WO2003087329A2

WO2003087329A2 - Targeted cytocidal virionoids for antiangiogenesis

Info

Publication number: WO2003087329A2
Application number: PCT/US2003/011142
Authority: WO
Inventors: Frederick L. Hall; Erlinda M. Gordon
Original assignee: University Of Southern California
Priority date: 2002-04-11
Filing date: 2003-04-11
Publication date: 2003-10-23
Also published as: AU2003230873A1; AU2003230873A8; WO2003087329A3

Abstract

The present invention provides an artificial VEGF peptide-bearing non-infectious retroviral particle. The viral particle lacks a fusogenic envelope and is capable of VEGF receptor-mediated binding to endothelial cells of a preselected species. The present invention also provides a method of selectively inhibiting tumor angiogenesis in a mammalian subject by administering to the subject an effective amount of a non-infectious viral particle comprising on its surface a VEGF peptide that specifically binds to a VEGF receptor of the preselected mammalian species. The particle selectively binds to a VEGF receptor-bearing vascular endothelial cell of the subject without delivery of genetic material from the particle. This binding is cytotoxic to the cell and results in a selective inhibition of angiogenesis in a tumor in the subject. The figure illustrates molecular engineering VEGF isoforms into the primary structure of modified MLV envelope proteins.

Description

TARGETED CYTOCIDAL VIRIONOIDS FOR ANTIANGIOGENESIS

BACKGROUND

1. Field of the Invention

The present invention relates to the medical arts, and in particular, to viral medicines.

2. Discussion of the Related Art

At the turn of the century, gene therapy remains poised at the threshold of modernizing medicine. While holding great promise for the treatment of numerous diseases, the field of gene therapy has been disappointingly slow in the development of safe and efficient gene delivery systems (Anderson, W.F., (1998) Human Gene Therapy, Nature (Suppl) 392, 25-30). Numerous attempts to target specific cell types have focused on modifying the receptor binding domain (SU) of the ecotropic Moloney murine leukemia virus MuLV) envelope (env) protein. (Kasahara, N, Dozy, AM, Kan, YW (1994) Tissue-specific targeting of retroviral vectors through ligand-receptor interactions, Science 266:1373-1376; MacKrell, AJ, Soong, NW, Curtis, CM, Anderson, WF (1996) Identification of a subdomain in the Moloney murine leukemia virus envelope protein involved in receptor binding, J Virol 70:1768-1774; Somia, NV, Zoppe, M, Verma, IM (1995) Generation of targeted retroviral vectors by using single-chain variable fragment: an approach to in vivo gene delivery, Proc Natl Acad Sci USA 92:7570-7574; Peng, KW, Russell, SJ (1999) Viral vector targeting, Curr Opin Biotechnol:454-457). In many cases, however, the molecular modification of the env protein prevents the obligatory conformational change that occurs in response to receptor binding which is required for viral fusion and core entry. (Zhao, Y, Zhu, Y, Lee, S, Li, L., Chang, E, Soong, NW, Douer, D, Anderson, WF (1999) Identification of the block in targeted retroviral-mediated gene transfer, Proc Nat'l Acad Sci USA 96: 4005-4010). Consequently, severe loss of viral infectivity often occurs. (Benedict, CA, Tun, RY, Rubinstein, DB, Guillaume, T, Cannon, PM, Anderson, WF (1999): Targeting retroviral vectors to CD34-expressing cells: binding to CD34 does not catalyze virus-cell fusion, Hum Gene Ther 10:545- 557; Morling, FJ, Peng, KW, Cosset, F-L, Russell, SJ (1997) Masking of retroviral envelope functions by oligomeri∑ing polypeptide adaptors, Virology 234:51-61 ; Cosset, FL, Russell, SJ (1996) Targeting retrovirus entry. Gene Ther 3:946-956). Among the only clinically useful targeting approaches heretofore involves the use of a pathology-targeted (pathotropic) vector which, when injected intravenously, seeks and accumulates in diseased or cancerous lesions within the body. (Hall, FL, Russell, SJ (1996) Targeting retrovirus entry, Gene Ther 3:946-956.; Gordon, EM , Liu, P, Chen, ZH, Liu L, Whitley, MD, Gee, C, Groshen, S, Hinton, DR, Beart, RW, Hall, FL (2000): Inhibition ofmetastatic tumor growth in nude mice by portal vein infusions of matrix-targeted retroviral vectors bearing a cytocidal cyclin Gl construct. Cancer Res 60:3343-3347; Gordon, EM, Liu, PX, Chen, ZH, Liu, L, Whitley, MD, Liu, L, Wei, D, Groshen, S, Hinton, DR, Anderson, WF, Beart, RW, Jr, Hall, FL (2001a) Systemic administration of a matrix- targeted retroviral vector is efficacious for cancer gene therapy in mice, Hum Gene Ther 12: 193- 204).

In recent years, "escort proteins" has been introduced to improve cell-specific targeting and gene delivery to activated endothelial cells. (Liu, L, Liu L, Anderson WF, Gordon EM, Hall FL. (2000) Incorporation of tumor vasculature targeting motifs (TVTMs) into MLV env escort proteins enhances retroviral binding and transduction of human endothelial cells, J Virol, 74:5320-5328; Masood, R, Gordon, EM, Whitley, MD, Wu, BW, Cannon, P, Evans, L, Anderson, WF, Gill, P, Hall, FL (2001) Retroviral vectors bearing IgG-binding motifs for antibody-mediated targeting of vascular endothelial growth factor receptors, Int'l J Mol Med 8:335-343). Escort proteins are defined as non- infectious retroviral env proteins that accompany the infectious wild type env to provide a gain-in- function-phenotype (i.e., targeting) to the composite vector. These escort proteins display specific ligands or targeting peptides, which essentially replace the deleted receptor binding domains of a modified ecotropic env construct. In previous studies, when a wild-type CAE env (encoding amphotropic 4070 gp70 envelope protein) was co-expressed with this modified env construct bearing a specific targeting motif, the vectors arrayed in the CAE plus "escort" env protein configuration exhibited the desired gain-of-function phenotype (e.g. enhanced viral binding to activated endothelial cells when compared to vectors bearing wild type receptor binding region of ecotropic gp70 protein [CEE] or the wild type amphotropic 4070 gp70 envelope protein [CAE]), as well as high titer amphotropic infectivity. (Hall FL, Liu L, Zhu NL, Stapfer M, Anderson WF, Beart RW, Gordon, EM., Molecular engineering of matrix-targeted retroviral vectors incorporating a surveillance function inherent in von Willebr and factor, Hum Gene Ther 11:983-993 [2000]; Liu et al., (2000) Incorporation of tumor vasculature targeting motifs (TVTMs) into ML V env escort proteins enhances retroviral binding and transduction of human endothelial cells_L J Virol, 74:5320-5328; Masood et al., (2001) Retroviral vectors bearing IgG-binding motifs for antibody-mediated targeting of vascular endothelial growth factor receptors, Int'l J Mol Med 8:335-343001)Liu et al., (2000) Incorporation of tumor vasculature targeting motifs (TVTMs) into ML V env escort proteins enhances retroviral binding and transduction of human endothelial cells_L J Virol, 74:5320-5328).

The utility of this approach is precedented in nature, as many types of wild-type viruses stably express dual envelope configurations, such as hemagglutinin or distinct attachment glycoproteins, in addition to membrane fusion proteins. Hence, these compound configurations may be utilized extensively in vector design to confer auxiliary targeting specificities. The "escort" env protein configurations may be particularly suitable for insertion of large or bulky polypeptides, such as growth factors or molecular bridges into retroviral vectors. (Masood et al., (2001) Retroviral vectors bearing IgG-binding motifs for antibody-mediated targeting of vascular endothelial growth factor receptors, Int'l J Mol Med 8:335-343).

Vascular endothelial growth factor (NEGF), also known as vascular permeability factor, is a secreted selective mitogen and prominent regulator of angiogeneisis and vascular permeability in vivo. (Kolch W, Martiiny-Baron G, Kieser A, Marme D. (1995) Regulation of the expression of the VEGF/VPF and its receptors: role in tumor angiogenesis, Breast Cancer Res Treat 36:139-155; Dvorak, H.F., L.F. Brown, M. Detmar, and A.M. Dvorak. (1995) Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis, Am. J. Pathol. 146: 1029-1039; Keck, P.J., S.D. Hauser, G. Krivi, K. Sanzo, T. Warren, J. Feder, and D.T. Connolly, (1989) Vascular permeability factor, an endothelial cell mitogen related to PDGF, Science 246: 1309- 1312; Connolly., D.T., D.M. Heuvelman, R. Nelson, J.V. Olander, B.L. Eppley, J. J. Delfino, N.R. Siegel, R. M. Leimgruber, and J. Feder, (1989) Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis, J. Clin. Invest. 84: 1470-1478; Thomas, K.A, (1996) Vascular endothelial growth factor, a potent and selective angiogenic agent, J. Biol. Chem. 271: 602- 606; Senger, D.R., S. J. Galli, A. M. Dvorak, CA. Perruzzi, V.S. Harvey, and H. F. Dvorak (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of as cites fluid, Science 219: 983-985; Senger, et al (1986) A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines, Cancer Res. 46: 5629-5632).

