Angiostatin fragments and methods of use

ES 2 292 174 T3

DESCRIPTION

Angiostatin fragments and procedures for use.

Field of the invention

The present invention relates to endothelial inhibitors, called angiostatin, which reversibly inhibit endothelial cell proliferation. More particularly, the present invention relates to angiostatin proteins that can be isolated from body fluids such as blood or urine, or that can be synthesized by chemical, enzymatic, or recombinant methods. Angiostatin can inhibit angiogenesis-related diseases and modulate angiogenic processes. Furthermore, the present invention relates to diagnostic assays and kits for the measurement of angiostatin, to histochemical kits for the localization of angiostatin, to DNA sequences that code for angiostatin and to molecular probes to monitor the biosynthesis of angiostatin, to antibodies that are specific for angiostatin, the development of protein agonists and antagonists for the angiostatin receptor, to agonists and antagonists of receptor-specific anti-angiostatin antibodies, and to cytotoxic agents bound to angiostatin proteins.

Background of the invention

As used herein, the term "angiogenesis" means the generation of new blood vessels in a tissue or organ. Under normal physiological conditions, humans or animals only experience angiogenesis in very specific restricted situations. For example, angiogenesis is commonly seen in wound healing, embryonic and fetal development, and formation of the corpus luteum, endometrium, and placenta. The term "endothelium" means a thin layer of flat epithelial cells that cover serous cavities, lymphatic vessels, and blood vessels.

Both controlled and uncontrolled angiogenesis are thought to be performed in a similar manner. Endothelial cells and pericytes, surrounded by a basement membrane, form capillary blood vessels. Angiogenesis begins with erosion of the basement membrane by enzymes released by endothelial cells and leukocytes. The endothelial cells, which line the lumen of the blood vessels, then protrude through the basement membrane. Angiogenic stimulants induce endothelial cells to migrate across the eroded basement membrane. The migrating cells form a "bud" outside the original blood vessel, in which the endothelial cells undergo mitosis and proliferate. The endothelial buds fuse together to form capillary loops, creating the new blood vessel.

Persistent, unregulated angiogenesis occurs in a multiplicity of disease states, tumor metastasis, and abnormal growth through endothelial cells and supports the pathologic damage seen in these states. The various disease states in which unregulated angiogenesis is present have been grouped together as angiogenic dependent or angiogenic associated diseases.

The hypothesis that tumor growth is dependent on angiogenesis was first proposed in 1971. (Folkman J., Tumor angiogenesis: Therapeutic implications., N. Engl. Jour. Med. 285: 1182-1186, 1971). In its simplest terms, it states: "Once the rumor has been 'taken on', each increase in the tumor cell population must be preceded by an increase in new capillaries that converge on the tumor." It is currently understood that tumor "seizure" indicates a prevascular phase of tumor growth in which a population of tumor cells occupying a volume of a few cubic millimeters and not exceeding a few million cells can survive in existing host microvessels. . Expansion of tumor volume beyond this phase requires the induction of new capillary blood vessels. For example, pulmonary micrometastases in the early prevascular phase in mice would be undetectable except by high-power microscopy in histological sections.

Examples of indirect evidence supporting this concept include:

(1) The growth rate of tumors implanted in subcutaneous clear chambers in mice is slow and linear before vascularization, and fast and almost exponential after neovascularization. (Algire GH, et al. Vascular reactions of normal and malignant tumors in vivo. I. Vascular reactions of mice to wounds and to normal and neoplastic transplants. J. Natl. Cancer Inst. 6: 73-85, 1945) (2) The growth of tumors in isolated perfused organs in which the blood vessels do not proliferate is limited to 1-2 mm³ but they rapidly expand to> 1000 times this volume when transplanted into mice and become neovascularized. (Folkman J, et al., Tumor behavior in isolated perfused organs: In vitro growth and metastasis of biopsy material in rabbit thyroid and canine intestinal segments. Annals of Surgery 164: 491-502, 1966) (3) Avascular cornea is performed slowly and at a linear speed, but changes to exponential growth after neovascularization. (Gimbrone, MA, Jr. et al., Tumor growth and neovascularization: An experimental model using the rabbit cornea. J. Natl. Cancer Institute 52: 41-427, 1974)

ES 2 292 174 T3 (4) Tumors suspended in the aqueous fluid of the anterior chamber of the rabbit eye, remain viable, avascular and limited in size to <1 mm³. Once they are implanted in the vascular bed of the iris, they become vascularized and grow rapidly, reaching 16,000 times their original volume within 2 weeks. (Gimbrone MA Jr., et al., Tumor dormancy in vivo by prevention of neovascularization. J. Exp. Med. 136: 261-276) (5) When tumors implant in the chorioallantoic membrane of the chick embryo, they grow slowly during an avascular phase of> 72 hours, but do not exceed a mean diameter of 0.93 ± 0.29 mm. Rapid tumor expansion occurs within 24 hours after the onset of neovascularization, and by day 7 these vascularized tumors reach a mean diameter of 8.0 ± 2.5 mm. (Knighton D., Avascular and vascular phases of tumor growth in the chick embryo. British J. Cancer, 35: 347-356, 1977) (6) Vascular impressions of metastases in rabbit liver reveal heterogeneity in the size of metastases, but show a relatively uniform cut-off point for the size at which the metastasis is present. vascularization. Tumors are generally avascular up to 1 mm in diameter, but are neovascularized beyond that diameter. (Lien W., et al., The blood supply of experimental liver metastases. II. A microcirculatory study of normal and tumor vessels of the liver with the use of perfused silicone rubber. Surgery 68: 334-340, 1970) (7) In transgenic mice that develop carcinomas in the beta cells of the pancreatic islets, the prevascular hyperplastic islets are limited in size to <1 mm. At 6-7 weeks of age, 4-10% of the islets become neovascularized, and from these islets arise large vascularized tumors of more than 1000 times the volume of the prevascular islets. (Folkman J, et al., Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature 339: 58-61, 1989) (8) A specific antibody against VEGF (vascular endothelial growth factor) reduces the density of the microvessels and causes a “significant or drastic” inhibition of the growth of three human tumors that depend on VEGF as its unique mediator of angiogenesis (in nude mice). The antibody does not inhibit tumor cell growth in vitro. (Kim KJ, et al., Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumor growth in vivo. Nature 362: 841-844, 1993) (9) Anti-bFGF monoclonal antibody causes 70% inhibition of growth of a mouse tumor that depends on the secretion of bFGF as its sole mediator of angiogenesis. The antibody does not inhibit tumor cell growth in vitro. (Hori A, et al., Suppression of solid tumor growth by immunoneutralizing monoclonal antibody against human basic fibroblast growth factor. Cancer Research, 51: 6180-6184, 1991) (10) Intraperitoneal injection of bFGF increases the growth of a primary tumor and its metastasis stimulating the growth of capillary endothelial cells in the tumor. Tumor cells themselves lack receptors for bFGF, and bFGF is not a mitogen for tumor cells in vitro. (Gross JL, et al. Modulation of solid tumor growth in vivo by bFGF. Proc. Amer. Assoc. Canc. Res. 31:79, 1990) (11) A specific angiogenesis inhibitor (AGM-1470) inhibits tumor growth and metastasis in vivo, but is much less active in inhibiting tumor cell proliferation in vitro. It inhibits the proliferation of vascular endothelial cells to half the maximum at a concentration lower than 4 logs than it inhibits the proliferation of tumor cells. (Ingber D, et al., Angioinhibins: Synthetic analogues of fumagillin which inhibit angiogenesis and suppress tumor growth. Nature, 48: 555-557, 1990). There is also indirect clinical evidence that tumor growth is dependent on angiogenesis.

(12) Human retinoblastomas that are metastatic to the vitreous develop into avascular spheroids that are restricted to less than 1 mm³ despite the fact that they are viable and incorporate ³Hthymidine (when removed from an enucleated eye and analyzed in vitro).

(13) Ovarian carcinoma metastasizes to the peritoneal membrane as tiny white avascular seeds (1-3 mm³). These implants rarely grow larger until one or more of them becomes neovascularized.

(14) The intensity of neovascularization in breast cancer (Weidner N, et al., Tumor angiogenesis correlates with metastasis in invasive breast carcinoma. N. Engl. J. Med. 324: 1-8, 1991, and Weidner N, et al., Tumor angiogenesis: A new significant and independent prognostic; indicator in early-stage breast carcinoma, J Natl. Cancer Inst. 84: 1875-1887, 1992) and in prostate cancer (Weidner N, Carroll PR, Flax J, Blumenfeld W, Folkman J. Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. American Journal of Pathology, 143 (2): 401-409, 1993) is highly correlated with the risk of future metastasis.

ES 2 292 174 T3 (15) Human cutaneous melanoma metastasis is rare before neovascularization. The beginning of neovascularization leads to an increase in the thickness of the lesion and an increased risk of metastasis. (Srivastava A, et al., The prognostic significance of tumor vascularity in: Intermediate thickness (0.76-4.0 mm thick) skin melanoma. Amer. J. Pathol. 133: 419-423, 1988) (16) In bladder cancer, the urine level of an angiogenic protein, bFGF, is a more sensitive indicator of the status and extent of a disease than is cytology. (Nguyen M, et al., Elevated levels of an angiogenic protein, basic fibroblast growth factor, in urine of bladder cancer patients. J. Natl. Cancer Inst. 85: 241-242, 1993).

WO 95/29242 discloses DNA sequences encoding angiostatin, corresponding to a kringle 1-4 fragment of a plasminogen. No mention is made of the DNA sequences that code for kringle regions 1, 2, 3, 1-2 or 2-3.

Therefore, it is clear that angiogenesis plays an important role in cancer metastasis. If this angiogenic activity could be suppressed or eliminated, then the tumor, although present, would not grow. In the pathological state, the prevention of angiogenesis could prevent the damage caused by the invasion of the new microvascular system. Therapies aimed at controlling angiogenic processes could lead to abrogation or mitigation of these diseases.

