nature 362:841-844; Min et al. (1996) Cancer Res. 56:2428-2433) including AIDS-related Kaposi sarcoma (Pontaliano et al. (1994) Biochemistry 33:10229-10248).

[0096] Pharmacological Compositions

[0097] The invention provides pharmacological compositions comprising the fusion polypeptides of the invention or polynucleotides encoding the polypeptides described herein. The pharmacological compositions may also contain any of the described variants of FGFR.

[0098] Compounds useful for formulating polypeptides and/or protein pharmaceutical compositions can be used with fusion proteins.

[0099] The pharmaceutical compositions will comprise a therapeutically effective amount of any of the proteins of the claimed invention.

[0100] The term “therapeutically effective amount” as used herein refers to an amount of a therapeutic agent to treat or prevent a disorder sufficient to exhibit a detectable preventive, ameliorative, curative or other therapeutic effect. The effect may include, for example, treatment, amelioration, or prevention of any physical or biochemical condition, for example, including but not limited to hyperproliferative growth, angiogenesis and cancer.

[0101] The effect can be detected by, for example, biochemical or histological means of assessing angiogenesis. Therapeutic effects also include reduction in physical symptoms, such as decreased tumor size. The precise effective amount for a subject will depend upon the subject's size and health, the nature and extent of the condition, and the therapeutics or combination of therapeutics selected for administration. Thus, it is not useful to specify an exact effective amount in advance. However, the effective amount for a given situation can be determined by routine experimentation.

[0102] In this aspect of the invention, by “FGF-mediated disorder,” “FGFR-mediated disorder,” or “angiogenesis-mediated disorder” is intended an adverse biological or biochemical condition that is exacerbated by FGF, FGFR, or angiogenesis. Examples of such disorders include, but are not limited to, tumorigenesis, neovascularization, hyper-proliferation of vascular smooth muscle cells, and the like.

[0103] Tumors include, but are not limited to, bladder, breast, node-negative breast, lung, rectal, colorectal, testis, and cervical tumors; glioblastoma; childhood brain tumors; squamous cancer of the tongue; etc.

[0104] Disorders involving neovascularization include, but are not limited to, diabetic retinopathy, retinopathy of prematurity (ROP), choroidal neovascularization, neovascular glaucoma, wound healing after surgery and injury, corneal scarring, ocular neoplasia, and breakdown of blood-retina barrier.

[0105] Disorders involving hyperproliferation of vascular smooth muscle cells include, but are not limited to, postangioplasty and postatherectomy restenosis.

[0106] It is recognized that depending on the type and stage of a particular disorder, the disorder may be mediated by FGF, FGFR, angiogenesis, or combinations thereof.

[0107] For purposes of the present invention, an effective dose will be from about 1 pg/kg to 10 mg/kg of the fusion protein in the individual to which it is administered.

[0108] A pharmaceutical composition can also contain a pharmaceutically acceptable carrier. The term “pharmaceutically acceptable carrier” refers to a carrier for administration of a therapeutic agent, such as antibodies or a polypeptide, genes, and other therapeutic agents. The term refers to any pharmaceutical carrier that does not itself induce the production of antibodies harmful to the individual receiving the composition, and which may be administered without undue toxicity. Suitable carriers may be large, slowly metabolized macromolecules such as proteins, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids, amino acid coknown to those of ordinary skill in the art.

[0109] Pharmaceutically acceptable salts can be used therein, for example, mineral acid salts such as hydrochlorides, hydrobromides, phosphates, sulfates, and the like; and the salts of organic acids such as acetates, propionates, malonates, benzoates, and the like. A thorough discussion of pharmaceutically acceptable excipients are available in Remington 's Pharmaceutical Sciences (Mack Pub. Co., NJ, 1991).

[0110] Pharmaceutically acceptable carriers in therapeutic compositions may contain liquids such as water, saline, glycerol and ethanol. Additionally, auxiliary substances, such as wetting or emulsifying agents, pH buffering substances, and the like, may be present in such vehicles. Typically, the therapeutic compositions are prepared as injectables, either as liquid solutions or suspensions; solid forms suitable for solution in, or suspension in, liquid vehicles prior to injection may also be prepared. Liposomes are included within the definition of a pharmaceutically acceptable carrier.

[0111] Administration Methods

[0112] Once formulated, the fusion protein compositions of the invention can be (1) administered directly to the subject; or delivered ex vivo, to cells derived from the subject.

[0113] Direct delivery of the compositions will generally be accomplished by injection, either subcutaneously, intraperitoneally, intravenously, or intramuscularly, or delivered to the interstitial space of a tissue. The compositions can also be administered into a tumor or lesion. Other modes of administration include oral and pulmonary administration, suppositories, and transdermal applications, needles, and gene guns, or hyposprays. Dosage treatment may be a single dose schedule or a multiple dose schedule.

