Vascular endothelial growth factor receptor targeting peptide-elastin fusion polypeptides

Abstract

Disclosed is a fusion polypeptide for inhibiting neovascularization, including a peptide specifically binding to vascular endothelial growth factor (VEGF) receptors, and a hydrophilic elastin-based polypeptide (hydrophilic EBP) linked to the peptide.

Claims

1. A fusion polypeptide for inhibiting neovascularization, comprising: a peptide of SEQ ID NO: 38 specifically binding to vascular endothelial growth factor (VEGF) receptors; and a hydrophilic elastin-based polypeptide (hydrophilic EBP) linked to the peptide, wherein the fusion polypeptide further comprises a hydrophobic elastin-based polypeptide (hydrophobic EBP) linked to the hydrophilic EBP, and the hydrophobic EBP is consisting of an amino acid sequence represented by Formula 1 or 2 below: TABLE-US-00010 Formula 1 [SEQ ID NO: 1] n; or   Formula 2 [SEQ ID NO: 2] n, wherein   SEQ ID NO: 1 is consisting of [VPGXG VPGXG VPGXG VPGXG VPGXG VPGXG];   SEQ ID NO: 2 is consisting of [VPAXG VPAXG VPAXG VPAXG VPAXG VPAXG]; n is an integer of 1 or more, and represents the number of repeats of SEQ ID NO: 1 or SEQ ID NO: 2; and X is an amino acid other than proline, is selected from any natural or artificial amino acids when the pentapeptide VPGXG or VPAXG is repeated, and at least one of X is a hydrophobic or aliphatic amino acid.

2. The fusion polypeptide according to claim 1, wherein the hydrophobic EBP is consisting of an amino acid sequence represented by Formula 1 or 2: in Formula 1, n is 1, and each X of the pentapeptide repeats is consisting of G (Gly), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 28], or in Formula 2, n is 1, and each X of the pentapeptide repeats is consisting of, G (Gly), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 29]; K (Lys), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 30]; D (Asp), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 31]; K (Lys) and F (Phe) in a ratio of 3:3 [SEQ ID NO: 32]; D (Asp) and F (Phe) in a ratio of 3:3 [SEQ ID NO: 33]; H (His), A (Ala), and I (Ile) in a ratio of 3:2:1 [SEQ ID NO: 34]; H (His) and G (Gly) in a ratio of 5:1 [SEQ ID NO. 35]; or G (Gly), C (Cys), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 36].

3. The fusion polypeptide according to claim 1, wherein the hydrophobic EBP is consisting of an amino acid sequence represented by Formula 2: in Formula 2, n is 12, and each X of the pentapeptide repeats is consisting of G (Gly), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 46], or in Formula 2, n is 24, and each X of the pentapeptide repeats is consisting of G (Gly), A (Ala), and F (Phe) in a ratio of 1:3:2 [SEQ ID NO: 47].

4. The fusion polypeptide according to claim 1, wherein the fusion polypeptide is consisting of an amino acid sequence corresponding to SEQ ID NO: 52 or SEQ ID NO: 53.

5. The fusion polypeptide according to claim 1, wherein the fusion polypeptide forms a self-assembled nanostructure having a core-shell structure, when the hydrophobic EBP forms a core structure and the hydrophilic EBP and the VEGF receptor-specific peptide form a shell structure by a temperature stimulus ranging from 18 to 50° C.

6. The fusion polypeptide according to claim 5, wherein the self-assembled nanostructure comprises a multivalent VEGF receptor-specific peptide as a shell.

7. A composition for treating diseases caused by neovascularization, comprising the fusion polypeptide of claim 1, wherein the fusion polypeptide forms a self-assembled nanostructure having a core-shell structure, when a hydrophobic EBP forms a core structure and a hydrophilic EBP and a VEGF receptor-specific peptide form a shell structure by a temperature stimulus ranging from 18 to 50° C., and the self-assembled nanostructure comprises a multivalent VEGF receptor-specific peptide as a shell, whereby binding affinity between the self-assembled nanostructure and a VEGF receptor increases, and VEGF fails to bind to the VEGF receptor, thereby inhibiting neovascularization.

8. The composition according to claim 7, wherein the diseases caused by neovascularization is any one or more selected from the group comprising diabetic retinopathy, retinopathy of prematurity, macular degeneration, choroidal neovascularization, neovascular glaucoma, eye diseases caused by corneal neovascularization, corneal transplant rejection, corneal edema, corneal opacity, cancer, hemangioma, hemangiofibroma, rheumatoid arthritis, and psoriasis.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The above and other objects, features and advantages of the present invention will become more apparent to those of ordinary skill in the art by describing in detail exemplary embodiments thereof with reference to the accompanying drawings, in which:

(2) FIG. 1 illustrates a schematic diagram and an adapter sequence for construction of plasmids encoding EBP gene libraries with different DNA sizes. (A) an adapter sequence for modification of a pET-21a plasmid, (B) a scheme for modification of a pET-21a plasmid for seamless gene cloning, (C) a scheme for inserting a monomer EBP gene into a modified pET-21a vector, and (D) a scheme for construction of plasmids encoding EBP gene libraries with different DNA sizes;

(3) FIG. 2 shows the agarose gel electrophoresis images of EBP gene libraries used in the present invention. (A) EBPE[A.sub.1G.sub.4I.sub.1], (B) EBPP[A.sub.1G.sub.4I.sub.1], (C) EBPE[K.sub.1G.sub.4I.sub.1], (D) EBPP[K.sub.1G.sub.4I.sub.1], (E) EBPE[D.sub.1G.sub.4I.sub.1], (F) EBPP[D.sub.1G.sub.4I.sub.1], (G) EBPE[E.sub.1G.sub.4I.sub.1], (H) EBPP[E.sub.1G.sub.4I.sub.1], (I) EBPP[G.sub.1A.sub.3F.sub.2], (J) EBPP[K.sub.1A.sub.3F.sub.2], (K) EBPP[D.sub.1A.sub.3F.sub.2], and (L) EBPP[H.sub.3A.sub.2I.sub.1]. The number of EBP repeat units was indicated below each DNA band. Two side-lanes on all agarose gels represent different DNA size markers (0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 1.0, 1.5, 2.0, and 3.0 kbp, from bottom to top);

(4) FIG. 3 shows the copper-stained SDS-PAGE gel (4 to 20% gradient) images of EBPs used in the present invention. (A) EBPE[A.sub.1G.sub.4I.sub.1], (B) EBPP[A.sub.1G.sub.4I.sub.1], (C) EBPE[K.sub.1G.sub.4I.sub.1], (D) EBPP[K.sub.1G.sub.4I.sub.1], (E) EBPE[D.sub.1G.sub.4I.sub.1] and (F) EBPP[D.sub.1G.sub.4I.sub.1]. Two side-lanes on SDS-PAGE gels contain standard protein size markers (7, 15, 24, 35, 40, 50, 65, 90, 110, and 150 kDa, from bottom to top);

(5) FIGS. 4A to 4F show the thermal profiles of EBPs used in the present invention. FIG. 4A shows the profiles of EBPE[A.sub.1G.sub.4I.sub.1].sub.n, FIG. 4B shows the profiles of EBPP[A.sub.1G.sub.4I.sub.1].sub.n, FIG. 4C shows the profiles of EBPE[K.sub.1G.sub.4I.sub.1].sub.n, FIG. 4D shows the profiles of EBPP[K.sub.1G.sub.4I.sub.1].sub.n, FIG. 4E shows the profiles of EBPP[D.sub.1G.sub.4I.sub.1].sub.n, and FIG. 4F shows the profiles of EBPP[G.sub.1A.sub.3F.sub.2].sub.n. To obtain thermal profiles, 25 μM EBP solutions were prepared in PBS buffer or PBS buffer supplemented with 1 to 3 M sodium chloride, and the optical absorbance of the EBP solution was measured at 350 nm while heating the solution at a heating rate of 1° C./min;

(6) FIGS. 5A to 5D are schematic diagrams of cloning, molecular structures, and functions of fusion polypeptides composed of EBPPs and an anti-Flt1 peptide. In FIG. 5A genes encoding EBPP diblocks were constructed, and a gene encoding an anti-Flt1 peptide was cloned into a plasmid including a gene encoding an EBPP diblock. In FIG. 5B fusion polypeptides composed of anti-Flt1 peptide-hydrophilic EBP; and anti-Flt1 peptide-hydrophilic EBP-hydrophobic EBP. In FIG. 5C fusion polypeptides composed of anti-Flt1 peptide-hydrophilic EBP-hydrophobic EBP were able to form a micellar structure by a temperature stimulus. In FIG. 5D (i) fusion polypeptides composed of anti-Flt1 peptide-hydrophilic EBP were able to bind to VEGFRs, and were able to inhibit interactions between VEGFR1 and VEGF. (ii) The micellar structures of fusion polypeptides composed of anti-Flt1 peptide-hydrophilic EBP-hydrophobic EBP were able to bind to VEGFRs, and was able to inhibit interactions between VEGFRs and VEGF with increased affinity due to the multivalency of the anti-Flt1 peptide;

(7) FIG. 6 shows (A) agarose gel (1%) images and (B) SDS-PAGE (4 to 20% gradient) gel images. (a) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, (b) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.6, (c) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and (d) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24;

(8) FIG. 7 shows the LCST of EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) as turbidity profiles. Turbidity profiles were determined by measuring the absorbance of (A to C) 25 μM EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) and (D to F) 25 μM anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer). The absorbance was measured at 350 nm in 10 mM PBS (A and D), 10 mM PBS supplemented with 1 M sodium chloride (B and E), and 10 mM PBS supplemented with 2 M sodium chloride (C and F), while heating samples at a heating rate of 1° C./min;

(9) FIG. 8 shows the (A) agarose gel (1%) images and the (B) SDS-PAGE (4 to 20% gradient) gel images of fusion polypeptides. (A) A modified pET21-a (+) plasmid encoding anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 or anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 was digested by XbaI and BseRI. (B) The fusion polypeptides were expressed in E. coli and purified by ITC. 4 to 20% gradient gels were visualized with copper stain. An expected molecular weight was indicated below the band;

(10) FIG. 9 shows the turbidity profiles of EBPP diblocks depending on the presence or absence of an anti-Flt1 peptide. (A) The turbidity profiles of (a) 25 μM EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and (b) 25 μM EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24. (B) The turbidity profiles of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and (C) the turbidity profiles of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 were obtained for concentrations of 12.5, 25, 50 and 100 μM in 10 mM PBS. Absorbance was measured at 350 nm while heating the samples at a rate of 1° C./min. A phase transition occurred twice. The first phase transition occurred as a result of hydrophobic block aggregation, the second phase transition was affected by polar EBPP[E.sub.1G.sub.4I.sub.1].sub.12. As EBPP diblock concentration increased, the first and second T.sub.t values thereof were lowered;