Expressed as four major splice variants of 121, 145, 165, 189, and 206 amino acids, respectively, VEGF isoforms differ in the efficiency of secretion and the potency of mitogenic activities, which are specific for vascular endothelial cells. (Houck, KA, Ferrara N, Winer J. (1991) The vascular endothelial growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA, Mol Endocrinol 5:1806-1814; Charnock-Jones, DS, Sharkey AM, Rajput-Williams J. (1993) Identification and localization of alternately spliced mRNAs for vascular endothelial growth factor in human uterus and estrogen regulation in endometrial carcinoma cell lines Biol Reprod 48:1120-1128).

VEGF gene expression is normally quite low in the absence of overt angiogenesis. (Berse, B, Brown LF, Van De Water L. (1992) Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages and tumors, Mol Biol Cell 3:21 1-220).

However, VEGF gene expression is found to be up-regulated both by oncogenic gene mutations and hypoxia present in ischemic tissues and solid tumors. (Senger, D.R., CA. Perruzzi, J. Feder, and H.F. Dvorak (1986) A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines, Cancer Res. 46: 5629-5632; Berse et al., (1992) Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages and tumors, Mol Biol Cell 3:211-220; Gitay-Goren, H, Halaban R, Neufeld G. (1993) Human melanoma cells but not normal melanocytes express vascular endothelial growth factor receptors, Biochem Biophys Res Commun 190:702-708; Brown, LF, Berse B, Jackman RW et al. (1993) Increased expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in kidney and bladder carcinomas, AM J Pathol 143:1255-1262; Berkman, RA, Merrill MJ, Reinhold WC. (1993) Expression of the vascular permeability factor/vascular endothelial growth factor gene in central nervous system neoplasms, J Clin Invest 91:153-159; Boocock, CA, Charnock- Jones DS, Sharkey AM, McLaren J, Barker PJ, Wright KA, Twentyman PR, Smith SK. (1995) Expression of vascular endothelial growth factor and its receptors fit and KDR in ovarian carcinoma, J Natl Cancer Inst 85:506-516; Shweiki,D, Itin A, Soffer D, Keshet E. (1992) Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis, Nature 359:843-845; Liu, Y, Cox SR, Morita T, Kourembanas S. (1995) Hypoxia regulates vascular endothelial growth factor gene expression in endothelial cells. Identification of a 5" enhancer, Circ Res 77:638-643; Mazure, NM, Chen, EY, Yeh, P, Laderoute, KR, Giaccia, AJ, (1996) Oncogenic transformation and hypoxia synergistically act to modulate vascular endothelial growth factor expression, Cancer Res. 56: 3436-3440).

The homodimeric VEGF polypeptides bind to one of two tyrosine kinase receptors: (1) a high affinity receptor (25 pM) receptor designated Flt-1 (de Vries, C, Escobedo JA, Ueno H, Houck KA, Ferrara N, Williams LT, (1992) The fins-like tyrosine kinase, a receptor for vascular endothelial growth factor, Science 255:989-990; Shibuya, M, Yamaguchi S, Yamane A, Ikada T, Tsushime H, Sato M, (1990) Nucleotide sequence and expression of human receptor-type tyrosine kinase (fit) closely related to the fins family, Oncogene 8:519-527), and (2) a lower affinity receptor (125 pM) designated KDR/flk-1 (Terman, Bl, Carrion ME, Kovacs E, Rasmussen BA, Shows TB. (1991) Identification of a new endothelial cell growth factor receptor tyrosine kinase, Oncogene 6:519-524). These tyrosine kinase receptors are selectively expressed in vascular endothelium. (Quinn., T, Peters KG, De Vries C, Ferrara N, Williams LT. (1993) Fetal liver kinase I is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium, Proc Nat'l Acad Sci 90:7533-7537). The KDR/Flk-1 receptor itself is also markedly up-regulated in endothelial cells under hypoxic conditions. (Waltenberger, J., U. Mayr, S. Pentz, and V. Hombach. (1996) Functional upregulation of the vascular endothelial growth factor receptor KDR by hypoxia. Circulation 94: 1647-1654).

VEGF is an important factor in driving the growth, metastasis, and angiogenesis of solid tumors. (Plate, K.H., G. Breier, H.A. Weich, H.D. Mennel, and W. Risau, (1994) Vascular endothelial growth factor and glioma angiogenesis: coordinate induction of VEGF receptors, distribution of VEGF protein and possible in vivo regulatory mechanisms, Int. J. Cancer 59: 520-529; Claffey, K.P., L.F. Brown, L.F. del Aguila, K. Tognazzi, K.T. Yeo, E.J. Manseau, and H.F (1996) Dvorak. Expression of vascular permeability factor/vascular endothelial growth factor by melanoma cells increases tumor growth, angiogenesis, and experimental metastasis, Cancer Res. 56: 172-181). Within the tumor microenvironment, there is a reported up-regulation of both VEGF and its cognate receptor(s) on tumor vascular endothelium. (Brekken et al., (1998) Vascular endothelial growth factor as a marker of tumor endothelium, Cancer Res. 58: 1952-1959). Both oncogenic transformation and hypoxic conditions that are found in most solid tumors act synergistically to modulate VEGF expression. (Mazure et al., (1996) Oncogenic transformation and hypoxia synergistically act to modulate vascular endothelial growth factor expression, Cancer Res. 56: 3436-3440); Forsythe, J.A., B.H. Jiang, N.N. Iyer, F. Agani, S.W. Leung, R.D. Koos, and G.L. Semenza, (1996) Activation of vascular endothelial growth factor gene transcription by hypoxia- inducible factor 1, Mol. Cell Biol. 16: 4604-4613).

Moreover, recent studies of VEGF expression in tumor stromal versus tumor cells have also focused on the importance of stromal fibroblasts as a source of VEGF expression and, hence, a contributor to tumor angiogenesis. (Fukumura, D., R. Xavier, T. Sugiura, Y. Chen, E. Park, Ν. Lu, M. Selig, G. Νeilsen, T. Taksir, R.K. Jain, and B. Seed, (1998) Tumor with VEGF promoter activity in stromal cells, Cell 94, 715-725). Additionally, both bFGF exposure and hypoxia serve to up-regulate the expression of VEGF receptor(s) on endothelial cells. (Pepper, M.S., and S. J. Mandriota (1998) Regulation of vascular endothelial growth factor receptor-2 (Flk-1) expression in vascular endothelial cells, Exp. Cell Res. 241: 414-425; Waltenberger et al., (1996) Functional upregulation of the vascular endothelial growth factor receptor KDR by hypoxia, Circulation 94: 1647-1654; Brogi, E., G. Schatteman, T. Wu, E.A. Kim, L. Varticovski, B. Keyt, J. M. Isner, (1996) Hypoxia-induced paracrine regulation of vascular endothelial growth factor receptor expression, J. Clin. Invest. 97: 469-476).

Thus, the VEGF/receptor complex is a highly specific marker of tumor endothelium. (Dvorak et al., (1991) H.F., J.A. Nagy, A.M. Dvorak, (1991) Structure of solid tumors and their vasculature: implications for therapy with monoclonal antibodies, Cancer Cells 3: 77-85; Ke-Lin et al., (1996) Vascular targeting of solid and ascites tumours with antibodies to vascular endothelial growth factor, Eur. J. Cancer 32A: 2467-2473). This implies that the VEGF/receptor complex can be utilized for the targeting and/or imaging of tumor vasculature. (Brekken et al., (1998) Vascular endothelial growth factor as a marker of tumor endothelium, Cancer Res. 58: 1952-1959). The VEGF receptor complex is among the most specific markers of human tumor-associated vasculature, as Flt-1 and KDR are expressed almost exclusively on endothelial cells. (Mustonen, T, Alitalo K. (1995) Endothelial receptor tyrosine kinases involved in angiogenesis, J Cell Biol 129:895-898). Moreover, up-regulation of both the ligand and the cognate receptor(s) of VEGF are observed within solid tumors, when compared with the endothelium of normal tissues. (Ke-Lin et al., [1996]). Consequently, antibodies directed against VEGF selectively stain tumor blood vessels after injection into tumor-bearing syngeneic mice. (Ke-Lin et al. [1996]). Thus, VEGF expression serves, not only as a specific marker of tumor endothelium (Brekken,R.A., X. Huang, S. W. King, and P.E. Thorpe (1998) Vascular endothelial growth factor as a marker of tumor endothelium, Cancer Res. 58: 1952-1959), but as a potential target for therapeutic intervention in the treatment of neoplastic disease.