Therefore, what is needed is a composition and method that can inhibit the unwanted growth of blood vessels, especially in tumors. A method is also needed to detect, measure and locate the composition. The composition must be able to overcome the activity of endogenous growth factors in premetastatic tumors and prevent the formation of capillaries in tumors thereby inhibiting tumor growth. The composition, fragments of the composition and specific antibodies to the composition must also be able to modulate capillary formation in other angiogenic processes, such as wound healing and reproduction. The composition and method for inhibiting angiogenesis should preferably be non-toxic and produce few side effects. There is also a need for a method to detect, measure and localize the binding sites for the composition as well as biosynthesis sites for the composition. The composition and fragments of the composition must be capable of being conjugated to other molecules for both radioactive and non-radioactive labeling purposes.

Summary of the invention

The present invention is defined by the claims.

In accordance with the present invention, compositions and methods are provided that are effective to modulate angiogenesis, and to inhibit unwanted angiogenesis, especially tumor growth-related angiogenesis. The present invention relates to a protein, which has been called "angiostatin", defined by its ability to overcome the angiogenic activity of endogenous growth factors such as bFGF, in vitro, and by its amino acid sequence homology and structural similarity with an internal part of plasminogen 1 begins at approximately amino acid 98 of plasminogen. Angiostatin comprises a protein having a molecular weight of between approximately 38 kilodaltons and 45 kilodaltons as determined by reductive polyacrylamide gel electrophoresis and having an amino acid sequence substantially similar to that of a murine plasminogen fragment beginning in the amino acid number 98 of an intact murine plasminogen molecule (SEQ ID NO: 2).

The amino acid sequence of angiostatin varies slightly between species. For example, in human angiostatin the amino acid sequence is substantially similar to the sequence of the murine plasminogen fragment described above, although an active human angiostatin sequence can start at either amino acid number 97 or amino acid 99 of an amino acid sequence. of human plasminogen. Furthermore, human plasminogen fragments have similar antiangiogenic activity as shown in a mouse tumor model. It should be understood that the number of amino acids in the active angiostatin molecule can vary and it is contemplated that all amino acid sequences having endothelial inhibitory activity are included in the present invention. The present invention provides a pharmaceutical composition as defined in claim 1-13.

The present invention provides methods and compositions for treating diseases and processes mediated by unwanted and uncontrolled angiogenesis by administering to a human or animal a composition comprising a substantially purified angiostatin, an angiostatin derivative, an angiostatin fragment, or an aggregate of angiostatin in a dosage sufficient to inhibit angiogenesis. The present invention is particularly useful for treating or suppressing tumor growth. Administration of angiostatin to a human or animal with tumors that have undergone prevascularized metastases will prevent the growth or expansion of those tumors.

The present invention also encompasses DNA sequences encoding angiostatin fragments, expression vectors containing DNA sequences encoding angiostatin fragments, and cells containing one or more expression vectors containing DNA sequences encoding angiostatin. The present invention further encompasses gene therapy methods by which DNA sequences encoding angiostatin fragments are introduced into a patient to modify angiostatin levels in vivo.

ES 2 292 174 T3

The present invention also describes diagnostic kits and methods for the detection and measurement of angiostatin in biological fluids and tissues, and for the localization of angiostatin in tissues and cells. The diagnostic kit and method can be in any configuration well known to those of ordinary skill in the art. The present invention also describes antibodies specific to the angiostatin molecule and parts thereof, and antibodies that inhibit the binding of antibodies specific to angiostatin. These antibodies can be polyclonal antibodies or monoclonal antibodies. Antibodies specific for angiostatin can be used in diagnostic kits to detect the presence and amount of angiostatin, which is a diagnosis or prognosis of the existence or recurrence of cancer or other angiogenesis-mediated disease. Angiostatin specific antibodies can also be administered to a human or animal to passively immunize the human or animal against angiostatin, thereby reducing angiogenic inhibition.

The present invention also describes diagnostic kits and methods for detecting the presence and amount of antibodies that bind to angiostatin in body fluids. The diagnostic kit and method can be in any configuration well known to those of ordinary skill in the art.

The present invention also describes receptor-specific anti-angiostatin antibodies that bind to the angiostatin receptor and transmit the appropriate signal to the cell and act as agonists or antagonists.

The present invention also describes angiostatin protein fragments and analogs that can be isotopically labeled or with other molecules or proteins for use in the detection and visualization of angiostatin binding sites with techniques including, but not limited to, tomography by positron emission, autoradiography, flow cytometry, radioreceptor binding assays, and immunohistochemistry.

These proteins and angiostatin analogs also act as agonists and antagonists at the angiostatin receptor, thereby increasing or blocking the biological activity of angiostatin. Such proteins are used in the isolation of the angiostatin receptor.

The present invention also includes angiostatin fragments for therapeutic and research applications. Still further, an angiostatin fragment is combined with pharmaceutically acceptable excipients, and optionally sustained release compounds or compositions, such as biodegradable polymers, to form therapeutic compositions.

The present invention describes molecular probes for ribonucleic acid and deoxyribonucleic acid involved in the transcription and translation of angiostatin. These molecular probes provide a means to detect and measure angiostatin biosynthesis in tissues and cells.

Accordingly, it is an object of the present invention to provide a composition comprising an angiostatin fragment.

It is another object of the present invention to provide a method for treating diseases and processes that are mediated by angiogenesis.

It is still another object of the present invention to provide a rare method and composition to treat diseases and processes that are mediated by angiogenesis including, but not limited to, hemangioma, solid tumors, blood-borne tumors, leukemia, metastasis, telangiectasia, psoriasis, scleroderma. , pyogenic granuloma, myocardial angiogenesis, Crohn's disease, plaque neovascularization, coronary collaterals, cerebral collaterals, arteriovenous malformations, angiogenesis in ischemic extremities, corneal diseases, rubeosis, neovascular glaucoma, diabetic retinopathy, retrolental fibroplasia, arthritis, diabetic neovascularization, macular degeneration, wound healing, gastrointestinal ulcer, Helicobacter-related diseases, fractures, keloids, vasculogenesis, hematopoiesis, ovulation , menstruation, placentation and benign lymphoreticulitis.

It is another object of the present invention to provide a composition for treating or suppressing the growth of cancer.

It is a further object of the present invention to provide angiostatin fragments by direct injection of DNA encoding angiostatin fragments in a human or animal in need of such angiostatin fragments.

It is yet another object of the present invention to provide a cancer therapy that has minimal side effects.

Another object of the present invention is to provide a method for the targeted delivery of angiostatin-related compositions to specific locations.

Still another object of the invention is to provide compositions and methods useful for gene therapy for the modulation of angiogenic processes.

ES 2 292 174 T3

These and other objects, features, and advantages of the present invention will become apparent upon review of the following detailed description of the disclosed embodiments and the appended claims.

Brief description of the figures

Figure 1 shows SEQ ID NO: 1, the amino acid sequence of complete murine plasminogen.

Figure 2 shows the angiostatin initiation sequence for murine (SEQ ID NO: 2) and compares the murine sequence with the corresponding human plasminogen protein fragments (SEQ ID NO: 3), from Rhesus monkey (SEQ ID NO: 4 ), porcine (SEQ ID NO: 5) and bovine (SEQ ID NO: 6). The mouse sequence is listed first, followed by Human, Rhesus, Porcine, and Bovine.

Figure 3 shows the BrdU marking index of tumor cells in the lung in the presence or absence of a primary tumor.

Figure 4 shows a Matrigel analysis of the influence of a primary Lewis lung tumor on bFGF-directed angiogenesis in vivo.

Figure 5 shows the dose response curve for a mouse bearing Lewis lung carcinoma (LLCLOW) versus serum from normal mice. Bovine capillary endothelial cells were tested in a 72 hour bFGF-directed proliferation assay.

Figure 6 shows that both high and low metastatic tumors contain endothelial mitogenic activity in their ascites, but not only the low metastatic tumor line has serum endothelial inhibitory activity.

Figure 7 shows a C4 reverse phase chromatographic profile of partially purified serum or urine from tumor bearing animals.

Figure 8 shows superficial lung metastasis after 13-day treatment of mice with intact human plasminogen plasminogen molecule, preparation of active fraction of a lysine-binding site I, concentrated urine from tumor-bearing mice, and concentrated urine from normal mice. .

Figure 9 shows lung weight after 13-day treatment of mice with intact human plasminogen plasminogen molecule, lysine-binding site I active fraction preparation, concentrated urine from tumor-bearing mice, and concentrated urine from normal mice. .

Figure 10 is a schematic representation of the vector pTrcHis.

Figure 11 depicts an immunoblot of E. coli expressed human angiostatin from a 10 L scaled-up fermentation, probed with monoclonal antibody to the kringle 1-3 region of human plasminogen. The arrow shows recombinant human angiostatin. A) shows recombinant angiostatin eluted with 0.2M aminocaproic acid; B) shows the last 1 X PBS wash of the lysine column; and C) shows clarified lysate of fractured cells.

Figure 12 is a graph depicting the percent inhibition of growing bovine capillary endothelial cells as a function of dilution of the stock solution; A1, A2, B1, B2, and E are recombinant clones expressing human angiostatin anti-angiogenesis activity; Controls C1, C2, D1 and D2 are negative control clones containing only vector without the human DNA sequence encoding angiostatin.

Figure 13 shows the inhibitory effect on proliferation of recombinant human angiostatin on bovine capillary endothelial cells in vitro.

Figure 14 shows the growth proliferation index and apoptotic index after removal of the primary tumor and treatment with saline or a fumagillin analog with anti-angiogenic activity.

Figure 15 shows the inhibition of the growth of a primary T241 tumor in mice by treatment with human angiostatin in vivo with a single injection of 40 mg / kg / day.

Figure 16 shows the inhibition of the growth of a primary LLC-LM tumor in mice by treatment with human angiostatin in vivo with two doses of 40 mg / kg per dose (80 mg / kg / day).

Figure 17 shows the effect of removal of a primary Lewis lung carcinoma tumor on the growth of its lung metastasis.

Figure 18 shows the proliferation of growth and apoptotic index after tumor resection.

Figure 19 shows the effect of administration of angiostatin protein to mice having implanted T241 fibrosarcoma cells on total tumor volume as a function of time.

ES 2 292 174 T3

Figure 20 shows the effect of administration of angiostatin protein to mice having implanted Lewis lung carcinoma (LM) cells on total tumor volume as a function of time.

Figure 21 shows the effect of administration of angiostatin protein to mice implanted with cells of the reticulum cellular sarcoma on total tumor volume as a function of time.