(11) FIG. 10 shows the hydrodynamic radius of (a) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and (b) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24, and the hydrodynamic radius was measured by a DLS instrument. The hydrodynamic radius of EBPP diblock polypeptides was measured at 25 μM in 10 mM PBS. The hydrodynamic radius of EBPP diblock polypeptides prior to the first phase transition is less than 10 nm, indicating that the polypeptides are present in a soluble unimer form;

(12) FIG. 11 shows the in vitro biological activities of EBPP[A.sub.1G.sub.4I.sub.1].sub.12, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.6, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24, which inhibit VEGFR1 binding to coated VEGF;

(13) FIG. 12 shows the in vitro biological activities of EBPP[A.sub.1G.sub.4I.sub.1].sub.12, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12, anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24, which inhibit VEGFR binding to coated VEGF. EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 were unimers at 37° C. On the other hand, anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 formed metastable micelles at 37° C., and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 formed stable micelles;

(14) FIG. 13 shows the results of the in vitro tube formation assay of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12. (A) the fluorescence microscope images of calcein-AM-labeled HUVECs and (B) the degree of inhibition of tube formation. The degree of inhibition of tube formation was quantified from the images of (A). The tube length of HUVECs treated with anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 decreased with increasing the concentration of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1]. The anti-Flt1-EBPP [A.sub.1G.sub.4I.sub.1].sub.12 inhibited migration and tube formation of HUVECs. Tubing lengths are average values±SE. *P≤0.05 by a t test; and

(15) FIG. 14 shows an in vivo inhibition effect of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 in a laser-induced choroidal neovascularization model. C57BL6 mice (n=3 per group) were treated with a vehicle (PBS), EBPP[A.sub.1G.sub.4I.sub.1].sub.12 (20 μg) or anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 (0.1, 1, 5 and 20 μg) after laser-induced injury, and the treatment was continued for 5 days. At day 14 after laser injury, mice were euthanized and fluorescein isothiocyanate (FITC)-dextran perfused whole choroidal flat-mounts were prepared. The CNV lesion size was quantified by Nano-Zoomer and FISH. (A) representative flat mount fluorescence microscopic images. (B) a graph of the CNV size of each treated group. Each point corresponds to a CNV lesion, and a horizontal bar corresponds to the average value of each group. *P≤0.05 by an unpaired t test. The data represents two independent experiments.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

Example 1: Materials

(16) A pET-21a (+) vector and BL21 (DE3) E. Coli cells were obtained from Novagen Inc. (Madison, Wis., U.S.). Top10 competent cells and calcein-AM were purchased from Invitrogen (Carlsbad, Calif., U.S.) and HUVECs were purchased from American Type Culture Collection (ATCC) (Virginia, U.S.). All customized oligonucleotides were synthesized by Cosmo GeneTech (Seoul, South Korea) and recombinant human VEGF-165 (rhVEGF.sub.165) was obtained from Sino Biological Inc. (Beijing, China). Calf intestinal alkaline phosphatase (CIP), BamHI and XbaI were obtained from Fermentas (Ontario, Canada). AcuI and BseRI were purchased from New England Biolabs (Ipswich, Mass., U.S.). T4 DNA ligase was obtained from Elpis Bio-tech (Taejeon, South Korea). DNA miniprep, gel extraction, and PCR purification kits were obtained from Geneall Biotechnology (Seoul, South Korea). “Dyne Agarose High” was obtained from DYNE BIO, Inc. (Seongnam, South Korea). Top10 cells were grown in “TB DRY” media obtained from MO BIO Laboratories, Inc. (Carlsbad. Calif., U.S.). BL21 (DE3) cells were grown in “CircleGrow” media obtained from MP Biomedicals (Solon, Ohio, U.S.). “Ready Gels, Tris-HCl 2-20% precast gels” were from Bio-Rad (Hercules, Calif., U.S.). Phosphate buffered saline (PBS, pH 7.4), kanamycin, polyethyleneamine (PEI), FITC-dextran, formalin and bovine serum albumin (BSA) were obtained from Sigma-Aldrich (St Louis, Mo., U.S.). Matrigel was purchased from BD Biosciences (San Diego, Calif., U.S.). Avastin, also known as bevacizumab was purchased from Roche Pharma Ltd. (Reinach, Switzerland). Ketamine was obtained from Huons (Seongnam, South Korea). Xylazine was purchased from BAYER (Leverkusen, Germany). Tropicamide was purchased from Santen Pharmaceutical Co. Ltd (Kita-ku, Osaka, Japan). A stereomicroscope was obtained from Leica (Wetzlar, Germany). Recombinant human VEGF.sub.165 protein and recombinant human VEGF R1/Flt-1 F.sub.c were purchased from R&D System (Minneapolis, Minn., U.S.). Rabbit anti-human IgG F.sub.c-HRP chimeric protein and 3,3′,5,5′-tetramethylbenzidine (TMB) was obtained from ThermoFisher (Massachusetts, U.S.).

Example 2: Notation for Different EBP Blocks and Block Polypeptides Thereof

(17) Different EBPs having a pentapeptide repeat unit of Val-Pro-(Gly or Ala)-X.sub.aa-Gly[VP (G or A)XG] are named as follows. X.sub.aa may be any amino acid except Pro. First, pentapeptide repeats of Val-Pro-Ala-X.sub.aa-Gly (VPAXG) with plasticity are defined as an elastin-based polypeptide with plasticity (EBPP). On the other hand, pentapeptide repeats of Val-Pro-Gly-X.sub.aa-Gly (VPGXG) are called elastin-based polypeptides with elasticity (EBPEs). Second, in [X.sub.iY.sub.jZ.sub.k].sub.n, the capital letters in the parentheses represent the single letter amino acid codes of guest residues, i.e., amino acids at the fourth position (X.sub.aa or X) of an EBP pentapeptide, and subscripts corresponding to the capital letters indicate the ratio of the guest residues in an EBP monomer gene as a repeat unit. The subscript number n of [X.sub.iY.sub.jZ.sub.k].sub.n represents the total number of repeats of an EBP corresponding to SEQ ID NO. 1 [VPGXG VPGXG VPGXG VPGXG VPGXG VPGXG] or SEQ ID NO. 2[VPAXG VPAXG VPAXG VPAXG VPAXG VPAXG] according to the present invention. For example, EBPP[G.sub.1A.sub.3F.sub.2].sub.12 is an EBPP block including 12 repeats of a pentapeptide unit, SEQ ID NO. 2 [VPAXG VPAXG VPAXG VPAXG VPAXG VPAXG], in which a ratio of Gly, Ala, and Phe at the fourth guest residue position (X.sub.aa) is 1:3:2. Finally, EBP-EBP diblock polypeptides are named according to the composition of each block in brackets with a hyphen between blocks as in EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12.

Example 3: Preparation of Modified pET-21a Vector for Cloning Seamless Gene

(18) 4 μg of a pET-21a vector was digested and dephosphorylated with 50 U of XbaI, 50 U of BamHI and 10 U of a thermosensitive alkaline phosphatase in FastDigest buffer for 20 minutes at 37° C. The digested plasmid DNA was purified using a PCR purification kit, and then was eluted in 40 μl of distilled and deionized water. Two oligonucleotides with XbaI and BamHI compatible sticky ends were designed, i.e., SEQ ID NO. 39 (5′-ctagaaataattttgtttaactttaagaaggaggagtacatatgggctactgataatgatcttcag-3′) and SEQ ID NO. 40 (5′-gatcctgaagatcattatcagtagcccatatgtactcctccttcttaaagttaaacaaaattattt-3′). To anneal the two types of oligonucleotides, each oligonucleotide was prepared at a concentration of 2 μM in 50 μl of T4 DNA ligase buffer, heat treated at 95° C. for 2 minutes and then slowly cooled to room temperature over 3 hours. To ligate the annealed dsDNA, i.e., a DNA insert, into multiple cloning sites within the linearized pET-21a vector, 20 pmol of the annealed dsDNA and 0.1 pmol of the linearized pET-21a vector were incubated in T4 DNA ligase buffer containing 1 U of T4 DNA ligase for 30 minutes at 37° C. The modified pET-21a (mpET-21a) vector for cloning and expressing a seamless gene was transformed into Top10 competent cells, followed by plating the Top10 competent cells on a super optimal broth with catabolite repression (SOC) plate supplemented with 50 μg/ml ampicillin. The DNA sequence of the mpET-21a vector was then verified by fluorescent dye terminator DNA sequencing (Applied Biosystems Automatic DNA Sequencer ABI3730).

Example 4: Synthesis of EBP Monomer Gene and Oligomerization Thereof

(19) EBP sequences having a pentapeptide repeat unit, Val-Pro-(Gly or Ala)-X.sub.aa-Gly, in which the fourth residues were varied in different molar ratios, were designed at the DNA level to optimize T.sub.t below a physiological temperature. The DNA and amino acid sequences of EBPs with various pentapeptide repeat units for 17 EBP libraries are shown in Tables 1 and 2, respectively.