Accordingly, neutralizing antibodies, soluble receptor constructs, and antisense strategies have each been shown to block angiogenesis and/or to suppress tumor growth by interference with VEGF signal transduction. (Sioussat, T,M, Dvorak HF, Brock TA, Senger DR (1993) Inhibition of vascular permeability factor (vascular endothelial growth factor) with antipeptide antibodies, Arch Biochem Biophys 301 :15-20; Kondo, S, Asano M, Suzuki H. (1993) Significance of vascular endothelial growth factor /vascular permeability factor for solid tumor growth and its inhibition by the antibody, Biochem Biophys Res Commun 194:1234-1241 ; Asano, M, Yukita A, Matsumoto T, Kondo S, Suzuki H. (1995) Inhibition of tumor growth and metastasis by an immunoneutralizing monoclonal antibody to human vascular endothelial growth factor/vascular permeability factor, Cancer Res 55:5296-5301; Presta, L,G, Chen H, O'Connor SJ, Chisholm V, Meng YG, Krummen L, Winkler M, Ferrara N. (1997) Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders, Cancer Res 57:4593-4599; Kendall, RL,Thomas KA. (1993) Inhibition of vascular endothelial cell growth factor activity by an endogenously encoded soluble receptor, Proc Natl Acad Sci USA 90:10705-10709; Atello, LP, Pierce EA, Foley ED, Takagi H, Chen H, Riddle L, Ferrara N, King GL, Smith LEH, (1995) Suppression of retinal neovascularizaiton in vivo by inhibition of vascular endothelial growth factor (VEGF) using soluble VEGF-receptor chimeric proteins, Proc Natl Acad Sci USA 92:10457-10461; Millauer,B, Longhi MP, Plate KH, Shawyer LK, Risau W, Ullrich A, Strawn LM, (1996) Dominant-negative inhibition of Flk-1 suppresses the growth of many tumor types in vivo, Cancer Res 56:1615-1620; Lin, P, Sankar S, Shan S, Dewhirst MW, Poverini PJ, Quinn TQ, Peters KG, (1998) Inhibition of tumor growth y targeting tumor endothelium using a soluble vascular endothelial growth factor receptor, Cell Growth Differ 9:49058; Saleh, M, Stacker SA, Wilks AF, (1996) Inhibition of growth of C6 gliomas cells in vivo by expression of antisense vascular endothelial growth factor sequence, Cancer Res 56:393-401; Cheng, SY, Huang HJ, Nagane M, Ji XD, Wand D, Shih CCY, Arap W, Huan CM, Cavenee WK, (1996) Suppression of glioblastoma angiogenicity and tumorigenicity by inhibition of endogenous expression of vascular endothelial growth factor, Proc Natl. Acad Sci USA 93:8502-8507).

The requirements for cellular selectivity and vector targeting combine with the emerging biochemical mechanisms of angiogenesis, which accompany tumor growth and metastasis, to advance the concept of targeting of tumor vasculature as a compelling therapeutic strategy. (Dvorak et al., (1991) Structure of solid tumors and their vasculature: implications for therapy with monoclonal antibodies, Cancer Cells 3: 77-85; Hanahan, D., and J. Folkman (1996) Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis, Cell 86, 353-364; Burrows, FJ and Thorpe PE, (1994) Vascular targeting-- a new approach to the therapy of solid tumors, Pharmacol Ther 64:155-174; Fidler, LJ. (1995) Modulation of the organ microenvironment for treatment of cancer metastasis, J. Natl. Cancer Inst. 87,1588-1592; Folkman, J. Antiangiogenic gene therapy, Proc. Natl. Acad. Sci. U.S.A. 95, 9064-9066 [1998]; Tanaka, T., Y. Cao, J. Folkman, and H.A. Fine (1998) Viral vector-targeted antiagniogenic gene therapy utilizing an angiostatin complementary DNA, Cancer Res. 58, 3362-3369; Griscelli, F., H. Li, A. Bennaceur-Griscelli, J. Soria, P. Opolon, C Soria, M. Perricaudet, P. Yeh, and H. Lu, (1998) Angiostatin gene transfer: inhibition of tumor growth in vivo by blockage of endothelial cell proliferation associated with mitosis arrest, Proc. Natl. Acad. Sci. U.S.A. 95, 6367-6372; Bergers, G., K. Javaherian, K.M. Lo, J. Folkman, and D. Hanahan. (1999) Effects of angiogenesis inhibitors on multistage carcinogenesis in mice, Science 248: 808-812).

However, in terms of retroviral vector targeting and prospective gene therapy per se, the development of a targeted injectable vector, which exhibits suitable affinities, selectivity, and stability for application in vivo remains a principal objective. (Anderson, et al., (1998) Human Gene Therapy, Nature (Suppl) 392, 25-30). The present invention addresses this need and provides VEGF-targeted virionoid vectors with cytocidal activity against angiogenesis in neoplastic tissue.

SUMMARY OF THE INVENTION

The present invention is based on an unexpected discovery made while studying a series of constructs comprising VEGF congeners that displayed within the context of Moloney murine leukemia virus (MLN) envelope "escort proteins". Using a 3 or 4 plasmid transient transfection system in human 293T producer cells, these modified envelope proteins were incorporated into a series of MLV- based vectors bearing a β-galactosidase marker gene. The performance of these chimeric retroviral vectors were then evaluated in vitro in terms of targeting and transduction of activated human endothelial cells. Comparative studies revealed critical structural features of the engineered env that favored stable virion incorporation and enhanced cell binding properties, thus identifying a subset of optimal VEGF receptor-targeted vectors. Normalized for equivalent titers, as assayed on NIH 3T3 cells, vectors displaying the VEGF 165 "escort protein" in combination with a wild type amphotropic 4070 gp70 envelope protein (CAE) env induced a three-fold increase in transduction efficiencies of both KSY1 and HUVE cell cultures. These data indicate that these VEGF-targeted vectors have therapeutic potential for targeting gene delivery to tumor-associated vasculature and/or metastatic cancer. Thus, in accordance with the present invention, vascular endothelial cell growth factor/vascular permeability factor (VEGFNPF) receptors selectively expressed on the surface of tumor-activated endothelial cells (EC) provide an advantageous locus for targeting vectors to angiogenic tissues and/or tumor vasculature.

Unexpectedly, paradoxical cytotoxicity was observed when the VEGF-targeted escort proteins were incorporated without a wild-type amphotropic env partner into virions, which thereby bound to endothelial cells without triggering fusion and entry of the retroviral core. The observed cytotoxicity was abolished when a wild-type amphotropic envelope protein was co-expressed with the VEGF- targeted escort construct. Thus, these findings identify a new class of therapeutic viral particles, which we have designated "targeted cytocidal virionoids." These inventive artificial predatory virionoids, lacking a fusogenic envelope, seek (by selective VEGF receptor-mediated targeting) and destroy target cells by contact cytotoxicity in the absence of gene delivery, thereby establishing the prototype of a new breed of engineered viral medicines that are distinguishable from vaccines and classical gene therapy vectors. The inventive targeted cytocidal virionoids have an advantage over retroviral vectors that have been used in the art to target angiogensis, in that they do not deliver exogenous genetic material into the target cell. Thus, their antiangiogenic efficacy does not depend upon the expression of a transgene or the direct manipulation of endogenous gene expression levels in the target cells. Cytotoxicity to endothelial cells, including vascular epithelial cells is induced merely by VEGF receptor-mediated binding of the inventive particles to the cell surface.

In accordance with the present invention a "virionoid" is a non-infectious cytotoxic or cytocidal viral particle that has its envelope artificially modified so that it displays to a specific ligand or targeting peptide that binds to a cell surface receptor target and induces cytotoxicity and/or death without gene delivery to the target cell. The native receptor binding moiety or domain of the viral particle is deleted and replaced with the specific targeting ligand. A most preferred embodiment is a virionoid constructed from a retroviral vector, such as, but not limited to, a lentiviral vector. Other embodiments of the inventive virionoids are constructed from adenoviral vectors, adeno-associated virus vectors, herpes virus vectors (e.g., herpes simplex virus-derived vectors, and pseudotyped viruses. The inventive virionoid can have a core containing a complete native viral genome or an artificially modified genome (e.g., a recombinant nucleic acid construct). Indeed preferred embodiments of the inventive retroviral particles can be constructed employing known techniques, such that they need not even contain nucleic acid (i.e., empty capsids; e.g., Seung SY et al, The 17 nucleotides downstream from the env gene stop codon are important for murine leukemia virus packaging, J Virol 74:8775- 8780 [2000]).