Figure 22 shows the effect of administration of angiostatin protein to immunodeficient SCID mice implanted with implanted human prostate carcinoma PC-3 cells on total tumor volume as a function of time over a 24-day period.

Figure 23 shows the effect of administration of angiostatin protein to immunodeficient SCID mice implanted with implanted human breast carcinoma MDA-MB cells on total tumor volume as a function of time over a period of 24 days.

Figure 24 is a schematic representation of the cloning of the mouse DNA sequence encoding mouse angiostatin protein derived from mouse plasminogen cDNA. Mouse angiostatin spans the kringle 1-4 regions of mouse plasminogen. PCR stands for polymerase chain reaction; P1 is the 5 'end oligonucleotide primer for PCR; P2 is the 3 'end oligonucleotide primer for PCR; SS calls the signal sequence; ATG is the translation initiation codon; TAA is the translation stop codon; HA represents the hemagglutinin epitope tag (YPYDVPDYASL); K1, K2, K3 and K4 represent kringle regions 1, 2, 3 and 4 of mouse plasminogen respectively. CMV is the promoter of cytomegalovirus; T7 is the bacteriophage promoter; PA represents proteins prior to activation; and SP6 is the promoter of Sp 6.

Figure 25 represents cell number as a function of days for untransfected cells (assay); cells transfected only with the vector, without the DNA sequence encoding angiostatin (Vector 5), and two clones expressing angiostatin (AST 31 and AST 37). Panel (a) represents the results of T241 cell transfection. Panel (b) represents the results of LL2 cells.

Figure 26 shows the results of the culture medium derived from E. coli cells containing the angiostatin clone on cell number. Non-transfected cells (assay); cells transfected only with the vector, without the DNA sequence coding for angiostatin (Vector 5) and three clones expressing angiostatin (AST 25, AST 31 and AST 37). Panel (a) represents the results of the incubation of the culture medium from control clones (test) and all angiostatin (expressing and non-expressing) in cell number. Panel (b) represents the results of the culture medium incubation of control (assay), vector only (vector 6) and angiostatin clones expressing mouse angiostatin in cell number. Panel (c) represents the results of the incubation of purified culture medium for control (assay) and angiostatin clones expressing mouse angiostatin in cell number, in which the culture medium was purified on a lysine column. sepharose to yield lysine binding components.

Figure 27 shows the effect on total tumor volume as a function of implantation time of T241 fibrosarcoma cells in mice, in which fibrosarcoma cells have been transfected with a vector containing a DNA sequence encoding protein from angiostatin, and wherein the vector can express angiostatin protein. "Non-transfected" represents unaltered T241 fibrosarcoma cells implanted in mice. "Vector 6" represents T241 fibrosarcoma cells transfected only with the vector, which does not contain the DNA sequence encoding angiostatin protein, implanted in mice. "Clone 25, Clone 31 and Clone 37" represent three angiostatin-producing clones of T241 fibrosarcoma cells transfected with a vector containing the DNA sequence encoding angiostatin protein implanted in mice.

Figure 28 shows a schematic representation of the structure of human plasminogen and its kringle fragments. Human plasminogen is a single chain protein containing 791 amino acids with an N-glycosylation side attached to Asn²⁸⁹. The non-protease binding region of human plasminogen consisting of the N-terminal 561 amino acids that exist in five separate domains, called kringles as shown in the circles (K1, K2, K3, K4, and K5), along with the proteins that separate these structures. Each triple disulfide bridged kringle contains 80 amino acids. Angiostatin covers the first 4 of these domains kringle (K1-4), kringle 1-3 (K1-3) and kringle 4 (K4) are obtained by digestion of human plasminogen with elastase. The rest of the kringle fragments are recombinant proteins expressed in E. coli. SS = signal sequence. PA = pre-activation protein.

Figure 29 shows SDS-PAGE analysis of purified native and recombinant kringle fragments from plasminogen under reducing conditions. (A) Purified individual recombinant kringle fragments from E. coli bacterial lysates were loaded onto a 15% SDS gel followed by Coomassie blue staining. Approximately 5 pg of each protein were loaded per lane (lane 2 = kringle 1 (K1); lane 3 = kringle 2 (K2); lane 4 = kringle 3 (K3); lane 5 = kringle 4 (K4); lane 1 = molecular weight markers). (B) Large fragments of kringle were stained with Coomassie blue. Kringles 1-4 (lane 2) and kringles 1-3 (lane 3) were obtained by digestion of human plasminogen with elastase and purified by lysine-sepharose chromatography. The recombinant fragment of kringles 2-3 (lane 4) was expressed in E. coli and refolded in vitro. Molecular weight markers are indicated on the left (lane 1).

ES 2 292 174 T3

Figure 30 shows an inhibition of endothelial cell proliferation by recombinant individual kringle fragments of angiostatin. Kringle fragments were assayed in bovine capillary endothelial cells in the presence of 1 ng / ml bFGF for 72 hours. (A) Anti-endothelial cell proliferative effects of two lysine-binding kringles, rK1 and rK4. The high affinity lysine-binding kringle, K1 (-O-), inhibited BCE cell proliferation in a dose-dependent manner. The intermediate affinity lysine-binding kringle, K4 (- · -), showed only little inhibitory effect at high concentrations. (B) Inhibition of non-BCE cell proliferation by binding to lysine K2 and K3. Both K2 (-) and K3 (- □ -) inhibited BCE cell proliferation in a dose-dependent manner. Data represent mean +/- SEM of triplicates.

Figure 31 shows an anti-endothelial proliferation activity of large angiostatin kringle fragments. The proteolytic fragments, K1-4 (angiostatin) (-O-) and K1-3 (-), inhibited BCE cell proliferation in a dose-dependent manner. Recombinant K2-3 (- · -) fragments showed less potent inhibition than K1-3 and K1-4. The data represent the mean of three determinations (+/- SEM) as percentages of inhibition.

Figure 32 shows additive inhibitory activity of recombinant kringle 2 and kringle 3. (A) The intact rK2-3 fragment (see also Figure 31) showed a weak inhibitory effect only at the 320 nM concentration. At the same concentration, additive inhibition was observed when mutant cysteine ​​fragments of rK2 were substituted for serine at position 169) and K3 (cysteine ​​substituted for serine at position 297) were tested together in bCe cells. Each value represents the mean +/- SEM of triplicates. (B) Schematic structure and amino acid sequence of K2 and K3. A disulfide bridge between kringle chains was previously reported to be present between cysteine¹⁶⁹ of K2 and the cistern²⁹⁷ de K3 (Sohndel, S., Hu, C.-K., Marti, D., Affolter, M., Schaller, J., Llinas, M., and Rickli, EE (1996) Biochem. in press).

Figure 33 shows an inhibition of endothelial proliferation by combinatorial kringle fragments. The assay was performed at a concentration of 320 nM for each kringle fragment. Fragments represent the mean of three determinations (+/- SEM) as percentages of inhibition. (A) Inhibitory effects of fragments by combining several individual kringles. (B) Combinatorial inhibitory activity of combined kringle fragments.

Figure 34 shows an inhibitory activity of angiostatin on endothelial cells after reduction and alkylation. (A) SDS-PAGE analysis of the reduced (lane 2) and non-reduced (lane 1) forms of human angiostatin. Purified human angiostatin was reduced with DTT followed by alkylation of the protein with an excess amount of iodoacetamide. Treated samples were dialyzed and assayed in BCE cells. (B) Inhibition of BCE cell proliferation by reduced and non-reduced forms of angiostatin at a concentration of 320 nM. Data represent mean +/- SEM inhibition of tripilicates.

Figure 35 shows an amino acid sequence alignment of possible human angiostatin kringle domains. The sequences of four kringle domains were aligned according to their conserved cysteines. Identical and conserved amino acids are shaded. Amino acids boxed in kringle 4 show the positively charged double lysines adjacent to conserved cysteine ​​residues 22 and 80.

Figure 36 shows lysine binding characteristics and reactivity of expressed angiostatin.

Figure 36A shows a Coomassie stained gel (40 µl load).

Figure 36B shows an immunoblot (20 µl load) of a similar gel. Lane: 1 sample broth from shake flasks from induced cultures showing angiostatin protein at approximately 50 kD and a few other proteins. The induced culture broth is diluted 1: 1 with buffer and loaded directly onto lysine-sepharose. Lane: 2 shows the unbound fraction that passed through the lysine column. All the angiostatin protein expressed by P. pastoris binds to the lysine column. Lane: 3 shows specific elution with 0.2 M aminocaproic acid showing that the angiostathira protein expressed by P. pastoris binds lysine and can be purified in a single step to obtain homogeneity on lysine-sepharose. In addition, the angiostatin protein expressed by P. pastoris is recognized by a conformationally dependent monoclonal antibody (VAP) grown against kringles 1 to 3.

Figure 37 shows angiostatin protein expressed by P. pastoris observed as a doublet migrating at 49 kD and 51.5 kD in denatured non-reduced Coomassie SDS-PAGE stained gels. Removal of the single N-linked complex chain from the angiostatin protein expressed with N-glycanase specific for mannose-rich structures results in a single 49.5 kD band. Panel A and panel B show a Coomassie stained gel and a similar gel Western blot respectively. Lane: 1 shows a purified P. pastoris expressed angiostatin protein. Lane: 2 shows a purified P. pastoris expressed angiostatin protein incubated under N-glycanase-free digestion conditions. Lane: 3 shows purified N-glycanase-digested P. pastoris expressed angiostatin protein.

Figure 38A shows 4 pg of angiostatin protein expressed by P. pastoris purified as a doublet on a Coomassie gel.

ES 2 292 174 T3

Figure 38B shows that the purified recombinant inhibits the proliferation of BCE. The BCE assay cell count obtained after 72 hours is shown, in the presence (·) or absence (o) of bFGF, and in the presence of bFGF with PBS as a control (Δ), and in the presence of bFGF with angiostatin protein expressed by P. pastoris (Δ).

Figure 38C shows that the inhibition is dose dependent.

Figure 39 shows that purified angiostatin expressed by P. pastoris was administered systemically (subcutaneously) to mice with primary tumors.