(20) TABLE-US-00003 TABLE 1 Gene sequences corresponding to EBP libraries. Both EBPs with plasticity (EBPPs) having a pentapeptide repeat of  Val-Pro-Ala-X.sub.aa-Gly, and EBPs with elasticity (EBPEs) having a pentapeptide repeat of Val-Pro-Gly-X.sub.aa-Gly were cloned to have the same guest residue composition and ratio. SEQ ID EBP Gene Sequence NO. EBPE[A.sub.1G.sub.4I.sub.1] GTC CCA GGT GGA GGT GTA CCC GGC GCG GGT GTC CCA GGT GGA GGT 3 GTA CCT GGG GGT GGG GTC CCT GGT ATT GGC GTA CCT GGA GGC GGC EBPP[A.sub.1G.sub.4I.sub.1] GTT CCA GCT GGC GGT GTA CCT GCT GCT GCT GTT CCG GCC GGT GGT 4 GTT CCG GCG GGC GGC GTG CCT GCA ATA GGA GTT CCC GCT GGT GGC EBPE[K.sub.1G.sub.4I.sub.1] GTT CCG GGT GGT GGT GTT CCG GGT AAA GGT GTT CCG GGT GGT GGT 5 GTT CCG GGT GGT GGT GGT GTT CCG GGT ATC GGT GTT CCG GGT GGC EBPP[K.sub.1G.sub.4I.sub.1] GTT CCG GCG GGT GGT GTT CCG GCG AAA GGT GTT CCG GCG GGT GGT 6 GTT CCG GCG GGT GGT GTT CCG GCG ATC GGT GTT CCG GCG GGT GGC EBPE[D.sub.1G.sub.4I.sub.1] GTT CCG GGT GGT GGT GTT CCG GGT GAT GGT GTT CCG GGT GGT GGT 7 GTT CCG GGT GGT GGT GGT GTT CCG GGT ATC GGT GTT CCG GGT GGC EBPP[D.sub.1G.sub.4I.sub.1] GTT CCG GCG GGT GGT GTT CCG GCG GAT GGT GTT CCG GCG GGT GGT 8 GTT CCG GCG GGT GGT GTT CCG GCG ATC GGT GTT CCG GCG GGT GGC EBPE[E.sub.1G.sub.4I.sub.1] GTT CCG GGT GGT GGT GTT CCG GGT GAA GGT GTT CCG GGT GGT GGT 9 GTT CCG GGT GGT GGT GGT GTT CCG GGT ATC GGT GTT CCG GGT GGC EBPP[E.sub.1G.sub.4I.sub.1] GTT CCG GCG GGT GGT GTT CCG GCG GAA GGT GTT CCG GCG GGT GGT 10 GTT CCG GCG GGT GGT GTT CCG GCG ATC GGT GTT CCG GCG GGT GGC EBPE[G.sub.1A.sub.3F.sub.2] GTC CCG GGT GCG GGC GTG CCG GGA TTT GGA GTT CCG GGT GCG GGT 11 GTT CCA GGC GGT GGT GTT CCG GGC GCG GGC GTG CCG GGC TTT GGC EBPP[G.sub.1A.sub.3F.sub.2] GTG CCG GCG GCG GGC GTT CCA GCC TTT GGT GTG CCA GCG GCG GGA 12 GTT CCG GCC GGT GGC GTG CCG GCA GCG GGC GTG CCG GCT TTT GGC EBPP[K.sub.1A.sub.3F.sub.2] GTG CCG GCG GCG GGC GTT CCA GCC TTT GGT GTG CCA GCG GCG GGA 13 GTT CCG GCC AAA GGC GTG CCG GCA GCG GGC GTG CCG GCT TTT GGC EBPP[D.sub.1A.sub.3F.sub.2] GTG CCG GCG GCG GGC GTT CCA GCC TTT GGT GTG CCA GCG GCG GGA 14 GTT CCG GCC GAT GGC GTG CCG GCA GCG GGC GTG CCG GCT TTT GGC EBPP[K.sub.3F.sub.3] GTT CCA GCG TTT GGC GTG CCA GCG AAA GGT GTT CCG GCG TTT GGG 15 GTT CCC GCG AAA GGT GTG CCG GCC TTT GGT GTG CCG GCC AAA GGC EBPP[D.sub.3F.sub.3] GTT CCA GCG TTT GGC GTG CCA GCG GAT GGT GTT CCG GCG TTT GGG 16 GTT CCC GCG GAT GGT GTG CCG GCC TTT GGT GTG CCG GCC GAT GGC EBPP[H.sub.3A.sub.3I.sub.1] GTG CCG GCG CAT GGA GTT CCT GCC GCC GGT GTT CCT GCG CAT GGT 17 GTA CCG GCA ATT GGC GTT CCG GCA CAT GGT GTG CCG GCC GCC GGC EBPP[H.sub.5G.sub.1] GTT CCG GCC GGA GGT GTA CCG GCG CAT GGT GTT CCG GCA CAT GGT 18 GTG CCG GCT CAC GGT GTG CCT GCG CAT GGC GTT CCT GCG CAT GGC EBPP[G.sub.1C.sub.3F.sub.2] GTG CCG GCG TGC GGC GTT CCA GCC TTT GGT GTG CCA GCG TGC GGA 19 GTT CCG GCC GGT GGC GTG CCG GCA TGC GGC GTG CCG GCT TTT GGC

(21) TABLE-US-00004 TABLE 2 Amino acid sequences corresponding to EBP libraries SEQ ID EBP Amino acid Sequence NO. EBPE[A.sub.1G.sub.4I.sub.1] VPGGG VPGAG VPGGG VPGGG VPGIG VPGGG 20 EBPP[A.sub.1G.sub.4I.sub.1] VPAGG VPAAG VPAGG VPAGG VPAIG VPAGG 21 EBPE[K.sub.1G.sub.4I.sub.1] VPGGG VPGKG VPGGG VPGGG VPGIG VPGGG 22 EBPP[K.sub.1G.sub.4I.sub.1] VPAGG VPAKG VPAGG VPAGG VPAIG VPAGG 23 EBPE[D.sub.1G.sub.4I.sub.1] VPGGG VPGDG VPGGG VPGGG VPGIG VPGGG 24 EBPP[D.sub.1G.sub.4I.sub.1] VPAGG VPADG VPAGG VPAGG VPAIG VPAGG 25 EBPE[E.sub.1G.sub.4I.sub.1] VPGGG VPGEG VPGGG VPGGG VPGIG VPGGG 26 EBPP[E.sub.1G.sub.4I.sub.1] VPAGG VPAEG VPAGG VPAGG VPAIG VPAGG 27 EBPE[G.sub.1A.sub.3F.sub.2] VPGAG VPGFG VPGAG VPGGG VPGAG VPGFG 28 EBPP[G.sub.1A.sub.3F.sub.2] VPAAG VPAFG VPAAG VPAGG VPAAG VPAFG 29 EBPP[K.sub.1A.sub.3F.sub.2] VPAAG VPAFG VPAAG VPAGG VPAAG VPAFG 30 EBPP[D.sub.1A.sub.3F.sub.2] VPAAG VPAFG VPAAG VPAGG VPAAG VPAFG 31 EBPP[K.sub.3F.sub.3] VPAFG VPAKG VPAFG VPAKG VPAFG VPAKG 32 EBPP[D.sub.3F.sub.3] VPAFG VPADG VPAFG VPADG VPAFG VPADG 33 EBPP[H.sub.3A.sub.3I.sub.1] VPAHG VPAAG VPAHG VPAIG VPAHG VPAAG 34 EBPP[H.sub.5G.sub.1] VPAGG VPAHG VPAHG VPAHG VPAHG VPAHG 35 EBPP[G.sub.1C.sub.3F.sub.2] VPACG VPAFG VPACG VPAGG VPACG VPAFG 36

(22) In Table 1, SEQ ID NO. 3 to 10 may be classified as gene sequences for hydrophilic EBP blocks, and SEQ ID NO. 11 to 19 may be classified as gene sequences for hydrophobic EBP blocks, in which Phe and His are incorporated. In Table 2, amino acid SEQ ID NO. 20 to 27 may be classified as hydrophilic EBP blocks, and amino acid SEQ ID NO. 28 to 36, in which Phe and His are incorporated, may be classified as hydrophobic EBP blocks. In particular, in Table 2, SEQ ID NO. 22 and 23 are classified as positively charged hydrophilic EBP blocks, and SEQ ID NO. 24 to 27 are classified as negatively charged hydrophilic EBP blocks. That is, as described above, when the LCST of an EBP is lower than the body temperature, the EBP exhibits hydrophobicity, and when the LCST of an EBP is higher than the body temperature, the EBP exhibits hydrophilicity. Due to this nature of EBPs, the hydrophilic and hydrophobic properties of EBPs may be relatively defined when EBPs are applied to biotechnology.

(23) Different EBPs having a pentapeptide repeat unit, Val-Pro-(Gly or Ala)-X.sub.aa-Gly [where X.sub.aa may be any amino acid except Pro], which are capable of responding to unique stimuli including temperature and pH, were designed at the DNA level. EBPs with plasticity (EBPPs) having a pentapeptide repeat unit of Val-Pro-Ala-X.sub.aa-Gly and EBPs with elasticity (EBPEs) having a pentapeptide repeat unit of Val-Pro-Gly-X.sub.aa-Gly were all cloned to have the same guest residue composition and ratio. Tables 1 and 2 represent the gene and amino acid sequences of different EBPs having respective pentapeptide units. For example, EBPE[G.sub.1A.sub.3F.sub.2].sub.12 and EBPP[G.sub.1A.sub.3F.sub.2].sub.12 not only show almost the same molar mass, but also the fourth residues of these EBP pentapeptide units represent the same combination. In addition, these EBP blocks have different mechanical properties because the third amino acid residues (Ala or Gly) of the pentapeptide units are different. Positively and negatively charged EBPs were prepared by introducing charged amino acids such as Lys, Asp, GIu, and His as guest residues.

(24) To anneal each pair of oligonucleotides encoding various EBPs, each oligonucleotide was prepared at a concentration of 2 μM in 50 μl of T4 DNA ligase buffer, heat treated at 95° C. for 2 minutes and then slowly cooled to room temperature over 3 hours. 4 μg of a modified pET-21a vector was digested and dephosphorylated with 15 U of BseRI and 10 U of FastAP thermosensitive alkaline phosphatase for 30 minutes at 37° C. The digested plasmid DNA was purified using a PCR purification kit, and then was eluted in 40 μl of distilled and deionized water. To ligate the annealed dsDNA, i.e., a DNA insert, into multiple cloning sites within the linearized mpET-21a vector, 90 pmol of the annealed dsDNA and 30 pmol of the linearized mpET-21a vector were incubated in T4 DNA ligase buffer containing 1 U of T4 DNA ligase for 30 minutes at 16° C. The ligated plasmid was transformed into Top10 chemically competent cells, followed by plating the Top10 competent cells on an SOC plate supplemented with 50 μg/ml ampicillin. DNA sequences were then confirmed by DNA sequencing. After all EBP monomer genes were constructed, each EBP gene was synthesized by ligating each of 36 types of repetitive genes (as an insert) into the corresponding vector containing each of the same 36 types of repetitive genes, as follows. A cloning procedure for EBP libraries and fusions thereof are illustrated in FIG. 1. Vectors harboring gene copies corresponding to EBP monomers were digested and dephosphorylated with 10 U of XbaI, 15 U of BseRI and 10 U of FastAP thermosensitive alkaline phosphatase in CutSmart buffer for 30 minutes at 37° C. The digested plasmid DNA was purified using a PCR purification kit, and then was eluted in 40 μl of distilled and deionized water. For preparation of an insert part, a total of 4 μg of an EBP monomer gene was digested with 10 U of XbaI and 15 U of AcuI in CutSmart buffer for 30 minutes at 37° C. After digestion, the reaction product was separated by agarose gel electrophoresis and the insert was purified using a gel extraction kit. Ligation was performed by incubating 90 pmol of the purified insert with 30 pmol of the linearized vector in T4 DNA ligase buffer containing 1 U of T4 DNA ligase for 30 minutes at 16° C. The product was transformed into Top10 chemically competent cells, and then the cells were plated on an SOC plate supplemented with 50 μg/ml ampicillin Transformants were initially screened by diagnostic restriction digestion on an agarose gel and further confirmed by DNA sequencing as described above.