For purposes of the present invention a "VEGF peptide" is an isoform, congener, or fragment of VEGF polypeptide that is at least about 110 contiguous amino acid residues long, or up to the length of a full length native VEGF polypeptide of interest. For example, a useful full-length human VEGF is 165 contiguous amino acid residues long. The useful VEGF peptide is one that will bind the VEGF receptor of interest on the target cell. A preferred embodiment is a human VEGF peptide including a VEGF receptor binding region, useful for binding human vascular endothelial cells, such as those comprised in the neovasculature of human malignant tumors. However, it will be apparent to the skilled artisan that a VEGF peptide can be preselected that binds the VEGF receptor of the mammalian species of interest, such as, but not limited to, a rodent, lagomorph, canine, feline, or non-human primate species. A useful VEGF peptide can also contain one or more amino acid substitutions or deletions, as long as the VEGF peptide still binds specifically to a VEGF receptor of interest, which binding capacity can be determined by routine in vitro screening using cultured vascular endothelial cells originating from a mammalian species of interest.

The present invention also provides a producer cell for producing the inventive VEGF peptide- bearing non-infectious retroviral particle.

The present invention also provides a method of inhibiting angiogenesis in a mammal, including in a human, which is especially useful for targeting the vascular endothelial cells comprised in neovasculature of, for example, malignant tumors, or in proliferative vascularization which develops in retinopathies, such as diabetic retinopathy.

In accordance with the inventive method, the virionoid is administered to a mammalian subject of a preselected (i.e., predetermined) mammalian species of interest, such as a human, the subject being in need of treatment for a tumor, retinopathy, or other lesion involving abnormal proliferative vascularization. A useful virionoid employed in the method, thus, comprises a VEGF peptide on its surface that specifically binds to a VEGF receptor of the preselected mammalian species. For example, a human VEGF peptide is incorporated into the inventive virionoid intended for administration to a human subject; a rabbit or rat VEGF peptide is incorporated into the virionoid intended for administration to a rabbit subject or rat subject, respectively. Alternatively, virionoids comprising a human VEGF peptide can also be used in treating rodents (e.g., mice, rats, guinea pigs, and hamsters) and non-human primates, because their VEGF receptors also specifically bind human VEGF. Whether a particular embodiment of the inventive virionoids can be used in the method for a particular mammalian species, e.g., a virionoid comprising a human VEGF peptide for administering to an animal of any particular other mammalian species, can be determined by prior in vitro screening using appropriate vascular endothelial cells corresponding to the mammalian species of interest.

The virionoid is administered by any suitable means, for example by injection. Injection can be intrarterial, intravenous, intrathecal, intraocular, intramuscular, intraperitoneal, or by direct (e.g., stereotactic) injection into a tumor or other lesion. The virionoid targets and specifically binds to vascular endothelial cells bearing the VEGF receptor.

In accordance with the method, an effective amount of the non-infectious viral particle (virionoid) inhibits the proliferation of new vasculature in a tumor or lesion in the subject compared to the rate of proliferation before treatment. This is determined by routine clinical means, and a clinical outcome such as inhibition of tumor growth can be used as a proxy in determining the effective amount. Preferably, the effective amount is about 1 x 10⁷ to about 1 x 10⁹ virionoid particles per kg body mass, more preferably about 4 x 10 to about 2 x 10 virionoid particles per kg body mass. BRIEF DESCRIPTION OF THE DRAWINGS Molecular Engineering VEGF isoforms into the primary structure of modifled MLV envelope proteins. Figure 1A: Human sequences of various lengths, including VEGF 1 10 ([SEQ ID NO:4], encoding human VEGF peptide having the first 110 amino acid residues [SEQ ID NO:5]), VEGF 121 ([SEQ ID NO:6], encoding human VEGF peptide having the first 121 amino acid residues [SEQ ID NO:7]), and VEGF165 ([SEQ ID NO:8], encoding complete human VEGF peptide [SEQ ID NO:9]), were generated by RT-PCR from a cDNA template, which also added the respective linkers and cloning sites as descriibed in Liu L et al., Incorporation of tumor vasculature targeting motifs (TVTMs) into ML V env escort proteins enhances retroviral binding and transduction of human endothelial cells, J Virol, 74:5320-5328 (2000]) (see also; Wu, BW et al, Characterization of the proline-rich region of murine leukemia virus envelope protein, J Virol 72: 5383-5391 [1998]). Figure IB: The PCR products were cloned in-frame into a strategically modified MLV-based envelope construct designated CEEC-BA as described in Liu et al. (2000) and Wu et al. (1998), to generate envelope escort proteins which were devoid of the ecotropic receptor binding domain. Figure 1C: VEGF165 was also cloned in-frame as N-terminal insertions into a 4070A amphotropic envelope construct designated CAEP-P that was modified to include a unique Pst 1 cloning site as described in Hall et al, Molecular engineering of matrix-targeted retroviral vectors incorporating a surveillance function inherent in von Willebrand factor, Hum Gene Ther 11 :983-993 (2000). Here the respective VEGF domains were flanked by "flexible" linkers to minimize stearic hindrances, and a His residue was added to promote an external conformation of the VEGF targeting domain. In Figure 1 B and 1C, "VEGF 95" identifies nucleotide positions 1-15 of SEQ ID NO:8 (i.e., nucleotide positions 95-109 of SEQ ID NO:3); "VEGF 589" identifies nucleotide positions 481-495 of SEQ ID NO:8 (i.e., nucleotide positions 575-589 of SEQ ID NO:3); "VEGF 424" identifies nucleotide positions 313-330 of SEQ ID NO:4 (i.e., nucleotide positions 407-424 of SEQ ID NO:3). (see Tables 3-6). Expression of the modified envelope proteins in 293T producer cells and incorporation of the envelopes into viral particles. (A,C) The level of expression of the retroviral env protein gp70 and the gag protein p30 in 293T cell lysates of wild type CAE env protein, a chimeric env protein bearing a VEGF isoform (V110-BA; V121-BA, V165-BA; V110-CAEP; V165-CAEP; brackets) was evaluated by Western blotting with and without co-expression with WT CAE env (V110-BA+CAE; V121-BA+CAE; V165-BA+CAE; V110-CAEP+CAE; V165-CAEP+CAE; brackets). (B,D) Comparative levels of env incorporation into retroviral particles bearing either WT CAE env protein, a chimeric env protein a VEGF construct (brackets) with and without WT CAE env protein were examined by Western analysis using anti-gp70 and anti-p30 antibodies.

Fig 3. Binding of retroviral vectors displaying VEGF/env constructs to KSYl Kaposi sarcoma cells. To examine the binding of VEGF-bearing retroviral vectors to activated human endothelial cells in vitro, we utilized human KSYl Kaposi sarcoma endothelial cells (ATCC), which exhibit a constitutive (autocrine) expression of both VEGF and VEGF-receptors. Test vectors were prepared with ecotropic CEE (rodent-specific) envelope partners, which do not by themselves recognize/infect human cells. (A) Binding of viral particles to cell suspensions was monitored by modified ELISA techniques, and (B) quantified by spectrophotometric readings. The results of these binding studies demonstrate high affinity binding of viral particles to KSYl cells (double asterisks ** = p<0.01), which was greater than that of the CEE (ecotropic; negative control) or CAE (amphotropic; positive control) envelope-bearing vectors.

Fig 4. Cytotoxicity of retroviral vectors bearing VEGF env escort constructs. (A:100X;B:400X)Normal morphological appearance of KSYl cells 24 hrs after transduction with a retroviral vector bearing a chimeric VEGF env escort construct (VI 10-BA) co-expressed with WT CAE env. (C,E: 100X; D,F:400X) Morphological appearance of disintegrating KSYl cells 24 hrs after transduction with retroviral vectors bearing only the VI 10-BA and V121-BA env escort constructs respectively. (G:200X) Morphological appearance of normal HUVE cells 24 hrs after transduction with a retroviral vector bearing a chimeric VEGF env escort construct (VI 10-BA) co-expressed with WT CAE env; (H: 200X) Morphological appearance of disintegrating HUVE cells 24 hrs after transduction with retroviral vectors bearing only the VI 10-BA env escort constructs.