[0114] Additional pharmaceutical methods may be employed to control the duration of action. Controlled release preparations may be achieved by the use of polymers to complex or absorb the compositions. The controlled delivery may be exercised by selecting appropriate macromolecules (for example, polyesters, polyamino acids, polyvinyl pyrrolidone, ethylene-vinylacetate, methylcellulose, carboxymethylcellulose, or protamine sulfate). The rate of drug release may also be controlled by altering the concentration of such macromolecules.

[0115] Another possible method for controlling the duration of action comprises incorporating the therapeutic agents into particles of a polymeric substance such as polyesters, polyamino acids, hydrogels, poly(lactic acid), or ethylene vinylacetate copolymers. Alternatively, it is possible to entrap the therapeutic agents in microcapsules prepared, for example, by coacervation techniques or by interfacial polymerization, for example, by the use of hydroxymethyl cellulose or gelatin-microcapsules or poly(methylmethacrylate) microcapsules, respectively, or in a colloid drug delivery system, for example, liposomes, albumin, microspheres, microemulsions, nanoparticles, nanocapsules, or in macroemulsions. Such teachings are disclosed in Remington's Pharmaceutical Sciences (1980).

[0116] A preferred controlled release composition for delivery of such compositions will encompass liposome encapsulation as described in, but not limited to U.S. Pat. No. 4,522,803; EP 0 280 503 B1; and WO 95/13796.

[0117] Detecting Cell Proliferation

[0118] The invention provides methods of diagnosing FGF-stimulated cell proliferation and angiogenesis by administering to test cells an effective amount of any of the fusion proteins and assessing inhibition against a control. The cells can be in vivo or in vitro. In this aspect, the administration could be directed at animal models or patients, including those described above, or directed at primary cultures or cell lines derived from tissues affected by disease states potentially involving FGF-stimulation. The method of diagnosis is, thus, useful for determining the propensity of an animal to respond to treatment with fusion protein antagonists as provided by the invention.

[0119] In this manner, the invention relates to providing receptor oligomer fusion constructs which can effect clinically determinable inhibition of angiogenesis or diseases involving angiogenesis when administered in effective amounts to patients or animal models in need of such administration.

EXPERIMENTAL

[0120] FGFR-Ig Fc fusion protein dimer (FGFR ECD Ig), according to Construct #1 (Example 6) was characterized with respect to bFGF binding in vitro (FIG. 1) and in whole cells (FIG. 2). Determined IC50 values indicate that the dimerized receptor is an effective antagonist of FGF binding at subnanomolar concentrations, and is 20-50 fold more potent than the monomeric form (FGFR ECD mono) in competing for bFGF binding.

Example 1

In Vitro Binding of 125I-bFGF to FGFR (FIG. 1)

[0121] FGFR monomer (U.S. Pat. No. 5,229,501) and FGFR-Ig Fc fusion dimer according to Construct #1 (Example 6) were tested for their ability to compete with 125I-bFGF for binding to immobilized FGF receptors. Immulon 2 microtiter plates were coated with 10 nM FGFR-IgFc fusion protein (Construct #1) in 50 mM Na2CO3, pH 9.6 overnight at 4° C. The plates were then blocked with 1% gelatin in Dulbecco's PBS (DPBS) for 1 hr at room temperature, and washed with DPBS plus 0.05% Tween 20. Serial dilutions in duplicate of FGFR-Ig fusion dimer (Construct #1, Example 6), FGFR monomer, or bFGF were added, followed by a subsaturating amount (0.5 nM final) 125I-bFGF in DPBS plus 0.1% gelatin, 0.1% triton X-100, and 10 μM Heparin, pH 7.75. After 2 hr at room temperature, wells were washed and the bound radioactivity determined by gamma counting. Displacement curves were analyzed with a four parameter fit to obtain IC50 values. The results indicate that the FGFR-Ig Fc fusion dimer is active as a FGF antagonist at subnanomolar concentrations, and is approximately 20-fold more potent than the FGFR monomer protein as a competitor of bFGF binding to immobilized FGF receptors.