(25) As described above, EBP gene libraries having different DNA sizes were synthesized using the designed plasmid vector and three different restriction endonucleases. FIG. 1 illustrates a recursive directional ligation (RDL) method, in which EBP monomer genes are ligated to form oligomerized EBP genes. For example, a gene construct encoding EBPP[G.sub.1A.sub.3F.sub.2].sub.12 was prepared by ligation, wherein a plasmid backbone and an insert derived from a plasmid-borne gene vector harboring a gene encoding EBPP[G.sub.1A.sub.3F.sub.2].sub.6 were used. The plasmid-borne gene vector harboring a gene encoding EBPP[G.sub.1A.sub.3F.sub.2].sub.6 was double-digested by XbaI and AcuI to obtain an insert, i.e., a gene fragment encoding EBPP[G.sub.1A.sub.3F.sub.2].sub.6. On the other hand, the plasmid-borne gene vector for EBPP[G.sub.1A.sub.3F.sub.2].sub.6 was double-digested by XbaI and BseRI to obtain a plasmid backbone and then the plasmid backbone was dephosphorylated by treatment with an alkaline phosphatase. The RDL method using two different double restriction enzymes has several advantages. First, due to the different shapes of the protrusions of both an insert and a digested vector, self-ligation of the digested vector did not occur, and the insert and the digested vector were efficiently linked in a head-tail orientation. Second, due to the mechanism of type III restriction endonuclease, an additional DNA sequence encoding each linker between blocks is not required. Each EBP gene was oligomerized to generate 36, 72, 108, 144, 180, and 216 EBP pentapeptide repeats. Using two restriction endonucleases XbaI and BamHI, oligomerized genes with sizes of 540, 1080, 1620, 2160, 2700, and 3240 base pairs (bps) were confirmed. As characterized by agarose gel electrophoresis, FIG. 2 depicts the digested DNA bands of EBP libraries with DNA size markers on both end lanes. For example, EBPE[A.sub.1G.sub.4I.sub.1] in FIG. 2(A) clearly shows a digested DNA band corresponding to a DNA region encoding an oligomerized pentapeptide sequence containing Ala, Gly, Ile in a ratio of 1:4:1 as a guest residue. All digested DNA bands are shown as corresponding lengths as compared to the molecular size markers.

(26) EBP genes and block co-polypeptides thereof were overexpressed in E. coli having a T7 promoter and purified by multiple cycles of inverse transition cycling (ITC). FIG. 3 shows copper-stained SDS-PAGE gel images of the purified EBPs. EBPs shifted at least 20% more than theoretically calculated molecular weights. Two side-lanes on SDS-PAGE gels contain standard protein size markers (7, 15, 24, 35, 40, 50, 65, 90, 110, and 150 kDa, from bottom to top). In FIGS. 3(A) and 3(B), EBPE[A.sub.1G.sub.4I.sub.1] and EBPP[A.sub.1G.sub.4I.sub.1] represent a series of corresponding proteins with a molecular weight greater than a theoretical molecular weight (for EBPE[A.sub.1G.sub.4I.sub.1], 14.0, 27.7, 41.3, 55.0, and 68.6 kDa, from left to right). In general, as shown in FIGS. 3(C) and 3(D), positively charged EBP libraries, including EBPE[K.sub.1G.sub.4I.sub.1] and EBPP[K.sub.1G.sub.4I.sub.1], showed higher molecular weights than nonpolar EBP libraries, including EBPE[A.sub.1G.sub.4I.sub.1] and EBPP[A.sub.1G.sub.4I.sub.1]. In addition, as shown in FIGS. 3(E) and 3(F), negatively charged EBP libraries, including EBPE[D.sub.1G.sub.4I.sub.1] and EBPP[D.sub.1G.sub.1I.sub.1], have differently charged characteristics, and thus exhibited higher molecular weights than positively charged EBP libraries.

(27) EBP libraries were characterized. FIGS. 4A to 4F show thermal transition behaviors of EBPs determined by measuring optical absorbance at 350 nm (absorbance.sub.350) at a heating rate of 1° C./min Inverse transition temperature (T.sub.t) is defined as a temperature at which the first derivative (d (OD.sub.350)/dT) of turbidity, which is a function of temperature, was the maximum. Based on environmental conditions such as a salt concentration and pH and the different third and fourth amino acids of an EBP pentapeptide repeat unit, the T.sub.t of an EBP was finely controlled in PBS and PBS was supplemented with 1 to 3 M sodium chloride. For example, EBPE[A.sub.1G.sub.4I.sub.1].sub.12 (FIG. 4A) with Gly at the third amino acid of an EBP pentapeptide repeat exhibited a T.sub.t about 15° C. higher than that of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 (FIG. 4B) with Ala at the third amino acid of an EBP pentapeptide repeat in PBS containing 1 M sodium chloride, because Gly at the third amino acid of an EBP pentapeptide repeat has a higher hydrophilicity than Ala. In general, charged EBP libraries have a higher T.sub.t than nonpolar EBP libraries because charged residues are introduced into the fourth amino acid of the EBP pentapeptide repeat of the charged EBPs. Negatively charged EBP libraries, such as EBPP[D.sub.1G.sub.4I.sub.1] (FIG. 4E), have different pK.sub.a values for Asp and Lys at the fourth amino acid of an EBP pentapeptide repeat, and thus have a higher T.sub.t than positively charged EBP libraries, such as EBPE[K.sub.1G.sub.4I.sub.1] (FIG. 4C) and EBPP[K.sub.1G.sub.4I.sub.1] (FIG. 4D). For reference, FIGS. 4A, 4B, 4C, 4D and 4E exhibit hydrophilicity, and FIG. 4F exhibits EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and EBPP[G.sub.1A.sub.3F.sub.2].sub.24 exhibit hydrophobicity.

Example 5: Gene Construction of Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.]n and Anti-Flt1-EBP Diblock Block (Copolypeptides)

(28) A pair of oligonucleotides encoding an anti-Flt1 peptide acting as a VEGFR1 antagonist were chemically synthesized by Cosmo Genetech (Seoul, Korea), and linked to an oligonucleotide cassette with cohesive ends including restriction sites recognized by AcuI and BseRI. An oligonucleotide cassette encoding the anti-Flt1 peptide was rationally designed to have no restriction sites recognized by BseRI, XbaI, AcuI and BamHI for seamless gene cloning, as shown in Table 3.

(29) TABLE-US-00005 TABLE 3 Gene and amino acid sequences of CPPs SEQ ID NO. Sequence Type Sequence 37 Gene Sequence GGC AAT CAG TGG TTT ATT 38 Amino acid  G N Q W F I Sequence

(30) In Table 4, the sequences, gene lengths and molecular weights of fusion polypeptides with a hydrophilic EBP block or an EBP diblock of hydrophilic EBP block-hydrophobic EBP block are shown.

(31) TABLE-US-00006 TABLE 4 Sequences, gene lengths and molecular weights of fusion polypeptides Nucleotide Fusion protein (SEQ ID NO.) length (bp) M.W (kDa) Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 (SEQ ID NO. 48) 288 8.19 Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.6 (SEQ ID NO. 49) 558 15.27 Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 (SEQ ID NO. 50) 1098 29.42 Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24 (SEQ ID NO. 51) 2178 57.72 Anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 2178 59.90 (SEQ ID NO. 52) Anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 3258 90.00 (SEQ ID NO. 53)

(32) Each plasmid containing an EBP with restriction sites recognized by BseRI, XbaI, AcuI and BamHI, and the oligonucleotide cassette were used to create genes for the fusion polypeptide libraries of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 and anti-Flt1-EBP diblock blocks. First, to anneal a pair of oligonucleotides encoding an anti-Flt1 peptide, each oligonucleotide was prepared at a concentration of 2 μM in 50 μl of T4 DNA ligase buffer, heat treated at 95° C. for 2 minutes and then the reaction solution was slowly cooled to room temperature over 3 hours. To clone the anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n, a plasmid vector encoding EBPP[A.sub.1G.sub.4I.sub.1].sub.3n was digested with 15 U of BseRI in CutSmart buffer for 30 minutes at 37° C. The digested plasmid DNA was purified using a PCR purification kit, and then dephosphorylated with 10 U of FastAP as a thermosensitive alkaline phosphatase in CutSmart buffer for 1 hour at 37° C. The digested and dephosphorylated plasmid DNA was purified using a PCR purification kit, and then eluted in 40 μl of distilled and deionized water. Ligation was performed by incubating 90 pmol of the purified insert and 30 pmol of the linearized vector in T4 DNA ligase buffer containing 1 U of T4 DNA ligase at 16° C. for 30 minutes. The product was transformed into Top10 chemically competent cells and the cells were plated on SOC plates supplemented with 50 μg/ml ampicillin Transformants were initially screened by diagnostic restriction digestion on an agarose gel and further confirmed by DNA sequencing as described above.

(33) Similarly, to clone anti-Flt1-EBP diblock blocks with hydrophobic blocks of different lengths, plasmid vectors encoding EBPP[G.sub.1A.sub.3F.sub.2].sub.n were digested with 10 U of XbaI and 15 U of BseRI in CutSmart buffer for 30 minutes at 37° C. The digested plasmid DNA was purified using a PCR purification kit, and then dephosphorylated with 10 U of FastAP as a thermosensitive alkaline phosphatase in CutSmart buffer for 1 hour at 37° C. The digested and dephosphorylated plasmid DNA was purified using a PCR purification kit, and then eluted in 40 μl of distilled and deionized water. 4 μg of EBPP[E.sub.1G.sub.4I.sub.1].sub.n genes were digested with 10 U of XbaI and 15 U of AcuI in CutSmart buffer for 30 minutes at 37° C. After digestion, the reaction product was separated by agarose gel electrophoresis and an insert was purified using a gel extraction kit. Ligation was performed by incubating 90 pmol of the purified insert and 30 pmol of the linearized vector in T4 DNA ligase buffer containing 1 U of T4 DNA ligase at 16° C. for 30 minutes. The product was transformed into Top10 chemically competent cells and the cells were plated on SOC plates supplemented with 50 μg/ml ampicillin Transformants were initially screened by diagnostic restriction digestion on an agarose gel and further confirmed by DNA sequencing. Plasmid vectors encoding anti-Flt1-EBP diblock blocks were prepared using BseRI, and ligation and confirmation of ligation were performed as described above.