Fig.5 (A) Cytocidal activity of targeted VEGF virionoids in KSYl cells. 4 x 10⁵ KSYl cells were transduced with a vector bearing either WT env or a virionoid bearing a VEGF congener. The cell count obtained 24 h after transduction, plotted on the vertical axis, is expressed as a function of type of VEGF congener, plotted on the horizontal axis. (B) Inhibitory activity of targeted VEGF virionoids in KSYl cells. 3 x 10⁴ KSYl cells were transduced with a vector bearing either WT env or a virionoid bearing a VEGF congener. The percent inhibitory activity obtained 24 h after transduction, plotted on the vertical axis, is expressed as a function of type of VEGF congener, plotted on the horizontal axis. Fig.6 (A) Cytocidal activity of targeted VEGF virionoids in HUVE cells. 1.5 x 10⁴ HUVE cells were transduced with a vector bearing either WT env or a virionoid bearing a VEGF congener. The cell count obtained 24 h after transduction, plotted on the vertical axis, is expressed as a function of type of VEGF congener, plotted on the horizontal axis. Fig. 6B Inhibitory activity of targeted VEGF virionoids in HUVE cells. 1.5 x 10⁴ HUNE cells were transduced with a vector bearing either WT env or a virionoid bearing a VEGF congener. The percent inhibitory activity, obtained 24 h after transduction, plotted on the vertical axis, is expressed as a function of type of VEGF congener, plotted on the horizontal axis.

EXAMPLES Example 1. Materials and Methods

Molecular cloning of MLV-based envelope proteins displaying VEGF Congeners (see Figure 1). VEGF coding sequence inserts with cohesive ends were cloned into the CEE+ Δ hinge (ecotropic)-envelope (env) construct (Wu, BW, Cannon PM, Gordon EM, Hall FL, Anderson WF, [1998] Characterization of the proline-rich region of murine leukemia virus envelope protein, J Virol 72: 5383-5391), designated CEEC, which was modified from CEE+ by substitution of a coding sequence for an amphotropic proline rich hinge region (PRR; SEQ ID ΝO:l; see Table 1A), such that at least 90% of the amino acid residues of the receptor binding region of the surface protein have been removed, and the hypervariable polyproline region of the ecotropic gp70 protein is replaced by the hypervariable polyproline region of the amphotropic gp70 protein (SEQ ID NO:2; see Table 2A), the coding sequence of which contains three unique restriction sites (Avr II, Pstl, Stul) and an NgoMl restriction site. (Wu et al. [1998]). The MLV-based env construct was cut with BstEII and Avrll and the linearized env plasmid was verified by restriction analysis on agarose gels and purified by the Gene Clean method (Bio 101, Vista, CA) prior to ligation with the respective VEGF insert and T4 DNA Ligase (New England Biolabs, Beverly, MA) for either 3 hours at room temperature or overnight at 4°C In the resulting "escort" constructs, the VEGF receptor binding domain flanked by glycine linkers replaced the entire receptor binding region of the MLV ecotropic env surface (SU) protein, between the BstEII site at the amino terminus and the Avrll site located proximal to the transmembrane (TM) domain. After ligation, the various constructs of plasmid DNA were transformed into XL1 Blue strain of E. coli and grown on LB agar plates under ampicillin selection. Plasmid DNA was extracted from selected transformed clones using QIAprep Miniprep Kits (Qiagen, Valencia, CA). Each construct was confirmed by enzyme digestion and analysis of the respective inserts, followed by direct DNA sequence analysis. N-terminal insertions of PCR-derived VEGF isoforms into the CAEP (amphotropic) envelope protein previously engineered to incorporate a unique Pst 1 site (Hall et al, (2000) Molecular engineering of matrix-targeted retroviral vectors incorporating a surveillance function inherent in von Willebrand factor, Hum Gene Ther 11 :983-993) were also constructed (see Figure 1) and verified by DNA sequence analysis.

In Figure 1A, nucleotide positions 1-592 of SEQ ID NO:3 (Table 3 below) correspond to nucleotide positions 41-632 of GenBank Accession M32977 (Human heparin-binding vascular endothelial growth factor [VEGF] coding sequence, Leung,D.W. et al, Vascular endothelial growth factor is a secreted angiogenic mitogen, Science 246:1306-1309 [1989]). Nucleotide positions 17-94 of SEQ ID NO:3 (i.e., SEQ ID NO: 10) encode a VEGF signal peptide (SEQ ID NO:l l)(Table 7 below).

Generation of viral particle stocks. Retroviral vectors bearing wild-type (WT) env and/or VEGF- bearing "escort" proteins were assembled using a three-plasmid or four-plasmid transient transfection system as described in Soneoka, Y., P.M. Cannon, E.E. Ramsdale, J.C Griffiths, R. Gaetano, S. M. Kingsman, and A.J. Kingsman (1995) A transient three-plasmid expression system for the production of high titer retroviral vectors, Nucl. Acid Res. 23, 628-633), respectively, depending on whether or not a wild type (WT), amphotropic or ecotropic env protein was co-expressed. The packaging components gag-pol, the WT env, the chimeric env, and a retroviral vector bearing a nuclear-targeted β- galactosidase expression construct were placed on separate plasmids, each containing the SV40 origin of replication. Ten μg of each plasmid were co-transfected by the calcium phosphate method into 293T cells, which express SV40 large T antigen. In particular, the plasmids used were:

(1) pcgp (described and designated pHIT660 in Soneoka, Y., et al., A transient three-plasmid expression system for the production of high titer retroviral vectors, Nucl. Acid Res. 23, 628-633 [1995]; pcgp is a plasmid encoding viral gag-pol);

(2) (i) pCAE (described in Morgan RA et al, Analysis of the functional and host range- determining regions of the murine ecotropic and amphotropic retrovirus envelope proteins, J Virol 67:4712-4721 [1993]; pCAE is a plasmid containing a polynucleotide sequence encoding the wild type amphotropic gp70 env protein), or (ii) pCEE (described in MacKrell, AJ, et al, Identification of a subdomain in the Moloney murine leukemia virus envelope protein involved in receptor binding, J Virol 70:1768-1774 [1996]; pCEE is a plasmid containing a polynucleotide sequence encoding the wild type ecotropic gp70 env protein), or (iii) pCEEC (described in Liu L et al., Incorporation of tumor vasculature targeting motifs (TVTMs) into MLV env escort proteins enhances retroviral binding and transduction of human endothelial cells, J Virol 74:5320-5328, [2000]; pCEEC is a plasmid containing a polynucleotide sequence encoding a retroviral envelope protein which is modified such that at least 90% of the amino acid residues of the receptor binding region of the surface protein have been removed, and the hypervariable polyproline region of the ecotropic gρ70 protein is replaced by the hypervariable polyproline region of the amphotropic gp70 protein [SEQ D NO:2]; see Table 2A);

(3) pESCORT (described in Liu, L et al, Incorporation of tumor vasculature targeting motifs (TVTMs) into MLV env escort proteins enhances retroviral binding and transduction of human endothelial cells, J Virol, 74:5320-5328 [2000]; a plasmid encoding the modified ecotropic env protein (CEEC) and a targeting peptide [in this case, human VEGF or VEGF receptor-binding fragments thereof]; and

(4) pcnBg (a retroviral vector bearing a nucleus targeted β-galactosidase gene; designated pHITl 12 in Soneoka [1995]).

The producer cells were subsequently treated with 10 mM sodium butyrate for 8 to 12 hours to facilitate virionoid production, the medium was changed, and retroviral vector supernatants were harvested at t =24 hours after transfection. The producer cells were designated "293T" (deposited with ATCC [Manasas, VA] as SD-3515).

Viral processing and incorporation of chimeric env proteins into retroviral vectors. The level of expression of the nascent WT e«v proteins gp70 and/or the chimeric env "escort" proteins in 293T cell lysates was evaluated by Western analysis, using a rat monoclonal 83A25 antibody against the C- terminus of the SU domain of gp70, as previously described (Zhu, NL, Cannon PM, Chen D and Anderson WF, (1998) Mutational analysis of the fusion peptide of Moloney murine leukemia virus transmembrane protein pi 5E, J Virol 72:1632-1639). To evaluate env incorporation into virions, viral particles were purified from soluble proteins and cell debris on a 20% sucrose gradient (in PBS), and the virion-associated proteins were subjected to Western analysis using anti-gp70 and anti-p30 antibodies (Zhu et al., (1998) Mutational analysis of the fusion peptide of Moloney murine leukemia virus transmembrane protein pi 5E, J Virol 72:1632-1639). Determination of viral titers. The infectious titers of test retroviral supematants on murine NIH3T3 cells were standardized and quantified based on the expression of a nuclear-targeted β-galactosidase reporter gene (Skotzko, MJ, Wu LT, Anderson WF, Gordon EM, Hall FL (1995) Retroviral vector- mediated gene transfer of antisense cyclin Gl (CYCG1) inhibits proliferation of human osteogenic sarcoma cells, Cancer Research 55:5493-5498), as determined by light microscopy. Briefly, 2.5 x 10⁴ NIH 3T3 cells were plated in each well of 6-well plates one day prior to transduction. Medium was replaced with 1 ml of serial dilutions of the respective retroviral supernatant with 8 μg/ml polybrene for 2 hrs, after which time 1 ml of fresh D10 medium was added to the cultures, which were maintained overnight at 37°C 5% C0₂. The respective cultures were then stained with X-Gal cytostain 48 hrs after transduction to detect the presence of nuclear targeted β-galactosidase as intensely blue-green nuclei. Viral titers were expressed as the number of β-galactosidase positive colony forming units per ml of applied vector supernatant (cfu ml).