Figures 39A and B show the number of metastases and lung weights respectively of mice treated daily with saline or angiostatin expressed by P. pastoris or with plasminogen-derived angiostatin protein. In contrast to the lungs of mice treated with saline, the lungs of mice treated with angiostatin protein expressed by P. pastoris or plasminogen-derived angiostatin protein were not vascularized and metastasis was potently suppressed.

Figure 40 shows that the lungs of mice treated with angiostatin expressed by P. pastoris were pink with micrometastases while the lungs of the saline control group were fully covered with vascularized metastases.

Figure 41 shows a photograph of reducing polyacrylamide gel electrophoresis (SDS-PAGE) of recombinant mouse angiostatin at various stages of purification and aggregation on nickel affinity column chromatography (lanes 1-3 inclusion body washes, lane 4 urea insoluble fraction, lane 5 starting materials on affinity column, lane 6 flow through column, lane 7 eluant angiostatin).

Figure 42 shows the effect of administering aggregated recombinant mouse angiostatin on bovine capillary endothelial cells.

Figure 43 shows the effect on tumor volume of administering 2 mg / kg / day of added angiostatin in mice inoculated with Lewis lung carcinoma.

Figure 44 shows the effect on tumor volume of administering 10 mg / kg / day of added angiostatin in mice inoculated with Lewis lung carcinoma.

Detailed description

The present invention describes compositions and methods for the treatment of diseases and processes that are mediated by or associated with angiogenesis. The composition is an angiostatin fragment or a combination of angiostatin fragments that can be isolated from body fluids including, but not limited to, serum, urine and ascites, or synthesized by chemical or biological methods (e.g. cell culture, recombinant gene expression, protein synthesis and in vitro enzymatic catalysis of plasminogen or plasmin to yield active angiostatin). Recombinant techniques include gene amplification from DNA sources using polymerase chain reaction (PCR), and gene amplification from RNA sources using reverse transcriptase / PCR. Angiostatin inhibits the growth of blood vessels in tissues such as non-vascularized or vascularized tumors.

The present invention encompasses a composition comprising a vector containing a DNA sequence encoding an angiostatin fragment or a combination of angiostatin fragments wherein the vector can express an angiostatin fragment or a combination of angiostatin fragments when is present in a cell, a composition comprising a cell containing a vector, wherein the vector contains a DNA sequence encoding angiostatin fragments or analogs thereof, and wherein the vector can express an angiostatin fragment or a combination of angiostatin fragments when present in the cell, and a method comprising, implanting in a human or non-human animal a cell containing a vector, wherein the vector contains a DNA sequence encoding an angiostatin fragment or a combination of angiostatin fragments, and wherein the vector can express an angiostatin fragment or a combination of angiostatin fragments when present in the cell .

Still further, the present invention encompasses angiostatin fragments that are combined with pharmaceutically acceptable excipients, and optionally sustained release compounds or compositions, such as biodegradable polymers, to form therapeutic compositions. Furthermore, the invention describes a composition comprising an antibody that specifically binds to angiostatin, wherein the antibody does not bind to plasminogen.

More particularly, the present invention describes a protein called angiostatin having a molecular weight of about 38 to 45 kilodaltons (kD) that can overcome the angiogenic activity of endogenous growth factors such as bFGF, in vitro. Angiostatin is a protein that has a molecular weight of between approximately 38 kilodaltons and 45 kilodaltons as determined by reducing polyacrylamide gel electrophoresis and has an amino acid sequence substantially similar to that of a murine plasminogen fragment beginning in the amino acid number 98 of an intact murine plasminogen molecule. The term “sus9

"ES 2 292 174 T3", when used in reference to the amino acid sequences of angiostatin, means an amino acid sequence that has anti-angiogenic activity and has a molecular weight of about 38 kD to 45 kD that also has a high degree of sequence homology to the mouse plasminogen protein fragment starting at approximately amino acid number 98 in mouse plasminogen and weighing 38 kD to 45 kD. A high degree of homology means amino acid homology of at least about 60%, desirably amino acid homology of at least about 70%, and more desirably amino acid homology of at least about 80%. The term "endothelial inhibitory activity" as used herein means the ability of a molecule to inhibit angiogenesis in general and, for example, to inhibit the growth of bovine capillary endothelial cells in a culture in the presence of the growth factor. fibroblast.

The amino acid sequence of the complete murine plasminogen molecule is shown in Figure 1 and SEQ ID NO: 1. The sequence for angiostatin begins at approximately amino acid 98. Active human angiostatin can begin at either amino acid 97 or 99 of the intact human plasminogen molecule. Figure 2 shows the amino acid sequence of the first 339 amino acids of mouse angiostatin, (SEQ ID NO: 2), and it is compared with the sequences of the corresponding protein fragments of human plasminogen plasminogen (SEQ ID NO: 3 ), Rhesus monkey (SEQ ID NO: 4), porcine (SEQ ID NO: 5) and bovine (SEQ ID NO: 6). Since these sequences are identical in more than 50% of their amino acids, it should be understood that the amino acid sequence of angiostatin is substantially similar between species. The total number of amino acids in angiostatin is not precisely known but is defined as the molecular weight of the active molecule. The amino acid sequence of angiostatin of the present invention may vary depending on which species the plasminogen molecule is derived. Thus, although the angiostatin of the present invention which is derived from human plasminogen has a slightly different sequence from mouse derived angiostatin, it has anti-angiogenic activity as shown in a mouse tumor model.

Angiostatin has been shown to inhibit endothelial cell growth in vitro. Angiostatin does not inhibit the growth of cell lines derived from other cell types. Specifically, angiostatin has no effect on Lewis lung carcinoma cell lines, mink lung epithelium, 3T3 fibroblasts, bovine aortic smooth muscle cells, bovine retinal pigment epithelium, MDCk cells (canine renal epithelium), WI38 cells (human fetal lung fibroblasts), EFN cells (murine fetal fibroblasts), and LM cells (murine connective tissue). Endogenous angiostatin in a tumor-bearing mouse is effective in inhibiting metastasis at a systemic concentration of approximately 10 mg angiostatin / kg body weight.

Angiostatin has a three-dimensional conformation that is defined by the kringle region of the plasminogen molecule. (Robbins, KC, “The palsminogen-plasmin enzyme system” Hemostasis and Thrombosis, Basic Principles and Practice, 2^to Edition, ed. by Colman, RW et al. JB Lippincott Company, pages 340-357, 1987). There are five such kringle regions, which are conformationally related motifs and have substantial sequence homology, in the Nh part.₂ terminal of the plasminogen molecule. The three-dimensional conformation of angiostatin is believed to encompass kringle regions 1 to 3 and a portion of the kringle 4 region of plasminogen. Each kringle region of the plasminogen molecule contains approximately 80 amino acids and contains 3 disulfide bridges. This cysteine ​​motif is known to exist in other biologically active proteins. These proteins include, but are not limited to, prothrombin, hepatocyte growth factor, scattering factor, and macrophage stimulating protein. (Yoshimura, T, et al., "Cloning, sequencing, and expression of human macrophage stimulating protein (MSP, MST1) confirms MSP as a member of the family of kringle proteins and locates the MSP gene on Chromosome 3" J. Biol. Chem, Vol. 268, No. 21, pp. 15461-15468, 1993). Any isolated protein or protein having a three-dimensional kringle-like conformation or cysteine ​​motif that has anti-angiogenic activity in vivo is contemplated as part of the present invention.

The present invention also describes the detection of angiostatin in body fluids and tissues for the purpose of diagnosis or prognosis of diseases such as cancer. The present invention also describes the detection of angiostatin binding sites and receptors in cells and tissues. The present invention also includes methods of treating or preventing angiogenic diseases and processes including, but not limited to, arthritis and tumors by stimulating the production of angiostatin, and / or administering angiostatin fragments to a patient. Additional treatment methods include the administration of angiostatin fragments, bound to cytotoxic agents. It should be understood that angiostatin can be of animal or human origin. Angiostatin fragments can also be produced synthetically by chemical reaction or by recombinant techniques in conjunction with expression systems. Angiostatin fragments can also be produced by enzymatically cleaving plasmin or isolated plasminogen to generate proteins that have anti-angiogenic activity. Angiostatin can also be produced by compounds that mimic the action of endogenous enzymes that cleave plasminogen to give angiostatin. Angiostatin production can also be modulated by compounds that affect the activity of enzymes that cleave plasminogen.

Passive antibody therapy can be employed using antibodies that specifically bind to angiostatin to modulate angiogenesis-dependent processes such as reproduction, wound development and healing, and tissue repair. In addition, antisera directed to the Fab regions of antibodies to angiostatin can be administered to block the ability of endogenous angiostatin antisera to bind to angiostatin.

ES 2 292 174 T3

The present invention also encompasses gene therapy by which the gene encoding angiostatin is regulated in a patient. Various methods for transferring or administering DNA to cells for expression of gene product protein, otherwise referred to as gene therapy, are disclosed in Gene Transfer into Mammalian Somatic Cells in vivo, N. Yang, Crit. Rev. Biotechn. 12 (4): 335-356 (1992), which is incorporated herein by reference. Gene therapy encompasses the incorporation of DNA sequences into somatic cells or non-human germ line cells in therapy either ex vivo or in vivo. Gene therapy works to stop, replace genes, increase normal or abnormal gene function, and fight infectious diseases and other pathologies.

Strategies for treating these medical problems with gene therapy include therapeutic strategies such as identifying the defective gene and then adding a functional gene to either replace the function of the defective gene or to augment a slightly functional gene; or prophylactic strategies, such as adding a gene for the product protein that will treat the condition or make the tissue or organ more susceptible to a treatment regimen. As an example of a prophylactic strategy, a gene such as angiostatin can be placed in a patient and thus prevent the production of angiogenesis; or a gene could be inserted that makes tumor cells more susceptible to radiation and then irradiation of the tumor would cause increased tumor cell killing.

Many protocols for the transfer of angiostatin DNA or angiostatin regulatory sequences are envisioned in this invention. Also envisioned as gene therapy methods are the transfection of promoter sequences, other than one commonly found specifically associated with angiostatin, or other sequences that could increase angiostatin protein production. An example of this technology is found in Transkaryotic Therapies, Inc., of Cambridge, Massachusetts, which uses homologous recombination to insert a "genetic switch" that is turned on in an erythropoietin gene in cells. See Genetic Engineering News, April 15, 1994. Such "genetic switches" could be used to activate angiostatin (or the angiostatin receptor) in cells that do not normally express angiostatin (or the angiostatin receptor).