Example 2

Binding of 125I-bFGF to FGFR: Whole Cells (FIG. 2)

[0122] FGFR monomer (U.S. Pat. No. 5,229,501) and FGFR-Ig fusion dimer according to Construct #1 (Example 6) were tested for their ability to compete with 125I-bFGF for binding to stable HEK293 cell lines overexpressing FGFR1. FGFR was overexpressed (0.3×106 receptors/cell) in HEK293 cells by transfection of the FGFR1 cDNA (U.S. Pat. No. 5,229,501) in the high copy number plasmid pcDNA3 and selecting clones resistant to G418. 1.5×105 cells were plated in 24 well plates in DMEM plus 10% serum. Following overnight incubation, the cells were washed twice with 1 ml DMEM plus 0.2% gelatin and 15 units/ml heparin. Serial dilutions of FGFR extracellular domain-Fg Fc fusion dimer, FGFR monomer, or bFGF were pre-mixed with 0.1 nM 125I-bFGF (1138 Ci/mmol) and 250 μl of this mixture was added to each well, and incubated at 37° C. for 30 minutes. The media was removed, and the cells washed three times with 1 ml DMEM containing 150 mM NaCl, 0.2% gelatin, and 15 units/ml heparin. The cells were lysed in 250 μl 0.1% SDS and lysates counted in a gamma counter. Displacement curves were analyzed with a four parameter fit to obtain IC50 values. In this assay, the FGFR-Ig fusion dimer is approximately 50-fold more potent than the FGFR monomer protein as a competitor of bFGF binding to cells.

Example 3

bFGF/Serum Stimulated HUVEC Proliferation (FIG. 3)

[0123] FGFR-Ig fusion dimer (FGFR ECD Ig), according to Construct #1 (Example 6) was tested for its ability to inhibit proliferation of human umbilical vein endothelial cells (HUVEC) in serum- and bFGF-containing media. HUVEC cells were plated in gelatin coated 96-well plates at a density of 2000 cells/well in 50 μl of EBM (endothelial basal media-Clonetics). The cells were incubated overnight (37° C. 5% CO2), the media was removed and 200 μl of media was added containing 10 ng/ml bFGF+90 μg/ml heparin+either 2% or 5% FBS and +/−0.1 μM FGFR fusion (Construct #1). The FGFR fusion was preincubated with bFGF and heparin for at least 30 minutes prior to addition to cell wells. Proliferation indexes were determined on days 0, 1, 2, 3, 4 and 5 using the WST-1 cell proliferation assay (Boehringer Mannheim), which measures number of viable cells. 100 nM FGFR-Ig fusion dimer completely inhibits bFGF-stimulated HUVEC proliferation, even in the presence of 5% serum.

Example 4

HUVEC (Proliferation: Comparison of FGFR Monomer and FGFR-Ig Dimer FIG. 4)

[0124] FGFR monomer (U.S. Pat. No. 5,229,501) and FGFR-Ig fusion dimer (FGFR ECD Ig), according to Construct #1 (Example 6) were compared for their ability to inhibit HUVEC proliferation at different doses in the presence of 10 ng/ml bFGF+90 μg/ml heparin+5% FBS as described in Example 3 above. The FGFR-Ig fusion dimer is more than 10-fold more potent than the FGFR monomer as an inhibitor of HUVEC proliferation in the presence of serum and bFGF, can inhibit proliferation completely, and can inhibit proliferation even below the level seen in the absence of FGF. These results are consistent with the data from in vitro and whole-cell binding assays of FIGS. 1-3.

[0125]
FIGS. 3 and 4 indicate that not only FGFR-Ig is a more potent inhibitor of this proliferation than monomeric extracellular domains of FGFR, the fusion protein is capable of 100% inhibition of the proliferation.

Example 5

Comparison of FGFR-Ig to Anti-bFGF Mab (FIG. 5)

[0126] The ability of the FGFR-Ig fusion dimer (FGFR ECD Ig), according to Construct #1 (Example 6) to compete with 125I-bFGF for binding to immobilized receptors was compared to the highest affinity neutralizing mouse monoclonal antibody to bFGF which is available (Upstate Biotechnology Inc). The assay was performed as in (A). The overall IC50s for these two protein is similar, but the antibody displays some non-competitive binding behavior, and is unable to completely inhibit 125I-bFGF binding, even at high concentrations. In contrast, the FGFR-Ig fusion protein completely inhibits binding.

Example 6

FGFR Extracellular Domain-Ig Fc Fusion Constructs

[0127] SEQ ID NOS:1-12 set forth nucleotide and amino acid sequences for fusion protein Constructs #1-6 comprising segments of an FGFR extracellular domain fused to the Fc region of an immunoglobulin molecule.

[0128] Construct #1

[0129] The polynucleotide and amino acid sequences of Construct #1 are set forth in SEQ ID NOS:1 and 2, respectively.

[0130] Construct #1 contains, in order from the 5′/NH2-terminus to the 3′/COOH-terminus: human FGFR1 signal peptide (comprised by amino acids 1-21); human FGFR1 extracellular domain (nucleotides 64-1123, amino acids 22-374) which contains the Ig I segment (comprised by nucleotides 163-303, amino acids 55-101), the acid box segment (nucleotides 376-399, amino acids 126-133), the Ig II segment (nucleotides 526-684, amino acids 176-228), and the IIIc variant of Ig III segment (nucleotides 823-1017, amino acids 275-339); a linker sequence with a thrombin cleavage site (nucleotides 1123-1170, amino acids 375-390); and the Fc portion of human IgG1 which includes the hinge region, and domains CH2 and CH3 (nucleotides 1171-1869, amino acids 391-622).