Viral binding to human endothelial cells. KSYl Kaposi sarcoma cells (CRL-11448) and HUVE human umbilical cord vascular endothelial cells (CC-2517) were obtained from the American Type Cell Culture Collection (ATCC, Bethesda MD) and Clonetics (San Diego, CA), respectively. For quantification of viral binding, 5 x 10⁶ KSYl or HUVE cells were suspended in RPMI 1640 in a microcentrifuge tube, and were spun down for 15 sec, after which time 1 ml of test vector supernatant was added (viral titers were generally normalized to ~1 x 10⁶ cfu/ml). The mixture was incubated with gentle shaking at room temperature for 30 min. The cells were washed twice with D10 (DMEM+10%FBS) medium, and then resuspended in 300 μl in the presence of a rat monoclonal 83 A25 antibodies directed against the C-terminus of the gp70 MLV env protein (Evans, LH, Morrison FG, Malik J, Portis J and Brittt WJ (1990) A neutralizable epitope common to the envelope glycoprotein of ecotropic, polytropic, xenotropic, and amphotropic murine leukemia viruses, J Virol 64: 6176-6183) and incubated at room temperature for one hr. The cells were again washed twice with D10 medium, and then incubated in 500 μl 1 :2500 HRP-goat anti-rat IgG (Zymed Laboratories Inc.) at RT for 30 min. The cells were washed and then incubated in 500 μl 1 :1000 rat peroxidase anti- peroxidase antibody (Stemberger Monoclonals, Inc.) at room temperature for 30 min. After additional washing, the cells were resuspended in 100 μl TMB single solution (Zymed Laboratories Inc.), and transferred to a 96-well ELISA plate, where the intensity of the color reaction (blue) was read at OD650 nm on a Rainbow Spectra ELISA reader (TECAN US, Inc., NC). Transduction of human endothelial cells. KSYl or HUNE cells were cultured on 1% gelatin-coated dishes in RPMI 1640 supplemented with either 2% (KSYl) or 10% (HUVE) fetal calf serum, 1% sodium pyruvate, 1% essential amino acids, 1% non-essential amino acids, 1 mM glutamine, and 1% penicillin-streptomycin. For transduction experiments, 2 x 10⁴ KSYl or 1.5 x 10⁴ HUVE cells in 3 ml RPMI-2% (2%FBS in RPMI 1640) were plated into each gelatin-coated well in a 6-well plate, and allowed to attach overnight at 37°C. The following morning, medium was replaced with 1 ml RPMI- 2%. The cultures were transduced with 1 ml of each test vector supernatant normalized for equivalent viral titers in the presence of polybrene (8 μg/ml) at 37^DC for 30 minutes. Thereafter, 2.5 ml fresh RPMI-2% was added to the cultures which were incubated overnight at 37^CC Medium was then replaced with fresh medium, and the cultures were further incubated at 37^~C for another 24 hrs. The cells were then stained with X-Gal to visualize the presence of nuclear- targeted β-galactosidase activity under light microscopy. To quantify the resulting transduction efficiency, the number of β -galactosidase positive cells (cells with blue-staining nuclei) was divided by the total cell number per well and expressed as % transduction efficiency.

Cytotoxicity Assays. A standard assay for inhibitory activity in transduced KSYl or HUVE cells was conducted as previously described ( Skotzko et al., (1995) Retroviral vector-mediated gene transfer of antisense cyclin Gl (CYCG1) inhibits proliferation of human osteogenic sarcoma cells, Cancer Research 55:5493-5498). Briefly, either 3 x 10⁴ KSYl cells or 1.5 x 10⁴ HUVE cells were exposed to 1 ml of the test vector or medium as control in the presence of Polybrene (8 μg/ml) for 2 h, with periodic rocking. Then, 1 ml of fresh D10 was added to each well, and the cultures were further incubated at 37°C in a 5% CO₂ incubator. The medium was replaced with 2 ml fresh D10 the next day. To assess the cytotoxic activity of the test vector, the transduced cells were evaluated for their proliferative potential by counting the number of viable cells in triplicate cultures at 24 h after transduction without G418 selection. Cytocidal activity was verified by a comparative decrease in cell number in the test vector-treated cultures compared to control vector- or control medium-treated cultures. The mean cell number in test vector- treated cultures at 24 h was then compared to that of control medium-treated cultures and expressed as % inhibitory activity, using the following formula:

% inhibitory activity = # of cells (medium-treated cultures^') - # of cells (test vector- treated cultures^') X 100

# of cells (medium-treated cultures) Statistical Analysis. The significance of differences among groups was evaluated using the Student t test .

Example 2. Results

Five distinct VEGF-bearing envelope constructs were selected for comparative evaluation (see Figure 1). Upon transient transfection, all of the 5 envelope proteins were expressed well in human 293T retroviral vector producer cells, each exhibiting an apparent molecular mass of about 70 to 80 kDa. As seen in Figure 2 (panel A), the expression of the envelope "escort" proteins was not impaired by co-transfection and co-expression of wild-type envelope proteins, which confer vector tropism and infectivity (i.e., ecotropic, CEE or amphotropic, CAE). Each of the VEGF "escort" proteins could be detected in purified viral particles, however, notable differences in incorporation efficiencies were observed (see Figure 2, panel B). While the VEGFl lO and VEGF121 escort proteins were stably incorporated into viral particles with a stoichiometry equivalent to that of wild type envelopes, the incorporation of the largest VEGF isoform (VEGF 165) was notably lower. In this case, the co- expression of a wild-type CAE envelope was seen to facilitate the incorporation of the modified "escort" protein, presumably due to structural complementation of the tertiary structures (Zhao, Y, Lee S and Anderson WF, (1997) Functional interactions between monomers of the retroviral envelope protein complex, J Virol 71:6967-6972, Anderson, et al., (1998) Human Gene Therapy, Nature (Suppl) 392, 25-30). Constructs bearing VEGF domains cloned as N-terminal insertions into a unique site engineered into the modified amphotropic envelope (CAEP) were expressed well in 293T producer cells (Figure 2C) and were stably incorporated into viral particles in the presence or absence of wild- type envelopes (Figure 2D).

Examination of corresponding viral titers on NIH 3T3 cells, which range from nil (VEGF "escort" envelopes alone) to 2.2 x 10⁶ cfu/ml for the VEGF escort plus CAE env constructs, confirm that the observed fusion and infectivity is provided solely by the co-expression and fusogenic properties of the wild-type (CEE or CAE) envelope partner. In contrast, VEGF insertions into CAEP yielded viral particles of varying infectivity, whereby the particles bearing the smaller congener (VEGFl lO) had markedly reduced infectivity while the larger VEGF isoform (VEGF 165) was moderately to highly infectious (Table 1). This interesting disparity in infectivity suggests that the minimal VEGFl lO construct might in this context interfere with protein folding and/or impose certain stearic hindrances that affect the conformation, hence the biological activity, of the CAEP envelope. Although the infectivity of the vector displaying the VEGF165-CAEP insert env construct was somewhat reduced compared to vector bearing wild type CAE env, this VEGF-bearing env construct, even when expressed alone, was apparently capable of mediating cell fusion and entry of the viral core. To examine the binding properties of the VEGF-bearing retroviral vectors to activated endothelial cells in vitro, we employed human KSYl Kaposi sarcoma endothelial cells, which exhibit a constitutive (autocrine) expression of both VEGF ligand and VEGF-receptors (Masood, R, Cai J, Zheng T, Smith DL, Naidu Y and Gill PS (1997) Vascular endothelial growth factor/vascular permeability factor is an autocrine growth factor for AIDS-Kaposi sarcoma, Proc Nat'l Acad Sci USA 94:979-984), and compared the cell binding ability of the "escort" constructs with that of both the amphotropic (CAE) and ecotropic (CEE) envelopes, the latter of which does not infect human cells. As shown in Figure 3, each of the VEGF-bearing escort constructs exhibited a striking gain of cell binding function. Likewise, the VEGF-bearing insert constructs also exhibited pronounced cell binding affinities, which was greater than that of the CAE (amphotropic) envelope-bearing vectors (data not shown).