Gene transfer methods for gene therapy fall into three broad categories, physical (e.g., electroporation, direct gene transfer, and particle bombardment), chemical (lipid-based vehicles, or other non-viral vectors), and biological (vector derived from virus and receptor uptake). For example, non-viral vectors including DNA-coated liposomes can be used. Such liposome / DNA complexes can be directly injected intravenously into the patient. Liposome / DNA complexes are believed to concentrate in the liver where they deliver DNA to macrophages and Kupffer cells. These cells are long-lived and therefore provide long-term expression of the administered DNA. Additionally, vectors or "naked" gene DNA can be injected directly into the desired organ, tissue, or tumor for targeted delivery of the therapeutic DNA.

Gene therapy methodologies can also be described by site of administration. Fundamental pathways for delivering genes include ex vivo gene transfer, in vivo gene transfer, and in vitro gene transfer. In ex vivo gene transfer, cells are taken from the patient and grown in culture. The DNA is transfected into the cells, the transfected cells are expanded in number, and then reimplanted into the patient. In in vitro gene transfer, the transformed cells are cells that grow in culture, such as tissue culture cells, and not particular cells from a particular patient. These "laboratory cells" are transfected, the transfected cells are selected and expanded either for implantation in a patient or for other uses.

In vivo gene transfer involves introducing the DNA into the patient's cells when the cells are within the patient. Methods include using virally mediated gene transfer using a non-infectious virus to deliver the gene into the patient or injecting naked DNA into a site in the patient and the DNA is taken up by a percentage of cells in which the gene product protein is expressed. . Additionally, the other methods described herein, such as the use of a "gene gun," can be used for in vitro insertion of angiostatin DNA or angiostatin regulatory sequences.

Chemical gene therapy methods can involve a lipid-based compound, not necessarily a liposome, transporting the DNA across the cell membrane. Lipofectins or cytofectins, lipid-based positive ions that bind to negatively charged DNA, form a complex that can cross the cell membrane and provide DNA inside the cell. Another chemical method uses receptor-based endocytosis, which involves binding a specific ligand to a cell surface receptor and enveloping and transporting it across the cell membrane. The ligand binds to DNA and the entire complex is transported into the cell. The ligand gene complex is injected into the bloodstream and then the target cells bearing the receptor will specifically bind to the ligand and transport the ligand-DNA complex into the cell.

Many gene therapy methodologies employ viral vectors to insert genes into cells. For example, altered retrovirus vectors have been used in ex vivo methods to introduce genes into peripheral lymphocytes and infiltrate tumors, hepatocytes, epidermal cells, myocytes, or other somatic cells. These altered cells are then introduced into the patient to provide the gene product from the inserted DNA.

Viral vectors have also been used to insert genes into cells using in vivo protocols. To direct tissue-specific expression of foreign genes, cis-acting promoters or regulatory elements known to be tissue-specific can be used. Alternatively, this can be accomplished using in situ delivery of DNA or viral vectors to specific anatomical sites in vivo. For example, gene transfer to vessels was achieved

ES 2 292 174 T3 blood cells in vivo by implanting in vitro transduced endothelial cells at chosen sites in arterial walls. The virus infected surrounding cells that also expressed the gene product. A viral vector can be delivered directly to the site in vivo, via a catheter for example, thus allowing only certain areas to be infected by the virus, and providing long-term, site-specific gene expression. In vivo gene transfer using retrovirus vectors has also been demonstrated in mammalian tissue and liver tissue by injection of altered virus into blood vessels leading to organs.

Viral vectors that have been used for gene therapy protocols include, but are not limited to, retroviruses, other RNA viruses such as poliovirus or Sindbis virus, adenovizus, adeno-associated virus, herpes virus, SV 40, vaccinia, and other DNA viruses. . Replication-defective murine retroviral vectors are the most widely used gene transfer vectors. Murine leukemia retroviruses are composed of a single-stranded RNA complexed with a nuclear nucleus protein and polymerase enzymes (pol), covered by a protein nucleus (gag) and surrounded by a glycoprotein envelope (env) that determines the interval guest. The cenomic structure of retroviruses includes the gag, pol and env genes enclosed by the long terminal repeats (LTRs) at 5 'and 3'. Retroviral vector systems exploit the fact that a minimal vector containing the 5 'and 3' LTRs and the packaging signal are sufficient to allow packaging, infection and integration of the vector into target cells as long as the viral structural proteins are delivered in trans in the packaging cell line. The fundamental advantages of retroviral vectors for gene transfer include efficient infection and gene expression in most cell types, precise single copy vector integration into target cell chromosomal DNA, and ease of manipulation of the retroviral genome.

Adenovirus is composed of linear, double-stranded DNA complexed with core proteins and surrounded by capsid proteins. Advantages in molecular virology have led to the ability to exploit the biology of these organisms to produce vectors that can transduce novel genetic sequences into target cells in vivo. Adenovirus-based vectors will express gene product proteins at high levels. Adenoviral vectors have high infectivity efficiencies, even with low virus titers. Additionally, the virus is fully infectious as a cell-free virion so injection of producer cell lines is not necessary. Another possible advantage for adenoviral vectors is the ability to achieve long-term expression of heterologous genes in vivo.

Mechanical methods of DNA delivery include fusogenic lipid vesicles such as liposomes or other membrane fusion vesicles, DNA lipid particles incorporating cationic lipids such as lipofectin, polylysine-mediated DNA transfer, direct injection of DNA, such as microinjection of DNA in germ or somatic cells, pneumatically administered DNA-coated particles, such as the gold particles used in a "gene gun" and inorganic chemical approaches such as calcium phosphate transfection. Another method, ligand-mediated gene therapy, involves complexing DNA with specific ligands to form ligand-DNA conjugates, to target DNA to a specific cell or tissue.

Injection of plasmid DNA into muscle cells has been found to yield a high percentage of cells that are transfected and have sustained expression of marker genes. The plasmid DNA may or may not integrate into the genome of the cells. Non-integration of transfected DNA would allow the transfection and expression of gene product proteins in non-proliferative, terminally differentiated tissues for a prolonged period of time without fear of mutational insertions, deletions, or alterations in the cellular or mitochondrial genome. Long-term, but not necessarily permanent, transfer of therapeutic genes into specific cells can provide treatments for genetic diseases or for prophylactic use. The DNA could be periodically reinjected to maintain the level of gene product without mutations occurring in the genomes of the recipient cells. Non-integration of exogenous DNA can allow the presence of several different exogenous DNA constructs within a cell with all constructs expressing various gene products.

Particle-mediated gene transfer methods were first used to transform plant tissues. With a particle bombardment device, or "gene gun," a driving force is generated to accelerate high-density DNA-coated particles (such as gold or tungsten) to a high speed that allows penetration of organs, tissues or target cells. Particle bombardment can be used in in vitro systems, or with ex vivo or in vivo techniques to introduce DNA into cells, tissues or organs.

Electroporation for gene transfer uses an electrical current to render cells or tissues susceptible to electroporation-mediated gene transfer. A brief electrical impulse with a given field strength is used to increase the permeability of a membrane such that DNA molecules can penetrate cells. This technique can be used for in vitro systems, or with ex vivo or in vivo techniques to introduce DNA into cells, tissues or organs.

In vivo vehicle-mediated gene transfer can be used to transfect foreign DNA into cells. The vehicle-DNA complex can be conveniently introduced into body fluids or the bloodstream and then specifically targeted to the target organ or tissue in the body. Both liposomes and polycations can be used, such as polylysine, lipofectins or cytofectyria. Liposomes that are cell specific or organ specific can be developed and therefore foreign DNA carried by the liposome will be taken up by target cells. Injection of immunoliposomes that are targeted to a specific receptor in certain cells can be used as a convenient method of inserting the DNA into cells that carry the receptor. Other vehicle system that has been used12

ES 2 292 174 T3 do is the conjugated asialoglycoprotein / polylysine system for transporting DNA to hepatocytes for in vivo gene transfer.

The transfected DNA can also be complexed with other types of vehicles such that the DNA is transported to the recipient cell and then resides in the cytoplasm or nucleoplasm. DNA can be coupled to carrier nuclear proteins in genetically engineered vesicle complexes and transported directly to the nucleus.

Angiostatin gene regulation can be achieved by administering compounds that bind to the angiostatin gene, or control regions associated with the angiostatin gene, or its corresponding RNA transcript to modify the rate of transcription or translation. Additionally, cells transfected with a DNA sequence encoding angiostatin can be administered to a patient to provide a source of angiostatin in vivo. For example, cells can be transfected with a vector containing a nucleic acid sequence encoding angiostatin. The term "vector" as used herein means a carrier that can contain or be associated with specific nucleic acid sequences, which functions to transport the specific nucleic acid sequences into a cell. Examples of vectors include plasmids and infectious microorganisms such as viruses, or non-viral vectors such as ligand-DNA conjugates, liposomes, lipid-DNA complexes. It may be desirable that a recombinant DNA molecule comprising an angiostatin DNA sequence be operably linked to an expression control sequence to form an expression vector that can express angiostatin. The transfected cells may be cells derived from normal tissue from a patient, diseased tissue from a patient, or may not be cells from a patient.

For example, tumor cells taken from a patient can be transfected with a vector that can express the angiostatin protein of the present invention, and reintroduced into the patient. The transfected tumor cells produce levels of angiostatin in the patient that inhibit tumor growth. Patients can be human or non-human animals. Cells can also be transfected by chemical, physical, or non-vector methods known in the art such as electroporation, ionoporation, or by a "gene gun." Additionally, angiostatin DNA can be injected directly, without the aid of a vehicle, into a patient. In particular, angiostatin DNA can be injected into the skin, muscle, or blood.

The gene therapy protocol for transfecting angiostatin in a patient can be either by integrating angiostatin DNA into the genome of cells, in minichromosomes, or as a DNA construct that does not replicate or replicates separately in the cytoplasm. or nucleoplasm of the cell. Angiostatin expression can continue for a long period of time or it can be periodically re-injected to maintain a desired level of angiostatin protein in the cell, tissue or organ or a given blood level.