Example 6: Expression of Genes Encoding EBPs, Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.].SUB.3n .and Anti-Flt1-EBP Diblock Block and Purification of Gene Expression Products

(34) E. coli strain BL21 (DE3) cells were transformed with each vector containing an EBP, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n or an anti-Flt1-EBP diblock block, and then inoculated in 50 ml of CircleGrow media supplemented with 50 μg/ml ampicillin Preculture was performed in a shaking incubator at 200 rpm overnight at 37° C. 500 ml of CircleGrow media with 50 μg/ml ampicillin was then inoculated with 50 ml of the precultured CircleGrow media and incubated in a shaking incubator at 200 rpm for 16 hours at 37° C. When optical density at 600 nm (OD.sub.600) reached 1.0, overexpression of an EBP gene or a block polypeptide gene thereof was induced by addition of IPTG at a final concentration of 1 mM. The cells were centrifuged at 4500 rpm for 10 minutes at 4° C. The expressed EBPs and block polypeptides thereof were purified by inverse transition cycling (ITC) as reported previously. The cell pellet was resuspended in 30 ml of HEPES buffer, and the cells were lysed by sonication for 10 s in 20 s intervals (VC-505, Sonics & Materials, Inc, Danbury, Conn.) on ice. The cell lysate was centrifuged in a 50 ml centrifuge tube at 13,000 rpm for 15 min at 4° C. to precipitate the insoluble debris of the cell lysate. Supernatant containing soluble EBPs was then transferred to a new 50 ml centrifuge tube and centrifuged with 0.5% w/v of PEI at 13,000 rpm for 15 minutes at 4° C. to precipitate nucleic acid contaminants. The inverse phase transition of the EBPs were triggered by adding sodium chloride at a final concentration of 4 M, and aggregated EBPs were separated from the lysate solution by centrifugation at 13,000 rpm for 15 minutes at 4° C. The aggregated EBPs were resuspended in cold PBS buffer, and the EBP solutions were centrifuged at 13,000 rpm for 15 minutes at 4° C. to remove any aggregated protein contaminants. These aggregation and resuspension processes were repeated 5 to 10 times until EBP purity reached about 95%, and the purity was determined by SDS-polyacrylamide gel electrophoresis (SDS-PAGE).

(35) FIG. 5A to 5D show a schematic diagram of molecular design, cloning and the anti-neovascularization function of fusion polypeptides according to the present invention. As shown in FIG. 5C, a fusion polypeptide corresponding to VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP forms a temperature-triggered core-shell micellar structure with a multivalent VEGFR-targeting peptide under physiological conditions.

(36) As shown in FIG. 5D (i), a fusion polypeptide corresponding to VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP may act as a therapeutic polypeptide due to strong non-covalent interactions between VEGFRs (in particular, VEGFR1) and the anti-Flt1 peptide. As shown in FIG. 5D (ii), a fusion polypeptide corresponding to VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP forms a micelle with a multivalent anti-Flt1 peptide, which increases the binding affinity of the fusion polypeptide for VEGFRs. Thus, use of the fusion polypeptide may enhance therapeutic efficacy for diseases associated with neovascularization. To minimize rapid degradation of anti-Flt1 peptides and to present anti-Flt1 peptides, as in vivo receptor antagonists, EBPs were introduced to an anti-Flt1 peptide as non-chromatographic purification polypeptide tags and as stabilizers.

(37) Modified pET-21a (mpET-21a) plasmids harboring EBPP[A.sub.1G.sub.4I.sub.1].sub.n, EBPP[E.sub.1G.sub.4I.sub.1].sub.n or EBPP[G.sub.1A.sub.3F.sub.2].sub.n (where the subscript number n of [X.sub.iY.sub.jZ.sub.k].sub.n is 6, 12, 18, 24, 30 or 36) were seamlessly cloned using standard molecular biology methodology. In particular, multimerization and fusion of EBPP genes were executed using recursive directional ligation (RDL) to construct genes encoding EBPPs with different molecular weights and EBPP block copolymers. FIG. 5A shows one method of gene cloning, by which genes for two different EBPPs and genes for an oligonucleotide cassette encoding an anti-Flt1 peptide were combined to prepare a gene for anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24. An mpET-21a plasmid harboring EBPP[G.sub.1A.sub.3F.sub.2].sub.24 was double-digested with XbaI and BseRI and dephosphorylated to prepare a linearized vector, whereas an mpET-21a plasmid harboring EBPP[E.sub.1G.sub.4I.sub.1].sub.12 was double-digested with XbaI and BseRI to prepare an insert. After ligation, a gene for a EBPP diblock of EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 was prepared. The cloned gene was digested with BseRI, dephosphorylated, and fused with an oligonucleotide cassette encoding an anti-Fil1 peptide to prepare anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24. Similarly, a series of genes for anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, 6, 12, 24 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12, 24 were cloned, and fusion polypeptides thereof were synthesized from plasmid-borne genes in E. coli, as shown in FIG. 1 (B).

(38) In VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-fusion polypeptide, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, 6, 12, 24 is soluble under physiological conditions and acts as a VEGFR antagonist to compete with VEGF, thereby inhibiting delivery of neovascularization signals to cells (FIG. 5D(i)). In an embodiment of the present invention, EBPP[A.sub.1G.sub.4I.sub.1].sub.n was selected because EBPP[A.sub.1G.sub.4I.sub.1].sub.n of all lengths is hydrophilic at body temperature without any charged amino acid residues, and because EBPP[A.sub.1G.sub.4I.sub.1].sub.n helps to provide an understanding of the correlation between EBPP length and binding affinity of an anti-Flt1 peptide according to EBPP blocks of four different lengths of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, 6, 12, 24. Furthermore, as shown in FIGS. 5c and 5d (ii), a fusion polypeptide[anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12, 24] of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP may form a temperature-triggered core-shell micellar structure with a multivalent VEGFR1-targeting peptide because of amphiphilic properties of hydrophilic EBPP[E.sub.1G.sub.4I.sub.1].sub.12 and hydrophobic EBPP[G.sub.1A.sub.3F.sub.2].sub.12, 24 under physiological conditions. In addition, these properties enhance the binding affinity of the fusion peptides to VEGFRs and allow the fusion peptides to have high adhesion. In particular, hydrophobic EBPP[G.sub.1A.sub.3F.sub.2].sub.12, 24 of two different lengths was used for micelle size control, and was used to study the effects of micelle size on the binding affinity of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12, 24 to VEGFRs.

Example 7: Characterization of EBPs, Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.].SUB.3n .and Anti-Flt1-EBP Diblock Block

(39) The purity of EBPs, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and anti-Flt1-EBP diblock blocks was determined by SDS-PAGE, and gel permeation chromatography (GPC) with a high-performance liquid chromatography (HPLC) 1260 series instrument (Agilent Technologies, Palo Alto, Calif., U.S.) using a Shodex GPC OHpak SB-804 HQ column (Showa Denko Co., Tokyo, Japan). Deionized water at 20° C. was used as an eluent at a flow rate of 1 ml/min and the GPC column was maintained at 20° C. Low dispersity pullulan in a range of 5,900 to 200,000 g/mol was used as a standard. A series of EBPs, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and anti-Flt1-EBP diblock blocks were analyzed using a refractive index detector (RID) and variable wavelength detector (VWD) at 280 nm. An effect of temperature on the inverse phase transition of various EBPs, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and anti-Flt1-EBP diblock blocks at 25 μM concentration in PBS was determined by measuring OD.sub.350 using a Cary 100 Bio UV/Vis spectrophotometer equipped with a multi-cell thermoelectric temperature controller (Varian Instruments, Walnut Creek, Calif.) between 10 to 85° C. at a heating rate of 1° C./min Self-assembly behaviors of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP [E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 from soluble unimers into micelles were characterized using a temperature-controlled Nano ZS90 (ZEN3690) dynamic light scattering (DLS) instrument (Malvern instruments, Worcestershire, UK), and the hydrodynamic radius (R.sub.H) thereof at 25 μM in PBS was measured in 11 successive runs at each temperature in a temperature range from 18 to 50° C. at a heating rate of 1° C./min. In addition, T.sub.t thereof is defined as the onset temperature for phase transition, and calculated from each DLS plot.

(40) Genes for fusion polypeptides composed of an anti-Flt1 peptide and hydrophilic EBP blocks with different lengths were constructed by molecular cloning and the lengths of those genes digested with XbaI and BseRI were confirmed by agarose gel electrophoresis as shown in FIG. 6(A). The DNA length of each gene encoding anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.6, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 or anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24 (354, 624, 1164 or 2244 bp, from left to right) is indicated below the respective gene fragments. Since DNA sequences digested by XbaI and BseRI are located outside genes encoding the fusion polypeptides, the DNA lengths of the genes are 66 base pairs longer than original gene lengths shown in Table 4. The fusion polypeptides composed of an anti-Flt1 peptide and EBP blocks with different chain lengths were expressed in E. coli and purified by ITC, as previously reported for the temperature-responsive EBPs. A copper-stained SDS-PAGE gel (4 to 20% gradient) shown in FIG. 6(B) shows the following: Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.n (subscript n is 3, 6, 12, or 24) was purified to have a homogeneity of at least 95% by an average of five rounds of ITC as characterized by HPLC. Compared to a standard protein migration distance, each fusion polypeptide shifted about 20% more than theoretical molecular weights shown in Table 4, which is in good agreement with previous studies. The expected molecular weights of the fusion polypeptides are indicated below each band (8.19, 15.27, 29.42 and 57.72 kDa, from left to right), and lanes at both ends of the SDS-PAGE gel represent standard protein size markers (7, 15, 24, 35, 40, 50, 65, 90, 110, and 150 kDa, from bottom to top).

(41) FIG. 7 shows the thermal transition behaviors of EBP blocks and the thermal transition behaviors of fusion polypeptides composed of an anti-Flt1 peptide and EBP blocks with different chain lengths. Based on the thermal transition behaviors, the effect of EBP block length, sodium chloride concentration and anti-Flt1 peptide fusion on transition temperature (T.sub.t) may be investigated. Turbidity profiles in FIG. 7 were obtained by measuring the absorbance of (A to C) 25 μM EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) and (D to F) 25 μM anti-Flt1-EBPP[A.sub.1G.sub.4I].sub.3n (n: integer) in 10 mM PBS (A and D) and in 10 mM PBS supplemented with 1 M sodium chloride (B and E) or 2 M sodium chloride (C and F) at 350 nm while heating samples at a rate of 1° C./min. T.sub.t is defined as the inflection point of each thermal plot in FIG. 7 and summarized in Table 5.