As in the cell binding studies described above, test vector supernatants were prepared for cellular transduction studies ~ with the exception that an amphotropic e«v partner (CAE) was utilized to enable the transfection of human cells — and then normalized for equivalency of titer, based on the transduction of NIH 3T3 cells (see Table 1). The results of a representative studies of KSYl and HUVE cell transduction are shown in Table 2. Under these comparative conditions, in which the time of exposure to the test vector was reduced from 120 minutes to 30 min in order to exaggerate the effect of cell targeting, vectors displaying the V165-BA "escort" protein induced a three-fold increase in transduction efficiencies of both KSYl and HUVE cell cultures (Table 2).

While the VEGF-envelope escort constructs alone enhanced endothelial cell binding, these envelopes, by themselves, did not support fusion and entry of the viral core, which requires the co- expression of a wild type envelope (Table 1). Unexpectedly, when some constructs were applied as viral particles to either KSYl or HUVE cells, a paradoxical cytotoxicity was observed, as determined by the appearance of numerous morphologically degenerative and apoptotic cells (Figures 4) and a marked decrease in the over-all number of viable cells (Figures 5A&6A). There was no difference in the viable cell counts between the medium- and the control WT CAE enveloped vector-treated KSYl or HUVE cell cultures. In contrast, single enveloped vectors bearing VEGF-110 and VEGF-121 but not VEGF-165 "escort proteins" demonstrated cytocidal activity in KSYl cells (p< 0.001 and 0.005 respectively), with the VEGF-110 vector exhibiting greater inhibitory activity (67.3 + S.D. 0.94%) than the VEGF-121 vector (48.7 ± 12.28%; p < 0.05; Figure 5B).

More dramatic cytocidal effects were observed in HUVE cell cultures exposed to single enveloped vectors bearing VEGF congeners, as attested by a marked decrease in viable cell counts in cultures exposed to vectors bearing VEGF-110 (pO.OOl), VEGF-121 (pO.OOl) and VEGF-165 (pO.OOl) when compared to those exposed to medium- and control vector bearing WT env (Figure 6A). Unlike KSYl cell cultures, the single enveloped VEGF-165 vector exhibited inhibitory activity in HUVE cell cultures (89 ± 6.53%) that was apparently greater than that of the VEGF-110 vector (79.67 ± 3.86%; p = 0.08) and somewhat less than that of VEGF-121 vector (94.3 + 1.89%; p = 0.16; Figure 6B). Moreover, this overt cytotoxicity was essentially eliminated by the co-expression of an infectious wild-type CAE envelope (Figure 4A & 4B for KSYl cells & 4G for HUNE cells). From these data, we conclude that the VEGF-bearing vector particles bind to endothelial cells via NEGF receptor-mediated mechanisms, and that in the absence of a fusogenic envelope, this physical contact without internalization of the viral particles is inherently cytotoxic. Although the present invention is not dependent on any particular mechanism, the observed contact cytotoxicity could be due to interference with normal NEGF signal transduction events or by disturbance of membrane stability.

Growth factor and/or adhesion receptors that are selectively expressed on surfaces of activated endothelial cells provide an advantageous locus for targeting drugs and gene therapy vectors to angiogenic tissues. Previous studies identified a series of fibronectin-derived ΝGR-bearing congeners that served to enhance the transduction of KSYl cells in vitro (Liu et al., (2000) Incorporation of tumor vasculature targeting motifs (TVTMs) into ML V env escort proteins enhances retroviral binding and transduction of human endothelial cells, J Nirol, 74:5320-5328). The results of the present study demonstrate that minimal VEGF receptor-binding ligands, when expressed in the context of MLV envelope proteins enhance both endothelial cell binding and endothelial cell transduction. Unexpectedly, the single enveloped vectors incorporating all VEGF congeners as escort proteins exhibited profound cytotoxicity in HUVE cell cultures and varying degrees of cytotoxicity in KSYl cell cultures (see Figures 4-6), while no overt cytotoxicity was noted in murine ΝIH3T3 cells. Further, the cytotoxic effects of single enveloped VEGF vectors were abrogated by co-expression of WT CAE env, indicating that the actual binding of the single enveloped VEGF vector in the absence of cell fusion and internalization may itself be deleterious. Moreover, retroviral vectors bearing the non- infectious VEGF/env insert construct (VEGF110-CAEP) were determined to be cytotoxic while vectors bearing the infectious VEGF/env insert construct (VEGF165-CAEP) were not (data not shown). Taken together, the results of these studies identify a new class of retroviral particles, referred to here as "targeted cytocidal virionoids", which exhibit cell-specific selectivity, which function in the absence of gene delivery, and which may exemplify a novel viral-based targeting approach to antiangiogenesis. Therefore, whether the clinical objective is to deliver a therapeutic (cytocidal) construct to tumor vasculature by VEGF-directed retroviral particles or to eliminate the tumor vasculature by contact cytotoxicity (see Figure 4), the present study provides the first proofs of principle that this approach is indeed feasible. The cytocidal virionoids of the present invention can be aimed at translating the cell specific targeting properties of these VEGF-bearing retroviral vectors to effective inhibition of tumorigenesis.

In summary, we utilized an engineering approach to examine the performance of a designed series of VEGF-bearing peptide congeners presented in the context of MLV env "escort" proteins, including strategic linkers and cloning sites, and were determined to be suitable for protein expression, retroviral vector production, and cell-binding kinetics. These VEGF-bearing env "escort" proteins were further demonstrated to function as targeting elements which serve to increase the cell binding affinity and transduction efficiency of the chimeric retroviral vectors, illustrating a potential utility for improving gene delivery in therapeutic angiogenesis and or antiangiogenesis/anticancer strategies. In the course of these studies a new class of retroviral particles, the targeted cytotocidal virionoid, was identified and characterized as a potential therapeutic agent for antiangiogenesis. These predatory virionoids seek (by selective targeting) and destroy target cells (by contact cytotoxicity) in the absence of gene delivery, thereby establishing the prototype of a new class of engineered viral medicines that are distinguishable from vaccines and classical gene therapy vectors.

Table 1

Viral Titers on Murine NIH3T3 Cells Using Targeted Vectors Displaying

VEGF Congeners with or without a WT CAE env Protein

N.D. = Not detected Table 2 Transduction Efficiencies of Targeted Retroviral Vectors Bearing VEGF VPF Congeners in Human Endothelial Cells

Table 1 A. INFORMATION FOR SEQ ID NO: 1

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 180 bases

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: polynucleotide (ix) FEATURE:

(A) NAME/KEY: polynucleotide encoding hypervariable polyproline region of amphotropic gp 70 protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1 :

GGACCCCGAG TCCCCATAGG GCCCAACCCA GTATTACCCG ACCAAAGACT CCCTTCCTCA 60 CCAATAGAGA TTGTACCGGC TCCACAGCCA CCTAGCCCCC TC AATACCAG TTACCCCCCT 120 TCCACTACCA GTACACCCTC AACCTCCCCT ACAAGTCCAA GTGTCCCACA GCCACCCCCA 180

Table 2A. INFORMATION FOR SEQ ID NO: 2: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 60 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: polypeptide (ix) FEATURE:

(A) NAME/KEY: Hypervariable polyproline region of amphotropic gp70 protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: Gly Pro Arg Val Pro He Gly Pro Asn Pro

5 10

Val Leu Pro Asp Gin Arg Leu Pro Ser Ser

15 20

Pro He Glu He Val Pro Ala Pro Gin Pro 25 30

Pro Ser Pro Leu Asn Thr Ser Tyr Pro Pro

35 40

Ser Thr Thr Ser Thr Pro Ser Thr Ser Pro

45 50

Thr Ser Pro Ser Val Pro Gin Pro Pro Pro

55 60 Table 3. Nucleotide positions 1-592 of SEQ ID NO:3. Includes coding sequence of VEGF signal peptide (nt. positions 17-94), VEGF 165 (nt. positions 95-589), and "tga" stop codon at nt. positions 590-592).

tcgggcctcc gaaaccatga actttctgct gtcttgggtg cattggagcc tcgccttgct 60 gctctacctc caccatgcca agtggtccca ggctgcaccc atggcagaag gaggagggca 120 gaatcatcac gaagtggtga agttcatgga tgtctatcag cgcagctact gccatccaat 180 cgagaccctg gtggacatct tccaggagta ccctgatgag atcgagtaca tcttcaagcc 240 atcctgtgtg cccctgatgc gatgcggggg ctgctgcaat gacgagggcc tggagtgtgt 300 gcccactgag gagtccaaca tcaccatgca gattatgcgg atcaaacctc accaaggcca 360 gcacatagga gagatgagct tcctacagca caacaaatgt gaatgcagac caaagaaaga 420 tagagcaaga caagaaaatc cctgtgggcc ttgctcagag cggagaaagc atttgtttgt 480 acaagatccg cagacgtgta aatgttcctg caaaaacaca gactcgcgtt gcaaggcgag 540 gcagcttgag ttaaacgaac gtacttgcag atgtgacaag ccgaggcggt ga 592 SEQ ID NO:3