Angiostatin can be isolated on a C4 HPLC column (see Table 3). Angiostatin protein is eluted at 30 to 35% in an acetonitrile gradient. On a sodium dodecyl sulfate polyacrylamide gel electrophoresis (PAGE) gel under reducing conditions, the protein band with activity eluted as a single peak at approximately 38 kilodaltons.

The inventors have shown that a growing primary tumor is associated with the release into the bloodstream of specific inhibitor (s) of endothelial cell proliferation, including angiostatin which can suppress anigiogenesis in a metastasis and thereby inhibit the growth of the metastasis itself. The source of the angiostatin associated with the primary tumor is unknown. The compound can be produced by degradation of plasminogen by a specific protease, or angiostatin could be produced by expression of a specific gene encoding angiostatin.

The angiogenic phenotype of a primary tumor depends on the production of angiogenic proteins in excess of cellular endothelial inhibitors that are made by normal cells, but are believed to be down-regulated during transformation to neoplasia. While angiostatin production may be down-regulated in an individual tumor cell relative to production by its parent cell type, the total amount of inhibitor made by the entire tumor may be sufficient to enter the circulation and suppress growth. endothelial cells at remote sites of micrometastasis. Angiostatin remains in the circulation for a significantly longer time than the angiogenic protein (s) released by a primary tumor. Thus, it appears that angiogenic proteins act locally, whereas angiostatin acts globally and circulates in the blood with a relatively long half-life. The half-life of angiostatin is approximately 12 hours to 5 days.

Although not wishing to be bound by the following hypothesis, it is believed that when a tumor becomes angiogenic it releases one or more angiogenic proteins (eg, aFGF, bFGF, VEGF, IL-8, GN-CSF, etc.), which act in a globally, target the endothelium in the vicinity of a primary tumor in an extravascular direction, and do not circulate (or circulate with a short half-life). These angiogenic proteins must be produced in an amount sufficient to overcome the action of the endothelial cell inhibitor (angiogenesis inhibitors) for a primary tumor to continue to expand its population. Once such a primary tumor is growing well, it continues to release cellular endothelial inhibitors into the circulation. According to this hypothesis, these inhibitors act remotely at a distance from the primary tumor, target the capillary endothelium of a metastasis from an intravascular direction, and continue to circulate. Thus, just at the time that a remote metastasis can begin to initiate angiogenesis, the capillary endothelium in its vicinity could be inhibited by incoming angiostatin.

ES 2 292 174 T3

Once a primary tumor has reached sufficient size to cause angiostatin to be continuously released into the circulation, it is difficult for a second tumor implant (or a micrometastasis) to initiate or increase its own angiogenesis. If a second tumor implant (for example, in the subcutaneous space, or in the cornea, or intravenously to the lung) occurs just after the primary tumor is implanted, the primary tumor will not be able to suppress the secondary tumor (because the angiogenesis in the secondary tumor will already be underway). If two tumors are implanted simultaneously (eg, on opposite flanks), the inhibitors can have an equivalent inhibitory effect on each other.

The angiostatin fragments of the present invention can:

(i) Administered to humans or animals bearing tumors as anti-angiogenic therapy;

(ii) Monitored in human or animal serum, urine, or tissues as prognostic markers; and (iii) Be used as the basis for analyzing serum and urine of cancer patients for similar angiostatic molecules.

It is contemplated as part of the present invention that angiostatin can be isolated from a body fluid such as blood or urine of patients or angiostatin can be produced by recombinant DNA methods or synthetic protein chemical methods well known to those skilled in the art. technique. Protein purification methods are well known in the art and a specific example of a method for purifying angiostatin, and testing for inhibitor activity is provided in the examples below. Isolation of endogenous human angiostatin is accomplished using similar techniques.

An example of a method for producing angiostatin using recombinant DNA techniques involves the steps of (1) identifying and purifying angiostatin as discussed above, and as described in more detail below, (2) determining the amino acid sequence N- terminal of the purified inhibitor, (3) synthetically generate 5 'and 3' DNA oligonucleotide primers for the angiostatin sequence, (4) amplify the angiostatin gene sequence using polymerase, (5) insert the amplified sequence into an appropriate vector such as an expression vector, (6) insert the vector containing the gene into a microorganism or other expression system that can express the inhibitor gene, and (7) isolate the inhibitor produced recombinantly. Suitable vectors include viral, bacterial, and eukaryotic (such as yeast) expression vectors. The above techniques are described in more detail in laboratory manuals such as "Molecular Cloning: A Laboratory Manual" Second Edition by Sambrook et al., Cold Spring Harbor Press, 1989. Human plasminogen DNA sequence has been published (Browne, MJ, et al., "Expression of recombinant human palsminogen and a glycoplasminogen in HeLa cells" Fibrinolysis Vol.5 (4). 257-260, 1991) and is incorporated herein by reference.

The gene for angiostatin can also be isolated from cells or tissue (such as tumor cells) that express high levels of angiostatin (1) by isolating messenger RNA from the tissue, (2) using reverse transcriptase to generate the corresponding DNA sequence and then ( 3) using polymerase chain reaction (PCR) with appropriate primers to amplify the DNA sequence encoding the active angiostatin amino acid sequence.

Still another method of producing angiostatin, or biologically active fragments thereof, is by protein synthesis. Once a biologically active fragment of an angiostatin is found using the assay system described in more detail below, it can be sequenced by, for example, automated protein sequencing methods. Alternatively, once the gene or DNA sequence encoding angiostatin is isolated, for example by the methods described above, the DNA sequence can be determined using manual or automated sequencing methods well known in the art. The nucleic acid sequence in turn provides information related to the amino acid sequence. Thus, if the biologically active fragment is generated by specific methods, such as tryptic digestion, or if the N-terminal fragment is sequenced, the remaining amino acid sequence can be determined from the corresponding DNA sequence.

Once the amino acid sequence of the protein is known, the fragment can be synthesized by techniques well known in the art, as exemplified by "Solid Phase Protein Synthesis: A Practical Approach" E. Atherton and RC Sheppard, IRL Press, Oxford , England. Similarly, multiple fragments can be synthesized which are subsequently joined together to form larger fragments. These synthetic protein fragments can also be prepared with amino acid substitutions at specific locations for testing for agonist and antagonist activity in vitro and in vivo. Protein fragments having high affinity binding to tissues can be used to isolate the angiostatin receptor on affinity columns. Isolation and purification of the angiostatin receptor is a critical step in elucidating the mechanism of action of angiostatin. Isolation of an angiostatin receptor and identification of angiostatin agonists and antagonists will facilitate the development of drugs to modulate angiostatin receptor activity, the final route to determining biological activity. Receptor isolation allows the construction of nucleotide probes to monitor receptor location and synthesis, using solution and in situ hybridization technology. In addition, the gene for the angiostatin receptor can be isolated, incorporated into an expression vector, and transfected into cells, such as tumor cells from a patient to increase the ability of a type of cell, tissue, or tumor to bind angiostatin and inhibit angiogenesis. local.

ES 2 292 174 T3

Angiostatin is effective in treating diseases or processes that are mediated by, or involve, angiogenesis. The present invention includes the method of treating angiogenesis-mediated disease with an effective amount of biologically active angiostatin fragments, or combinations of angiostatin fragments that simultaneously possess anti-angiogenic activity. Diseases beset by angiogenesis include, but are not limited to, solid tumors; blood-borne tumors such as leukemias; tumor metastasis; benign tumors, for example hemangiomas, acoustic neuromas, neurofibromas, trachomas and pyogenic granulomas; rheumatoid arthritis; psoriasis; angiogenic eye diseases, eg, diabetic retinopathy, retrolental fibroplasty, macular degeneration, corneal graft rejection, neovascular glaucoma, retrolental fibroplasia, rubeosis; Osler-Webber syndrome; myocardial angiogenesis; plaque neovascularization; telangiectasia; hemophilic joints; angiofibroma; and wound granulation. Angiostatin is useful in the treatment of abnormal or excessive stimulation of endothelial cells. These diseases include, but are not limited to, intestinal adhesions, Crohn's disease, atherosclerosis, scleroderma, and hypertrophic scars, ie, keloids. Angiostatin can be used as a birth control agent by preventing the vascularization required for implantation of the embryo. Angiostatin is useful in the treatment of diseases that have angiogenesis as a pathological consequence such as benign lymphoreticulitis (Róchele minalia quintosa) and ulcers (Helicobacter pylori).

Synthetic protein fragments of angiostatin have a variety of uses. Protein that binds to the angiostatin receptor with high specificity and avidity is radiolabelled and used for visualization and quantification of binding sites using membrane binding and autoradiographic techniques. This application provides important diagnostic and research tools. Knowledge of the binding properties of the angiostatin receptor facilitates investigation of the receptor-bound transduction mechanism.

Furthermore, labeling of angiostatin proteins with short-lived isotopes allows visualization of receptor binding sites in vivo using positron emission tomography or other modern radiographic techniques to localize tumors with angiostatin binding sites.

Systematic amino acid substitution in these synthesized proteins yields high-affinity protein agonists and antagonists for the angiostatin receptor that increase or decrease the binding of angiostatin to its receptor. Such agonists are used to suppress the growth of micrometastases, thus limiting the spread of cancer. Angiostatin antagonists are applied in situations of inadequate vascularization, to block the inhibitory effects of angiostatin and promote angiogenesis. For example, this treatment may have therapeutic effects to promote wound healing in diabetics.

Angiostatin proteins are used to develop affinity columns for the isolation of the angiostatin receptor from cultured tumor cells. Isolation and purification of the angiostatin receptor are followed by amino acid sequencing. Using this information, the gene or genes encoding the angiostatin receptor can be identified and isolated. The cloned nucleic acid sequences are then disclosed for insertion into vectors capable of expressing the receptor. These techniques are well known to those skilled in the art. Transfection of the nucleic acid sequence (s) encoding the angiostatin receptor in tumor cells, and the expression of the receptor by the transfected tumor cells enhances the responsiveness of these cells to endogenous or exogenous angiostatin and thus decreases the metastatic growth rate.