(42) TABLE-US-00007 TABLE 5 (a) (b) (c) (d) (e) (f) (g) (h) 0M NaCl N/A N/A N/A 68 N/A N/A 67 57 1M NaCl N/A 80 52 42 N/A 57 45 39 2M NaCl N/A 53 34 28 49 34 28 23 T.sub.1 of (a) EBPP[A.sub.1G.sub.4I.sub.1].sub.3, (b) EBPP[A.sub.1G.sub.4I.sub.1].sub.6, (c) EBPP[A.sub.1G.sub.4I.sub.1].sub.12, (d) EBPP[A.sub.1G.sub.4I.sub.1].sub.24, (e) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3, (f) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.6, (g) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and (h) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24

(43) T.sub.t values in Table 5 are determined by measuring the inflection points of thermal profiles in FIG. 7. Transition temperature was changed depending on EBPP[A.sub.1G.sub.4I.sub.1] block length and sodium chloride concentration.

(44) In general, EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and anti-Flt1-EBPP [A.sub.1G.sub.4I.sub.1].sub.3n without polar amino acid residues exhibit T.sub.t higher than 37° C. under physiological conditions because Ala, Gly and Ile were introduced to the EBPPs as the guest residue of the repetitive pentapeptide unit of the EBPPs in a ratio of 1:4:1. Anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n was hydrophilic and VEGFR binding-fusion polypeptides thereof were soluble under physiological conditions, which allowed the polypeptides to specifically bind to VEGFRs without any steric hindrance. Thus, the fusion polypeptides of the present invention may act as VEGFR antagonists against VEGF. Furthermore, when the effect of EBPP block length and ionic strength on thermal responsiveness was analyzed, as the EBP block length of EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n, and sodium chloride concentration in PBS increased, T.sub.t thereof decreased. In particular, the T.sub.t of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n was much lower than that of EBPP[A.sub.1G.sub.4I.sub.1].sub.3n because Gly-Asn-Gln-Trp-Phe-Ile (GNQWFI) of an anti-Flt1 peptide sequence for targeting VEGFRs was hydrophobic, resulting in a decrease in the T.sub.t of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n. For example, the T.sub.t of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and the T.sub.t of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.24 were about 18 and 11° C. lower than those of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and EBPP[A.sub.1G.sub.4I.sub.1].sub.24 in PBS, respectively. A T.sub.t difference (DT.sub.t) between EBPP[A.sub.1G.sub.4I.sub.1].sub.3 and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 was more than 36° C. in PBS with 2 M sodium chloride, whereas DT.sub.t between EBPP[A.sub.1G.sub.4I.sub.1].sub.3 and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 was 23° C. in PBS with 1 M sodium chloride. Therefore, as EBPP[A.sub.1G.sub.4I.sub.1] block length became shorter, the T.sub.t of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n was greatly decreased irrespective of various concentrations of sodium chloride. This data indicates that the effect of hydrophobicity of the anti-Flt1 peptide on the thermal transition of the EBPP[A.sub.1G.sub.4I.sub.1] block is potentially greater.

(45) Next, the properties of fusion polypeptides composed of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP are described. Two different genes, which encode a fusion polypeptide composed of “anti-Flt1 peptide” and “amphiphilic EBP diblock” of hydrophilic EBP-hydrophobic EBP having hydrophobic EBP blocks with various chain lengths, were constructed using RDL, a seamless molecular cloning method. The full lengths of those genes digested by XbaI and BseRI were confirmed by agarose gel electrophoresis as shown in FIG. 8(A). The DNA length of each gene encoding (a) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 or (b) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 (2244 and 3324 bp, from left to right) is indicated below each gene fragment. Since DNA sequences digested by XbaI and BseRI are located outside genes encoding the fusion polypeptides, the lengths of these genes are 66 base pairs longer than the original gene lengths of the fusion polypeptides the shown in Table 4. Two different anti-Flt1-EBP diblock copolypeptides including (a) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and (b) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 were expressed in E. coli and purified by one among non-chromatographic purification methods, ITC as described above for purification of a series of temperature-responsive anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n. The image of a copper-stained SDS-PAGE gel (4 to 20% gradient) in FIG. 8(B) shows the following: Both anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 were purified by an average of five rounds of ITC, each with one major polypeptide band. Compared to a standard protein migration distance, each fusion polypeptide shifted about 20% more than theoretical molecular weights shown in Table 4. In addition, as characterized by HPLC, after an average of five rounds of ITC runs, each polypeptide had a homogeneity of at least 95%. The expected molecular weights of the polypeptides are indicated below each band (59.9 and 90.0 kDa, from left to right), and lanes at both ends of the SDS-PAGE gel represent standard protein size markers (7, 15, 24, 35, 40, 50, 65, 90, 110, and 150 kDa, from bottom to top).

(46) FIG. 9 shows the thermal transition behaviors of anti-Flt1-EBP diblock copolypeptides of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 depending on the length and concentration of a hydrophobic EBPP[G.sub.1A.sub.3F.sub.2] block. Turbidity profiles were obtained by measuring absorbance at 350 nm at four different concentrations (12.5, 25, 50 and 100 μM) in 10 mM PBS at a heating rate of 1° C./min. As described above, T.sub.t was measured as the inflection point of each thermal plot in FIG. 9 and summarized in Table 6 below.

(47) TABLE-US-00008 TABLE 6 (a) (b) (c) (d) Conc. (uM) 25 25 12.5 25 50 100 12.5 25 50 100 First T.sub.t 39.02 29.12 34.2 36.2 37.4 39.5 26.7 28.0 29.1 29.2 (° C.) Second T.sub.t N/A N/A 78.2 81.2 82.3 84.4 74.6 78.0 79.2 81.2 (° C.) T.sub.1 of (a) EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12, (b) EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24, (c) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and (d) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24

(48) T.sub.t values in Table 6 are determined by measuring the inflection points of thermal profiles in FIG. 7. The first phase transition occurred as a result of hydrophobic block aggregation, and was greatly affected by the length of EBPP[A.sub.1G.sub.3F.sub.2]. The fusion polypeptides thereof had the same polar EBPP[E.sub.1G.sub.4I.sub.1].sub.12 block. The second phase transition was affected by a polar EBPP[E.sub.1G.sub.4I.sub.1].sub.12 block, and the fusion polypeptides thereof had a similar second T.sub.t

(49) As the concentration of anti-Flt1-EBP diblock blocks increased, the first T.sub.t and the second T.sub.t gradually decreased. In general, the temperature-triggered phase transition of anti-Flt1-EBP diblock copolypeptides occurs twice, because aliphatic- and hydrophobic EBPP[A.sub.1G.sub.3F.sub.2] block having a low T.sub.t and polar- and hydrophilic EBPP[E.sub.1G.sub.4I.sub.1] block having a high T.sub.t exhibit different thermal properties. The phase transition of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 occurred at 36.2 and 81.2° C. at the 25 μM concentration, whereas the phase transition of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 occurred at 28.0 and 78.0° C. at the same concentration. This data indicates that the doubled block length of the hydrophobic EBPP[A.sub.1G.sub.3F.sub.2] has a significant effect on the first T.sub.t and the second T.sub.t, lowering the same by 8.2 and 3.2° C., respectively. In particular, diblock polypeptides of EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 without anti-Flt1 fusion, as a control, exhibited a first T.sub.t of only 39.0 and 29.1° C. without an additional phase transition, as shown in FIG. 9(A). On the other hand, anti-Flt1-EBP diblock copolypeptides clearly exhibited a lowered first T.sub.t and second T.sub.t as opposed to the phase transition behavior of diblock polypeptides without anti-Flt1, because fusion of a hydrophobic anti-Flt1 peptide (Gly-Asn-Gln-Trp-Phe-Ile (GNQWFI)) and the hydrophilic EBPP[E.sub.1G.sub.4I.sub.1] block of diblock polypeptides greatly decreased the first T.sub.t of EBPP[G.sub.1A.sub.3F.sub.2] and the second T.sub.t of a hydrophilic EBPP[E.sub.1G.sub.4I.sub.1] block, which was due to the proximity of these blocks. Furthermore, in anti-Flt1-EBP diblock copolypeptides, EBPP[A.sub.1G.sub.3F.sub.2] and EBPP[E.sub.1G.sub.4I.sub.1], with block lengths adjusted at exactly 1:1 and 1:2 ratios, created a unique metastable micelle phase right above the first T.sub.t thereof, which indicated that the thermally-triggered amphiphilic anti-Flt1-EBP diblock copolypeptides self-assembled into a metastable micelle. At this time, the metastable micelle continued to develop as a stable micelle in a temperature range from the first T.sub.t to the second T.sub.t. This is in good agreement with the self-assembly behaviors of diblock polypeptides of EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 that are not fused with anti-Flt1, as in a EBP-based diblock copolymer-based micelle reported previously.

(50) In accordance with the unique thermal transition of anti-Flt1-EBP diblock copolypeptides, the self-assembly behaviors of anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24 from soluble unimers into micelles were characterized by dynamic light scattering (DLS). The hydrodynamic radius (R.sub.H) thereof at 25 μM in PBS was measured in 11 successive runs at each temperature in a temperature range from 18 to 50° C. at a heating rate of 1° C./min. T.sub.t thereof was defined as the onset temperature for phase transition, calculated from each DLS plot in FIG. 10, and summarized in Table 7 below.

(51) TABLE-US-00009 TABLE 7 (a) (b) First T.sub.t (° C.) Second T.sub.t (° C.) First T.sub.t (° C.) Second T.sub.t (° C.) Absorbance 36.2 82.3 28.0 79.2 DLS 36.0 N/A 27.0 N/A T.sub.t of (a) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and (b) anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24

(52) Referring to Table 7, the first aggregation of fusion polypeptides increases the hydrodynamic radius thereof due to micelle formation.