[0131] Construct #2

[0132] The polynucleotide and amino acid sequences of Construct #2 are set forth in SEQ ID NOS:3 and 4, respectively.

[0133] Construct #2 comprises deleting from Construct #1 the Ig I segment plus additional flanking sequences (nucleotides 91-357, amino acids 31-119, as numbered in Construct #1), and part of the linker encompassing the thrombin cleavage site (nucleotides 1123-1146, amino acids 375-382, as numbered in Construct #1).

[0134] Accordingly, Construct #2 comprises from the 5′/NH2-terminus to the 3′/COOH-terminus: human FGFR1 signal peptide, human FGFR1 extracellular domain which contains the acid box segment, the Ig II segment and the IIIc variant of Ig III segment; and the Fc portion of human IgG1 which includes the hinge region, and domains CH2 and CH3.

[0135] Construct #3

[0136] The polynucleotide and amino acid sequences of Construct #3 are set forth in SEQ ID NOS:5 and 6, respectively.

[0137] Construct #3 comprises deleting from Construct #1 the Ig I segment as well as the acid box and flanking sequences (nucleotides 91-441, amino acids 31-147, as numbered in Construct #1), and part of the linker encompassing the thrombin cleavage site (nucleotides 1123-1146, amino acids 375-382, as numbered in Construct #1).

[0138] Accordingly, Construct #3 comprises from the 5′/NH2-terminus to the 3′/COOH-terminus: human FGFR1 signal peptide, human FGFR1 extracellular domain which contains the Ig II segment and the IIIc variant of the Ig III segment; and the Fc portion of human IgG1 which includes the hinge region, and domains CH2 and CH3.

[0139] Construct #4

[0140] The polynucleotide and amino acid sequences of Construct #4 are set forth in SEQ ID NOS:7 and 8, respectively.

[0141] Construct #4 is the same as Construct #2 with two additional changes:

[0142] a) Nucleotides 937 to 938 were changed from “CT” to “GA”, which changes amino acid 313 from LEU to GLU (as numbered in Constructs #2 and 4). This mutation decreases the affinity of the Fc portion for Fc receptors.

[0143] b) Nucleotide 1225 was changed from “C” to “T”, which changes amino acid 409 from PRO to SER (as numbered in sequences #2 and 4). This mutation decreases the affinity of the Fc portion for complement.

[0144] The positions correspond to amino acids 235 (LEU to GLU) and 331 (Pro to Ser) of human IgG1.

[0145] Construct #5

[0146] The polynucleotide and amino acid sequences of Construct #5 are set forth in SEQ ID NOS:9 and 10, respectively.

[0147] Construct #5 is the same as Construct #3 with two additional changes:

[0148] a) Nucleotides 853 to 854 have been changed from “CT” to “GA”, which changes amino acid 285 from LEU to GLU (as numbered in sequences #3 and 5).

[0149] b) Nucleotide 1141 has been changed from “C” to “T”, which changes amino acid 385 from PRO to SER.

[0150] The positions changed correspond to amino acids 235 (LEU to GLU) and 331 (Pro to Ser) of human IgG1, and result in decreased affinities of the Fc portion for Fc receptors and complement respectively.

[0151] Construct #6

[0152] The polynucleotide and amino acid sequences of Construct #6 are set forth in SEQ ID NOS:11 and 12, respectively.

[0153] Construct #6 is the same as Construct #5 with one change: Nucleotides 772-798 (amino acids 258-266), as numbered in Construct #5, were deleted. Thus, this construct lacks all the residues from the linker sequence encompassing the thrombin cleavage site as described in Construct #1, and is potentially the least immunogenic construct.

[0154] The purpose of this construct was to eliminate all residual amino acids left over from the original linker segment in Construct #1. Construct #6 has the same activity as the other constructs and is the preferred construct for administration as a therapeutic, because it is potentially the least immunogenic construct.

[0155] Other modifications and embodiments of the invention will come to mind in one skilled in the art to which this invention pertains having the benefit of the teachings presented herein. Therefore, it is to be understood that the invention is not to be limited to the specific embodiments disclosed. Although specific terms are employed, they are used in generic and descriptive sense only and not for purposes of limitation, and that modifications and embodiments are intended to be included within the scope of the appended claims.

	Number	Date	Country
Parent	09499846	Feb 2000	US
Child	10683255	Oct 2003	US

Fibroblast growth factor

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

US Classifications

International Classifications

Abstract

Description

Claims

CROSS-REFERENCE TO RELATED APPLICATION

Provisional Applications (1)

Divisions (1)