Cytotoxic agents such as ricin are bound to angiostatin, and high affinity angiostatin protein fragments, thus providing a tool for killing cells that bind to angiostatin. These cells can be found in many locations, including but not limited to micrometastases and primary tumors. Proteins bound to cytotoxic agents are infused in a manner designed to maximize delivery to the desired location. For example, ricin-bound high affinity angiostatin fragments are delivered via cannula to vessels providing the target site or directly on the target. Such agents are also administered in a controlled manner through osmotic pumps coupled to infusion cannulas. A combination of angiostatin antagonists with angiogenesis stimulators can be co-applied to increase tissue vascularity. This therapeutic regimen provides an effective means of destroying metastatic cancer.

Angiostatin fragments can be used in combination with other compositions and methods for treating disease. For example, a tumor can be treated conventionally with surgery, radiation, or chemotherapy combined with angiostatin fragments and then the angiostatin fragments can then be administered to the patient to extend the latency of micrometastasis and to stabilize and inhibit the growth of any primary tumor. residual. Additionally, angiostatin fragments or combinations thereof are combined with pharmaceutically acceptable excipients, and optionally sustained release matrices, such as biodegradable polymers, to form therapeutic compositions.

A sustained release matrix, as used herein, is a matrix made of materials, usually polymers, that can be degraded by enzymatic or acid / base hydrolysis or by dissolution. Once inserted into the body, the matrix is ​​activated by enzymes and body fluids. The sustained release matrix is ​​desirably chosen from biocompatible materials such as liposomes, polylactides (polylactic acid), polyglycolides (glycolic acid polymer), polylactide co-glycolide (copolymers of lactic acid and glycolic acid), polyanhydrides, poly (ortho) esters, polyproteins, hyaluronic acid, collagen, chondroitin sulfate, carboxylic acids, fatty acids, phospholipids, polysaccharides, nucleic acids, polyamino acids, amino acids such as phenylalanine, tyrosine, isoleucine, polynucleotides, polyvinylpropylene, polyvinylpyrrolidone, and silicone. A bio matrix15

Preferred degradable ES 2 292 174 T3 is a matrix of either polylactide, polyglycolide, or polylactide co-glycolide (copolymers of lactic acid and glycolic acid).

The therapeutic composition that modulates angiogenesis of the present invention can be a solid, liquid or aerosol and can be administered by any known route of administration. Examples of solid therapeutic compositions include pills, creams, and implantable dosage units. Pills can be administered orally, therapeutic creams can be administered topically. Implantable dosage units can be administered locally, for example at a tumor site, or implantable for systemic delivery of the angiogenesis modulating composition, for example subcutaneously. Examples of liquid composition include formulations adapted for injection subcutaneously, intravenously, intraarterially, and formulations for topical and intraocular administration. Examples of aerosol formulation include inhaler formulation for administration to the lungs.

The angiostatin fragments of the present invention can also be used to generate antibodies that are specific for the inhibitor and its receptor. The antibodies can be either polyclonal antibodies or monoclonal antibodies. These antibodies that specifically bind to angiostatin or angiostatin receptors can be used in diagnostic kits and methods that are well known to those of skill in the art to detect or quantify angiostatin or angiostatin receptors in a body fluid or tissue. The results of these tests can be used to diagnose or predict the occurrence or recurrence of cancer and other angiogenesis-mediated diseases.

Angiostatin fragments can also be used in a diagnostic kit and method to detect and quantify antibodies that can bind to angiostatin. These kits would allow the detection of circulating anti-angiostatin antibodies indicating the spread of micrometastasis in the presence of angiostatin secreted by primary tumors in situ. Patients who have such circulating anti-angiostatin antibodies are more susceptible to developing multiple tumors and cancers, they are more susceptible to recurrence of cancer after treatment or periods of remission. Fab fragments of these anti-angiostatin antibodies can be used as antigens to generate anti-angiostatin Fab fragment antiserum that can be used to neutralize anti-angiostatin antibodies. Such a method would reduce the clearance of circulating angiostatin by anti-angiostatin antibodies, thus effectively raising circulating angiostatin levels.

The present invention describes a method of blocking the action of excess endogenous angiostatin. This can be done by passively immunizing a human or animal with antibodies specific for unwanted angiostatin in the system. This treatment may be important in treating abnormal ovulation, menstruation and placentation and vasculogenesis. This provides a useful tool to examine the effects of angiostatin removal on metastatic processes. The Fab fragment of antibodies to angiostatin contains the binding site for angiostatin. This fragment is isolated from antibodies to angiostatin using techniques known to those of skill in the art. Angiostatin antiserum Fab fragments are used as antigens to generate serum production of anti-Fab fragments. Infusion of this antiserum against angiostatin Fab fragments prevents angiostatin from binding to anti-angiostatin antibodies. Therapeutic benefit is obtained by neutralizing endogenous anti-angiostatin antibodies by blocking the binding of angiostatin to antiangiostatin Fab fragments. The net effect of this treatment is to facilitate the ability of endogenous circulating angiostatin to reach target cells, thus slowing the spread of metastasis.

It should be understood that the present invention is contemplated to include derivatives of angiostatin fragments that have endothelial inhibitory activity. The present invention includes derivatives of the biologically active fragments of the angiostatin protein. These include proteins with angiostatin activity that have amino acid substitutions or have sugars or other molecules attached to functional groups of amino acids.

Protein fragments with angiostatin activity described above can be provided as substantially purified and isolated proteins and protein fragments in pharmaceutically acceptable formulations using formulation methods known to those of skill in the art. These formulations can be administered by conventional routes. In general, the combinations can be administered topically, transdermally, intraperitoneally, intracranially, intracerebroventricular, intracerebral, intravaginal, intrauterine, oral, rectal, or parenteral (eg, intravenous, intraspinal, subcutaneous, or intramuscular). In addition, Angiostatin Fragments can be incorporated into biodegradable polymers allowing sustained release of the compound, the polymers being implanted in the vicinity of where drug delivery is desired, for example, at the site of a tumor, or implanted in such a way that it is released. angiostatin slowly systemically. Osmotic minipumps can also be used to provide controlled delivery of high concentrations of angiostatin through cannulas to the site of interest, such as directly into a metastatic growth or vascular supply to that tumor. Biodegradable polymers and their uses are described, for example, in detail in Brem et al., J. Neurosurg. 74: 441-446 (1991), which is incorporated herein by reference in its entirety.

The dosage of the angiostatin fragments of the present invention will depend on the disease state or condition being treated and on other clinical factors such as the weight and condition of the human or animal and on the route of administration of the compound. To treat humans or animals, about 0.5 mg / kilogram to 500 mg / kilogram of the angiostatin fragments can be administered. Depending on the half-life of the angiostatin fragments in the particular animal or human, the angiostatin fragments can be administered between

ES 2 292 174 T3 several times a day and once a week. It should be understood that the present invention has application for both human and veterinary use. The methods of the present invention contemplate single as well as multiple administrations, administered either simultaneously or over an extended period of time.

Angiostatin fragment formulations include those suitable for oral, rectal, ophthalmic (including intravitreal or intracameral), nasal, topical (including buccal and sublingual), intrauterine, vaginal, or parenteral (including subcutaneous, intraperitoneal, intramuscular, intravenous, intradermal) administration. , intracranial, intratracheal and epidural). Angiostatin fragment formulations can be conveniently presented in unit dosage form and can be prepared by conventional pharmaceutical techniques. Such techniques include the step of combining the active ingredient and the pharmaceutical carrier (s) or excipient (s). In general, formulations are prepared by uniformly and intimately associating the active ingredient with liquid carriers or finely divided solid carriers or both, and then, if necessary, shaping the product.

Formulations suitable for parenteral administration include aqueous and non-aqueous sterile injection solutions that may contain antioxidants, buffers, bacteriostats, and solutes that render the formulation isotonic with the blood of the desired recipient; and aqueous and non-aqueous sterile suspensions which may include suspending and thickening agents. The formulations may be presented in unit dose or multiple dose containers, eg, sealed ampoules and vials, and may be stored in a freeze-dried (lyophilized) state that only requires the addition of the sterile liquid carrier, eg, water for injections, immediately before use. Improvised injection solutions and suspensions can be prepared from sterile powders, granules and tablets of the type described above.

Preferred unit dosage formulations are those containing a daily unit or dose, daily subdose, or an appropriate fraction thereof, of the administered component. It should be understood that in addition to the components, particularly mentioned above, the formulations of the present invention may include other agents conventional in the art that are related to the type of formulation in question. Optionally, cytotoxic agents may be incorporated or otherwise combined with angiostatin proteins, or biologically functional protein fragments thereof, to provide dual therapy to the patient.

The proteins that inhibit angiogenesis of the present invention can be synthesized in a conventional microchemical facility and verified for purity with HPLC and mass spectrometry. Those skilled in these techniques are commonly aware of the methods of protein synthesis, HPLC purification, and mass spectrometry. Angiostatin proteins and angiostatin receptor proteins are also produced in yeast or E. coli, and purified by column chromatography.

Different protein fragments of the intact angiostatin molecule can be synthesized for use in various applications including, but not limited to the following; as antigens for the development of specific antisera, as active agonists and antagonists at angiostatin binding sites, as proteins to bind to, or used in combination with, cytotoxic agents for the targeted killing of angiostatin-binding cells. The amino acid sequences that comprise these proteins are selected based on their position in the outer regions of the molecule and are accessible for binding to antisera. The amino and carboxy terminus of angiostatin are represented separately, as well as the middle region of the molecule between the fragments to be synthesized.

These protein sequences are compared to known sequences using protein sequence databases such as GenBank, Brookhaven Protein, SWISS-PROT, and PIR to determine possible sequence homologies. This information facilitates the elimination of sequences that show a high degree of sequence homology with other molecules, thus enhancing the potential for high specificity in the development of antisera, agonists and antagonists against angiostatin.

Angiostatin and angiostatin-derived proteins can be coupled to other molecules using conventional methods. The amino and carboxyl-termini of angiostatin contain tyrosine and lysine residues and are isotopically and non-isotopically labeled with many techniques, for example radiolabelling using conventional techniques (tyrosine residues - chloramine T, iodogen, lactoperoxidase; lysine residues - reagent of Bolton-Hunter). These coupling techniques are well known to those of skill in the art. Alternatively, tyrosine or lysine is added to fragments lacking these residues to facilitate labeling of amino and hydroxyl groups on the protein. The coupling technique is chosen based on the functional groups available on the amino acids including, but not limited to, amino, sulfhydryl, carboxyl, amide, phenol, and imidazole. The various reagents used to perform these couplings include, but are not limited to, glutaraldehyde, diazotized benzidine, carbodiimide, and p-benzoquinone.