(53) Anti-Flt1-EBP diblock copolypeptides existed in soluble unimer forms below the first T.sub.t of 36° C. for anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 and below the first T.sub.t of 27° C. for anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.24, and the hydrodynamic radius (R.sub.H) thereof at 25 μM in PBS was about 10 nm. As temperature increased above the first T.sub.t, the R.sub.H thereof instantaneously increased in a range of 160 and 240 nm at a slightly higher temperature than the first T.sub.t, then decreased to 28.4 and 43.6 nm. The anti-Flt1-EBP diblock copolypeptides formed metastable micelles due to non-equilibrium thermodynamics of amphiphile-based self-assembly and different hydrophilic-to-hydrophobic block length ratios, and then the copolypeptides formed stable micelles with constant R.sub.H values even at 50° C. because self-assembly thereof reached equilibrium. The R.sub.H of an anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 micelle was 15.2 nm larger than that of the anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 micelle due to the doubled block length of EBPP[G.sub.1A.sub.3F.sub.2] and the bigger aggregated domain of the EBPP[G.sub.1A.sub.3F.sub.2] block at the core of the micellar structure thereof. Furthermore, to determine the critical micelle concentrations (CMCs) of the anti-Flt1-EBP diblock copolypeptides, the micelle sizes thereof at various concentrations in a range of 0.1 to 25 μM were measured at 20° C. below T.sub.t and 37° C. above T.sub.t. Under environmental conditions of 0.5 μM and 37° C., the anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-[G.sub.1A.sub.3F.sub.2].sub.12 still formed a metastable micelle with a R.sub.H of ˜125 nm and the anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 formed a stabilized micelle with a R.sub.H of ˜44 nm. However, no micelle formation was observed for the two anti-Flt1-EBP diblock copolypeptides at 0.1 μM and 37° C., indicating that the 0.1 μM concentration was lower than CMCs thereof, and the CMCs were in a range of 0.1 to 0.5 μM. Therefore, the anti-Flt1-EBP diblock copolypeptides formed temperature-triggered core-corona micellar structures with multivalent anti-Flt1 peptides for targeting Flt1 under physiological conditions because of the amphiphilic properties of hydrophilic EBPP[E.sub.1G.sub.4I.sub.1] and hydrophobic EBPP[G.sub.1A.sub.3F.sub.2]. In particular, in the anti-Flt1-EBP diblock copolypeptides, different block lengths of hydrophobic EBPP[G.sub.1A.sub.3F.sub.2] finely controlled micellar size, which affected the binding affinity thereof to Flt1, resulting in high adhesion.

Example 8: Determination of Specific Binding of Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.].SUB.3n .and Anti-Flt1-EBP Diblock Block to Flt1

(54) Specific binding of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: 1, 2, 4, and 8) and anti-Flt1-EBP diblock copolypeptides to Flt1 was determined by enzyme-linked immunosorbent assay (ELISA). First, to coat a 96 well plate with recombinant human VEGF165 protein (rhVEGF.sub.165) (M.W. 38.4 kDa) present in a disulfide-linked homodimer, 50 μl of a solution containing the rhVEGF.sub.165 at a concentration of 0.5 μg/ml was added to the 96 well plate, and the plated was incubated at 4° C. overnight. The wells of the 96 well plate coated with the rhVEGF.sub.165 were washed with PBS containing 0.05% Tween-20 to completely remove unattached rhVEGF.sub.165, and then the wells were incubated with PBS containing 3 wt % BSA at room temperature for 2 hours to block the surface of each well, which was not coated with the rhVEGF.sub.165. After incubation, the wells were washed with PBS containing 0.05% Tween-20 to remove unbound BSA. Next, to impart specific binding affinity between an anti-Flt1 peptide and Flt1 (VEGFR1), a recombinant human Flt1-F.sub.c chimeric protein (M.W. 200.0 kDa) present in a disulfide-linked homodimer at a concentration of 0.5 μg/ml was pre-incubated with (1) anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: 1, 2, 4, and 8) in PBS containing 1 wt % BSA or with (2) anti-Flt1-EBP diblock copolypeptides (anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP [G.sub.1A.sub.3F.sub.2] 12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24) with hydrophobic blocks of different lengths. In this case, the pre-incubation was carried out at room temperature for 2 hours at different concentrations within a range of 0.5 to 500 μM. Thereafter, the mixed solution was added to rhVEGF.sub.165-coated wells, followed by additional incubation at room temperature for 2 hours. The EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and EBP diblock copolypeptide (EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24) with hydrophobic blocks (having the same concentration) of different lengths were used as a standard. Each well was washed with PBS supplemented with 0.05% Tween-20 to remove Flt1-F.sub.c protein that was not bound to rhVEGF.sub.165 on the surface of the well. Whether human Flt1-F.sub.c protein was specifically bound to the rhVEGF.sub.165-coated well was determined by measuring the absorbance of oxidized chromogenic substrates upon protein-antibody binding at 450 nm using rabbit anti-human IgG F.sub.c-horseradish peroxidase (HRP) conjugates as a secondary antibody. PBS (containing 0.3 w % BSA) diluted with anti-human IgG F.sub.c-HRP was added to each well and incubated for 1 hour at room temperature, followed by washing 8 times with PBS containing 0.05 Tween-20. 3,3′,5,5′-tetramethylbenzidine (TMB) was added to each well to indirectly determine the degree of specific binding of Flt1-F.sub.c protein to VEGF by measuring the specific interaction between the Flt1-F.sub.c protein and the anti-human IgG F.sub.c-HRP protein, and HRP-catalyzed oxidation of the TMB. The color intensity of the oxidized TMB was measured at 450 nm. Each ELISA experiment was performed three times for reproducibility.

(55) The specific binding properties of a fusion polypeptide of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP are examined. As shown in FIG. 11, the specific binding of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) fusion polypeptides was characterized by enzyme-linked immunosorbent assay (ELISA). First, 38.4 kDa recombinant human VEGF.sub.165 protein present in a disulfide-linked homodimer was coated on wells, and then the wells were blocked by bovine serum albumin (BSA). A 200.0 kDa recombinant human Flt1-F.sub.c chimeric protein present in a disulfide-linked homodimer was incubated with anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: 1, 2, 4, and 8) at different concentrations within a range of 0.5 to 500 μM to induce specific binding between each other, and then the mixed solution was added to the VEGF-coated wells, followed by incubation for 2 hours at room temperature. Human Flt1-F.sub.c chimeric protein was specifically bound to the VEGF-coated wells was determined by measuring the absorbance of oxidized chromogenic substrates upon protein-antibody binding at 450 nm using rabbit anti-human IgG F.sub.c-horseradish peroxidase (HRP) conjugates as a secondary antibody. Regardless of different concentrations, EBPP[A.sub.1G.sub.4I.sub.1].sub.12 did not significantly inhibit specific binding between the Flt1-F.sub.c chimeric protein and VEGF. Contrary to the minimal inhibitory effect of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 with respect to the specific binding, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n copolypeptides significantly inhibited an interaction between Flt1-F.sub.c and VEGF in a dose-dependent manner independent of EBPP[A.sub.1G.sub.4I.sub.1] block length. These results indicate that anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n has a high specific binding capacity to a human Flt1-F.sub.c chimeric protein, which may prevent the human Flt1-F.sub.c chimeric protein from binding to VEGF. In particular, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 showed a maximum inhibitory effect of about 75% at 500 μM, whereas anti-Flt1-EBPP [A.sub.1G.sub.4I.sub.1].sub.24 had a lower inhibitory effect than anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12, which might be the consequence of steric hindrance caused by an extended EBPP[A.sub.1G.sub.4I.sub.1] chain length. Although hydrophilic EBPP[A.sub.1G.sub.4I.sub.1].sub.3 blocks with different chain lengths were introduced to anti-Flt1 peptides as VEGFR1-specific antagonists, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n copolypeptides retained high specificity of the anti-Flt1 peptide for Flt1, due to the inert nature of EBPs. In contrast to conventional peptide-polymer conjugates such as anti-Flt1 peptide-hyaluronate (HA) conjugates, the anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n copolypeptides were prepared at the gene level, and imparted the monodisperse molecular weight and enhanced stabilization of an anti-Flt1 peptide due to the inert nature of EBPs acting like PEG This monodisperse molecular weight and stability might increase the half-life of the anti-Flt1 peptide in vivo.

(56) Next, the binding properties of fusion polypeptides of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP are examined With specific binding of soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n to a human Flt1-F.sub.c chimeric protein, anti-Flt1-EBP diblock blocks (anti-Flt1-EBPP [E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 and anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP [G.sub.1A.sub.3F.sub.2].sub.24) with hydrophobic blocks of different lengths formed temperature-triggered core-shell micellar structures with multivalent anti-Flt1 peptides under physiological conditions. Multivalent anti-Flt1 located on the outer shell of the formed self-assembled micelles increased the binding affinity of the fusion polypeptides to human Flt1 (VEGFR1). As measured by enzyme-linked immunosorbent assay (ELISA) in FIG. 12, as the concentrations of fusion polypeptides of anti-Flt1 peptide-hydrophilic EBP-hydrophobic EBP increased, specific binding between Flt1-F.sub.c and VEGF was significantly inhibited by the fusion polypeptides, which is in good agreement with the results of the example for soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n. Unlike the degree of inhibition of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n fusion polypeptides with respect to specific binding between Flt1-F.sub.c and VEGF, anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 micelles with a R.sub.H of ˜125 nm in a metastable state showed a dramatically enhanced inhibitory effect (˜95%) on specific binding between Flt1-F.sub.c and VEGF depending on the spatial multivalent display of Flt1-targeting peptides on the micelles. On the other hand, anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.24 micelles with a R.sub.H of ˜44 nm in a stable state exhibited a similar inhibition degree compared with the degree of inhibition of soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n blocks. Although the two anti-Flt1-EBP diblock copolypeptides formed micelles in a concentration range from 0.5 to 500 μM at 37° C. under physiological conditions, the peptides showed a much different degree of inhibition with respect to specific binding between Flt1-F.sub.c and VEGF based on the stability of micelles thereof, potentially due to different binding affinities between the Flt1-F.sub.c and the controlled spatial display of the multivalent anti-Flt1 peptides of the micellar nanostructures. Importantly, anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 micelles at 250 μM had a greater inhibitory effect than anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 fusion polypeptides at 500 μM, which suggested that a lower dose of the anti-Flt1-EBPP[E.sub.1G.sub.4I.sub.1].sub.12-EBPP[G.sub.1A.sub.3F.sub.2].sub.12 might have a higher binding affinity to the human Flt1 protein in vivo for anti-neovascularization.

Example 9: In Vitro Tubing Assay of HUVECs Using Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.].SUB.12

(57) In vitro tubing assay of HUVECs using soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 was performed to evaluate effects of the soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 copolypeptides on proliferation, migration and tube formation of endothelial cells. For Matrigel coating, 200 μl of 8.7 mg/ml Matrigel was added to a 48 well plate and incubated at 37° C. for 1 hour to become solidified. To label HUVECs with fluorescence, HUVECs were incubated with 10 μM calcein-AM at 37° C. for 15 minutes and washed with PBS several times. The calcein-labeled HUVECs at 2×10.sup.4 cells/well were grown on the Matrigel-coated wells, and incubated at 37° C. for 4 hours with 50 ng/ml recombinant human rhVEGF.sub.165 and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 as a Flt1-specific antagonist at different concentrations. After incubation, it was determined whether proliferation, migration and tube formation of endothelial cells were stimulated. To clarify to what extent anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 could inhibit tube formation of HUVECs, the same concentration of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 was assessed as a control. In addition, Avastin, a recombinant humanized monoclonal antibody (mAb) against VEGF, was used as another control to compare therapeutic efficacy for anti-neovascularization based on the therapeutic efficacy of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12, as a Flt1-specific antagonist. The tube formation of HUVECs was photographed with Micromanipulator (Olympus, Tokyo, Japan), and quantified by measuring whole tube lengths in three random fields per well with Image lab software (Bio-Rad Laboratories, Hercules, Calif., USA). When the tubing assay was performed, the tube formation of HUVECs incubated in PBS for 4 hours was used as a control. The experiment was repeated three times.