Table 4. VEGF 110 coding sequence (SEQ DO NO:4) and peptide (SEQ ID NO:5) gca ccc atg gca gaa gga gga ggg cag aat cat cac gaa gtg gtg aag 48 Ala Pro Met Ala Glu Gly Gly Gly Gin Asn His His Glu Val Val Lys 1 5 10 15 ttc atg gat gtc tat cag cgc age tac tgc cat cca ate gag ace ctg 96 Phe Met Asp Val Tyr Gin Arg Ser Tyr Cys His Pro lie Glu Thr Leu 20 25 30 gtg gac ate ttc cag gag tac cet gat gag ate gag tac ate ttc aag 144 Val Asp lie Phe Gin Glu Tyr Pro Asp Glu lie Glu Tyr lie Phe Lys 35 40 45 cca tec tgt gtg ccc ctg atg cga tgc ggg ggc tgc tgc aat gac gag 192 Pro Ser Cys Val Pro Leu Met Arg Cys Gly Gly Cys Cys Asn Asp Glu 50 55 60 ggc ctg gag tgt gtg ccc act gag gag tec aac ate ace atg cag att 240 Gly Leu Glu Cys Val Pro Thr Glu Glu Ser Asn lie Thr Met Gin lie 65 70 75 80 atg egg ate aaa cet cac caa ggc cag cac ata gga gag atg age ttc 288 Met Arg lie Lys Pro His Gin Gly Gin His lie Gly Glu Met Ser Phe

85 90 95 eta cag cac aac aaa tgt gaa tgc aga cca aag aaa gat aga 330 SEQ ID NO: 4 Leu Gin His Asn Lys Cys Glu Cys Arg Pro Lys Lys Asp Arg SEQ ID NO: 5 100 105 110 Table 5. VEGF 121 coding sequence (SEQ ID NO:6) and peptide (SEQ ID NO:7) gca ccc atg gca gaa gga gga ggg cag aat cat cac gaa gtg gtg aag 48

Ala Pro Met Ala Glu Gly Gly Gly Gin Asn His His Glu Val Val Lys

1 5 10 15 ttc atg gat gtc tat cag cgc age tac tgc cat cca ate gag ace ctg 96

Phe Met Asp Val Tyr Gin Arg Ser Tyr Cys His Pro lie Glu Thr Leu

20 25 30 gtg gac ate ttc cag gag tac cet gat gag ate gag tac ate ttc aag 144 Val Asp lie Phe Gin Glu Tyr Pro Asp Glu lie Glu Tyr lie Phe Lys 35 40 45 cca tec tgt gtg ccc ctg atg cga tgc ggg ggc tgc tgc aat gac gag 192 Pro Ser Cys Val Pro Leu Met Arg Cys Gly Gly Cys Cys Asn Asp Glu 50 55 60 ggc ctg gag tgt gtg ccc act gag gag tec aac ate ace atg cag att 240 Gly Leu Glu Cys Val Pro Thr Glu Glu Ser Asn lie Thr Met Gin lie 65 70 75 80 atg egg ate aaa cet cac caa ggc cag cac ata gga gag atg age ttc 288 Met Arg lie Lys Pro His Gin Gly Gin His lie Gly Glu Met Ser Phe

85 90 95 eta cag cac aac aaa tgt gaa tgc aga cca aag aaa gat aga gca aga 336 Leu Gin His Asn Lys Cys Glu Cys Arg Pro Lys Lys Asp Arg Ala Arg 100 105 110 caa gaa aat ccc tgt ggg cet tgc tea 363 SEQ ID NO: 6 Gin Glu Asn Pro Cys Gly Pro Cys Ser SEQ ID NO: 7 115 120

Table 6. VEGF 165 coding sequence (SEQ ID NO:8) and peptide (SEQ ID NO:9) gca ccc atg gca gaa gga gga ggg cag aat cat cac gaa gtg gtg aag 48 Ala Pro Met Ala Glu Gly Gly Gly Gin Asn His His Glu Val Val Lys 1 5 10 15 ttc atg gat gtc tat cag cgc age tac tgc cat cca ate gag ace ctg 96 Phe Met Asp Val Tyr Gin Arg Ser Tyr Cys His Pro lie Glu Thr Leu 20 25 30 gtg gac^' ate ttc cag gag tac cet gat gag ate gag tac ate ttc aag 144 Val Asp lie Phe Gin Glu Tyr Pro Asp Glu lie Glu Tyr lie Phe Lys 35 40 45 cca tec tgt gtg ccc ctg atg cga tgc ggg ggc tgc tgc aat gac gag 192 Pro Ser Cys Val Pro Leu Met Arg Cys Gly Gly Cys Cys Asn Asp Glu 50 55 60 ggc ctg gag tgt gtg ccc act gag gag tec aac ate ace atg cag att 240 Gly Leu Glu Cys Val Pro Thr Glu Glu Ser Asn lie Thr Met Gin lie 65 70 75 80 atg egg ate aaa cet cac caa ggc cag cac ata gga gag atg age ttc 288 Met Arg lie Lys Pro His Gin Gly Gin His lie Gly Glu Met Ser Phe

85 90 95 eta cag cac aac aaa tgt gaa tgc aga cca aag aaa gat aga gca aga 336 Leu Gin His Asn Lys Cys Glu Cys Arg Pro Lys Lys Asp Arg Ala Arg 100 105 110 caa gaa aat ccc tgt ggg cet tgc tea gag egg aga aag cat ttg ttt 384 Gin Glu Asn Pro Cys Gly Pro Cys Ser Glu Arg Arg Lys His Leu Phe 115 120 125 gta caa gat ccg cag acg tgt aaa tgt tec tgc aaa aac aca gac teg 432 Val Gin Asp Pro Gin Thr Cys Lys Cys Ser Cys Lys Asn Thr Asp Ser 130 135 140 cgt tgc aag gcg agg cag ctt gag tta aac gaa cgt act tgc aga tgt 480 Arg Cys Lys Ala Arg Gin Leu Glu Leu Asn Glu Arg Thr Cys Arg Cys 145 150 155 160 gac aag ccg agg egg 495 SEQ ID NO: 8 Asp Lys Pro Arg Arg SEQ ID NO : 9 165 Table 7. Coding sequence (SEQ ID NO: 10)for VEGF signal peptide (SEQ ID NO: 1 1) atg aac ttt ctg ctg tct tgg gtg cat tgg age etc gcc ttg ctg etc 48

Met Asn Phe Leu Leu Ser Trp Val His Trp Ser Leu Ala Leu Leu Leu 1 5 10 15 tac etc cac cat gcc aag tgg tec cag get 78 SEQ ID NO : 10

Tyr Leu His His Ala Lys Trp Ser Gin Ala SEQ ID NO : 11

20 25

Claims

We claim:

1. An artificial VEGF peptide-bearing non-infectious retroviral particle, said particle capable of VEGF receptor-mediated binding to endothelial cells of a preselected species, wherein the viral particle lacks a fusogenic envelope.

2. The non-infectious retroviral particle of Claim 1, wherein the VEGF peptide is a human VEGF peptide.

3. A method of selectively inhibiting tumor angiogenesis in a mammalian subject of preselected mammalian species, comprising:

(a) providing a non-infectious viral particle comprising on its surface a VEGF peptide that specifically binds to a VEGF receptor of the preselected mammalian species, said viral particle lacking a fusogenic envelope; and

(b) administering to the subject an effective amount of the non-infectious viral particle.

4. The method of Claim 3, wherein the delivered particle selectively binds to a VEGF receptor-bearing vascular endothelial cell of the subject without delivery of genetic material from the particle.

5. The method of Claim 4, wherein the VEGF receptor-mediated binding of the viral particle to the vascular endothelial cell is cytotoxic to the cell, thereby selectively inhibiting angiogenesis in a tumor in the subject.

6. The method of Claim 3, wherein the VEGF peptide is a human VEGF peptide.

7. The method of Claim 3, wherein the administering step comprises injection.

8. The method of Claim 3, wherein the effective amount is from about 10⁷ to about 10⁹ of the viral particles per kg body mass.

9. The method of Claim 8, wherein the effective amount is from about 4xl0⁷ to about

2x10 of the viral particles per kg body mass.

10. A producer cell for producing a VEGF peptide-bearing non-infectious retroviral particle.

11. The producer cell of Claim 10, wherein the cell has the characteristics of a cell deposited with ATCC as SD-3515.