(58) As shown in FIG. 11, based on enzyme-linked immunosorbent assay (ELISA) results, it was confirmed that anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n (n: integer) fusion polypeptides had a high specific binding capacity to the human Flt1-F.sub.c chimeric protein, and 29.4 kDa anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 showed a maximum degree of inhibition compared with soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n fusion polypeptides with different EBP chain lengths. Accordingly, the effects of soluble anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 fusion polypeptides on proliferation, migration and tube formation of endothelial cells were assessed in HUVECs in vitro. Calcein-labeled HUVECs at 2×10.sup.4 cells/well were grown on a 48 well plate pre-coated with Matrigel, 50 ng/ml recombinant human rhVEGF.sub.165 was treated to stimulate proliferation, migration and tube formation of endothelial cells, and anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 acting as a Flt1 (VEGFR1)-specific antagonist was treated at different concentrations, followed by incubation. The same concentration of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 was used as a control to clearly show to what extent anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 could inhibit the tube formation of HUVECs. In addition, Avastin (also named bevacizumab), a recombinant humanized monoclonal antibody (mAb) against VEGF, was used as a control to compare therapeutic efficacy for anti-neovascularization. Avastin has been widely used to treat various neovascular eye diseases, such as age-related macular degeneration (AMD) and diabetic retinopathy, based on specific binding between Avastin and VEGF. Based on specific binding of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3 to the human Flt1 protein in the membrane of HUVECs, it was assumed that the inhibitory effects of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n and Avastin on the tube formation of HUVECs were caused by specific protein-protein interactions, while Avastin bound to rhVEGF.sub.165 and minimized rhVEGF.sub.165-triggered cellular signaling for the tube formation and neovascularization of HUVECs. As characterized by in vitro tubing assay of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12, fluorescence images of calcein-AM labeled HUVECs and the degree of inhibition of tube formation based on the normalized tube length of HUVECs in the images were shown in FIG. 13. The tube length of HUVECs was measured by tracking the fluorescence signal of HUVECs, averaged under each condition, and the tube length of HUVECs incubated in PBS for 4 hours was used as a baseline. A tube length when HUVECs were treated with rhVEGF.sub.165 for 4 hours was set at 100%, and tube length at each concentration was normalized. The tube length of HUVECs incubated with EBPP[A.sub.1G.sub.4I.sub.1].sub.12 as a control was similar to that of HUVECs treated with rhVEGF.sub.165, indicating that EBPP[A.sub.1G.sub.4I.sub.1].sub.12 had no significant effect on the tube formation of HUVECs. On the other hand, in the case of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 as a Flt1-specific antagonist, as the concentration of the anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 increased in a range of 0.1 to 10 μM, the tube length of HUVECs gradually decreased. In accordance with decrease of the tube length of HUVECs when incubated with anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12, fluorescence images clearly show that the degree of inhibition of migration and tube formation of HUVECs became evident as the concentration of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 increased. These results indicate that anti-Flt-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 inhibits the tube formation of HUVECs and the degree of inhibition is greatly controlled by the concentration of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12. In particular, HUVECs incubated with 10 μM anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 showed no significant migration and tube formation even in the presence of rhVEGF.sub.165, which was similar to use of Avastin at 0.2 mg/ml. Therefore, as validated by ELISA, the anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 fusion polypeptides still retained high specificity of an anti-Flt1 peptide against Flt1.

Example 10: In Vivo Anti-Neovascularization Using Anti-Flt1-EBPP[A.SUB.1.G.SUB.4.I.SUB.1.].SUB.12 .in Laser-Induced Choroidal Neovascularization Model

(59) 6- to 8-week-old female C57BL-6 mice were anesthetized with intraperitoneal injection of ketamine at 100 mg/kg and xylazine at 10 mg/kg, and the pupils were dilated with 5 mg/ml tropicamide, and 532 nm laser diode (150 to 210 mW, 0.1 sec, 50 to 100 μM) was applied to each fundus to induce choroidal neovascularization in vivo. Multiple burns were performed in the 6, 9, 12, and 3 o'clock positions of the posterior pole of the eye with a slit-lamp delivery system. Production of bubbles at the time of laser, which indicates Bruch's membrane rupturing, is an important factor in obtaining the CNV model. To evaluate an effect of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n copolypeptides on anti-neovascularization in a laser-induced choroidal neovascularization model in vivo, the CNV model mice were injected in an intravitreal manner with PBS as a vehicle, EBPP[A.sub.1G.sub.4I.sub.1].sub.12 or various concentrations of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 once a day for 5 days and anesthetized after 14 days with an intraperitoneal injection of ketamine at 100 mg/kg and xylazine at 10 mg/kg. The mice were treated with retro-orbital injection of 100 μl ultrapure water containing 25 mg/ml FITC-dextran. Enucleated eyes were then fixed in 10% formalin for 30 minutes at room temperature. The cornea, iris, lens, and vitreous humor were gently removed under a stereomicroscope (Leica, Wetzlar, Germany). Four radial incisions were made in the dissected retina, which was then flattened with a coverslip. Each in vivo anti-neovascularization experiment was performed with three replicates.

(60) By ELISA and HUVEC tubing assay, it was demonstrated that anti-Flt1-EBPP[A G.sub.4I.sub.1].sub.3n fusion polypeptides retained anti-neovascularization activity as an antagonist against VEGFR1. The present inventors hypothesized that anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n fusion polypeptides might show a therapeutic activity with respect to neovascularization-related eye diseases (in particular, retinal neovascular disease, age-related macular degeneration (AMD)). Intravitreal injection of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 was evaluated for the suppression of laser-induced choroidal neovascularization (CNV), which was an animal model for AMD, in C57BL-6 mice. Daily injection of protein solutions started immediately after laser injury and maintained for 5 days. Injection of a vehicle (PBS) or EBPP[A.sub.1G.sub.4I.sub.1].sub.12 was used as a negative control. CNV lesion volumes were imagined and evaluated with fluorescein isothiocyanate (FITC)-dextran perfused whole choroidal flat-mounts at day 14 after laser injury (FIG. 14A). Quantitative analysis showed that doses of 0.1, 1, 5 and 20 μg of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 per day for 5 days (in total 0.5, 5, 25 and 100 μg) suppressed CNV lesion size by 32%, 52.3% (P<0.05), 54.4% (P<0.05), and 25.9%, respectively, as compared with PBS control mice (FIG. 14B). The CNV lesion sizes of an EBPP[A.sub.1G.sub.4I.sub.1].sub.12-treated animal had values similar to those of a PBS-treated animal. The suppressive effect of EBPP[A.sub.1G.sub.4I.sub.1].sub.12 on a CNV lesion showed a dose dependent manner in a range from 0.5 to 25 μg in total. However, 100 μg anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 showed a reduced effect on suppression of the CNV lesion, potentially due to an excessive dose of anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12.

(61) Binding affinity of a targeting ligand against a growth factor receptor (GFR) in cells is important for various diseases associated with cell growth such as neovascularization, because the binding affinity determines whether intracellular signaling will proceed. In the present invention, VEGFR-targeting fusion polypeptides, which are composed of thermally responsive elastin-based polypeptides (EBPs) and vascular endothelial growth factor receptor (VEGFR)-targeting peptides, were genetically manipulated, expressed, and purified and the physicochemical properties thereof were analyzed. The EBPs were introduced as non-chromatographic purification tags and also introduced as a stabilizer, like a poly(ethylene glycol) conjugate, for minimizing rapid in vivo degradation of VEGFR-targeting peptides. In addition, the VEGFR-targeting peptide was introduced to function as a receptor antagonist by specifically binding to VEGFRs.

(62) A fusion polypeptide composed of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP exhibited a soluble unimer form. On the other hand, a fusion polypeptide composed of VEGFR-targeting peptide (anti-Flt1 peptide)-hydrophilic EBP-hydrophobic EBP exhibited a temperature-triggered core-shell micellar structure with a multivalent VGFR-targeting peptide under physiological conditions. As analyzed by enzyme-linked immunosorbent assay (ELISA), this structure greatly increased the binding affinity of the fusion polypeptide for VEGF receptors. Depending on the spatial display of a VEGFR-targeting peptide, the binding affinity of the fusion polypeptides to VEGFRs was greatly regulated.

(63) An anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.3n fusion polypeptide (anti-Flt1 peptide-hydrophilic EBP), which existed as a soluble unimer form below a transition temperature, showed a high anti-neovascularization effect in a CNV model as compared with a EBPP block as a control. In addition, an anti-Flt1-EBP diblock fusion polypeptide (anti-Flt1 peptide-hydrophilic EBP-hydrophobic EBP) formed a temperature-triggered, self-assembled multivalent micellar nanostructure under physiological conditions, resulting in a great difference in the degree of inhibition with respect to specific binding between Flt1-F.sub.c and VEGF depending on the stability of the micellar nanostructure thereof. In the tube formation assay of HUVECs in vitro, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 greatly reduced tube formation, whereas EBPP[A.sub.1G.sub.4I.sub.1].sub.12 had no significant effect on tube formation, which was due to specific interactions between the anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 and Flt1 (VEGFR1) on the HUVEC membrane. Finally, in the laser-induced CNV model of mice, anti-Flt1-EBPP[A.sub.1G.sub.4I.sub.1].sub.12 showed a high anti-neovascularization effect. Therefore, this fusion polypeptide and the self-assembled multivalent micellar nanostructure thereof with an anti-Flt1 may be used as a therapeutic polypeptide targeting neovascularization, such as treatment of retinal, corneal, choroidal neovascularization, tumor growth, cancer metastasis, diabetic retinopathy, and asthma.

(64) A fusion polypeptide for inhibiting neovascularization of the present invention can provide a new direction for a drug delivery system for anti-neovascularization, such as treatment of retinal, corneal, choroidal neovascularization, tumor growth, cancer metastasis, diabetic retinopathy, and asthma.