Angiostatin proteins are chemically coupled to isotopes, enzymes, carrier proteins, cytotoxic agents, fluorescent, chemiluminescent, bioluminescent molecules, and other compounds for a variety of applications. The efficiency of the coupling reaction is determined using different techniques appropriate for the specific reaction. For example, radiolabeling of an angiostatin protein is accomplished with¹²⁵I using chloramine T and Na¹²⁵I of high specific activity. The reaction is terminated with sodium metabisulfite and the mixture is desalted on disposable columns. Labeled protein is eluted from the column and fractions are collected. Aliquots are removed from each fraction and radioactivity is measured in a gamma counter. In this way, the Na is separated¹²⁵Unreacted I of the labeled angiostatin protein. Protein fractions with the highest specific radioactivity are stored for later use such as testing for the ability to bind antisera to angiostatin.

ES 2 292 174 T3

Another application of protein conjugation is for the production of polyclonal antisera. For example, angiostatin proteins containing lysine residues bind to purified bovine serum albumin using glutaraldehyde. The efficiency of the reaction is determined by measuring the incorporation of the radiolabeled protein. Protein and unreacted glutaraldehyde are determined by dialysis. The conjugate is stored for later use.

Antisera can be generated against angiostatin, angiostatin analogs, angiostatin protein fragments and the angiostatin receptor. Following protein synthesis and purification, both monoclonal and polyclonal antisera are obtained using established techniques known to those of skill in the art. For example, polyclonal antisera can be obtained from rabbits, sheep, goats, or other animals. Angiostatin proteins conjugated to a carrier molecule such as bovine serum albumin, or angiostatin itself, are combined with a mixture of adjuvants, emulsified, and injected subcutaneously at multiple sites on the spine, neck, flanks, and sometimes into the the footpads. The booster injections are done at regular intervals, such as every 2 to 4 weeks. Blood samples are obtained by venipuncture, for example using the marginal veins of the ear after dilation, approximately 7 to 10 days after each injection. Blood samples are allowed to clot overnight at 4 ° C and centrifuged at approximately 2400 X g at 4 ° C for approximately 30 minutes. The serum is removed, an aliquot is removed and stored at 4 ° C for immediate use or at -20 to -90 ° C for later analysis.

All serum samples from polyclonal antisera generation or media samples from monoclonal antiserum production are analyzed for antibody titer determination. The titer is established by various means, for example, using dot blots and density analysis, and also by precipitation of radiolabeled protein-antibody complexes using protein A, secondary antisera, cold ethanol, or charcoal dextran followed by activity measurement with a gamma counter. Higher titer antisera are also purified on affinity columns that are commercially available. Angiostatin proteins are coupled to the gel on the affinity column. The antiserum samples are passed through the column and the anti-angiostatin antibodies remain bound on the column. These antibodies are subsequently eluted, collected, and evaluated for titer and specificity.

The highest titer anti-angiostatin antisera are tested to establish the following; a) optimal antiserum dilution for the highest antigen specific binding and lowest non-specific binding, b) the ability to bind increasing amounts of angiostatin protein in a conventional shift curve, c) possible cross-reactivity with related proteins and proteins, including plasminogen and also angiostatin of related species, d) ability to detect angiostatin proteins in extracts of plasma, urine, tissues and in cell culture media.

Kits for the measurement of angiostatin and the angiostatin receptor are also described. Antisera that possess the highest titer and specificity and that can detect angiostatin proteins in plasma, urine, tissue extracts, and cell culture media are further screened to establish easy-to-use kits for rapid angiostatin measurement and localization. reliable, sensitive and specific. These test kits include, but are not limited to, the following techniques; competitive and non-competitive assays, radioimmunoassays, bioluminescence and chemiluminescence assays, fluorometric assays, intercalation assays, immunoradiometric assays, dot blots, enzyme-associated assays including ELISA, microtiter plates, antibody coated strips or test strips for rapid monitoring urine or blood, and immunocytochemistry. The range, sensitivity, precision, reliability, specificity, and reproducibility of the assay are established for each kit. Inter-assay and within-assay variation is set to 20%, 50%, and 80% points on standard displacement or activity curves.

An example of a test kit commonly used in research and medicine is a radioimmunoassay (RIA) kit. An RIA for an angiostatin is illustrated below. Following successful radioiodination and purification of angiostatin or an angiostatin protein, the antiserum having the highest titer is added in various dilutions to tubes containing a relatively constant amount of radioactivity, such as 10,000 cpm, in a suitable buffer system. Other tubes contain buffer or pre-immune serum to determine non-specific binding. Following incubation at 4 ° C for 24 hours, protein A is added and tubes are vortexed, incubated at room temperature for 90 minutes, and centrifuged at approximately 2000 - 2500 X g at 4 ° C to precipitate complexes. of antibody bound to labeled antigen. The supernatant is removed by aspiration and the radioactivity in the pellet is counted in a gamma counter. The antiserum dilution that binds approximately 10 to 40% of the labeled protein after subtraction of non-specific binding is further characterized.

Next, a dilution range (approximately 0.1 pg to 10 ng) of the angiostatin protein used for antiserum development is evaluated by adding known amounts of the protein to tubes containing the antiserum and radiolabeled protein. After a further incubation period, for example 24 to 48 hours, protein A is added and the tubes are centrifuged, the supernatant is removed and the radioactivity in the pellet is counted. Displacement of the binding of radiolabeled angiostatin protein by unlabeled (pattern) angiostatin protein provides a standard curve. Various concentrations of other angiostatin protein fragments, plasminogen, angiostatin from different species, and homologous proteins are added to the test tubes to characterize the specificity of the antiserum against angiostatin.

Extracts from various tissues are prepared, including but not limited to primary and secondary tumors, Lewis lung carcinoma, cell cultures that produce angiostatin, placenta, uterus, and other tissues such as

ES 2 292 174 T3 brain, liver and intestine using extraction techniques that have been used successfully to extract angiostatin. After lyophilization or Speed ​​Vac of the tissue extracts, assay buffer is added and different aliquots are placed in the RIA tubes. Known angiostatin producing cell extracts produce displacement curves that are parallel to the standard curve, whereas non-angiostatin producing tissue extracts do not displace radiolabeled angiostatin from the angiostatin antiserum. In addition, extracts of urine, plasma and cerebrospinal fluid from animals with Lewis lung carcinoma are added to the test tubes in increasing amounts. Parallel displacement curves indicate the utility of the angiostatin assay to measure angiostatin in tissues and body fluids.

Tissue extracts containing angiostatin are further characterized by subjecting the aliquots to reverse phase HPLC. Eluate fractions are collected, dried in Speed ​​Vac, reconstituted in RIA buffer, and analyzed in the angiostatin RIA. The maximum amount of angiostatin immunoreactivity is located in the fractions corresponding to the angiostatin elution position.

The assay kit provides instructions, antiserum, angiostatin or angiostatin protein and angiostatin and / or possibly radiolabeled reagents for the precipitation of bound angiostatin-angiostatin antibody complexes. The kit is useful for the measurement of angiostatin in biological fluids and tissue extracts from animals and humans with and without tumors.

Another kit is used for the localization of angiostatin in tissues and cells. This Angiostatin Immunohistochemistry Kit provides instructions, antiserum to angiostatin and possibly blocking serum and secondary antiserum bound to a fluorescent molecule such as fluorescein isothiocyanate, or to some other reagent used to visualize the primary antiserum. Immunohistochemical techniques are well known to those of skill in the art. This angiostatin immunohistochemistry kit allows the localization of angiostatin in tissue sections and cultured cells using both light and electron microscopy. It is used for both research and medical purposes. For example, tumor biopsies are performed or tissue sections are harvested and cut with a microtome to examine angiostatin production sites. Such information is useful for diagnostic and possibly therapeutic purposes in the detection and treatment of cancer. Another method of visualizing angiostatin biosynthesis sites involves radiolabelling of nucleic acids for use in in situ hybridization to probe for angiostatin messenger RNA. Similarly, the angiostatin receptor can be localized, visualized, and quantified with immunohistochemical techniques.

This invention is further illustrated by the following examples, which are in no way to be construed as imposing limitations on the scope thereof.

Example 1

Choice of an animal tumor system in which the growth of metastasis by the primary tumor is inhibited and accelerated after removal of the primary tumor

By examining a variety of murine tumors that can inhibit their own metastasis, a Lewis lung carcinoma was selected in which the primary tumor most effectively inhibited lung metastasis. Six week old syngeneic male C57BI6 / J mice (subcutaneous back) were injected with 1 x 10⁶ tumor cells. Visible tumors first appeared after 3-4 days. When the tumors were about 1500 mm in size³, the mice were randomly divided into two groups. The primary tumor was completely excised in the first group and left intact in the second group after a sham operation. Although tumors from 500 mm³ up to 3000 mm³ inhibited the growth of metastases, 1500 mm³ it was the largest primary tumor that could be resected with a high survival rate and without local recurrence.

After 21 days, all mice were sacrificed and autopsied. In mice with an intact primary tumor, there were four +2 visible metastases, compared to fifty +5 metastases in the mice in which the tumor had been excised (p <0.0001). These data were confirmed by lung weight, which closely corresponds to tumor burden, as previously demonstrated. There was a 400% increase in wet lung weight in mice that had their tumors removed compared to mice in which the tumor remained intact (p <0.0001).

This experimental model provided reproducible data and the described experiment is reproducible. This tumor is marked as "Lewis lung carcinoma - little metastatic" (LLC-Low). The tumor also suppressed metastases in a nearly identical pattern in SCID mice, which are deficient in both B and T lymphocytes.

Example 2

Isolation of a highly metastatic Lewis lung carcinoma tumor variant, whether or not the primary tumor is removed

A highly metastatic variant of Lewis lung carcinoma arose spontaneously from the LLC-Low cell line of Example 1 in a group of mice and has been isolated according to the methods described in Example 1 and repeatedly transplanted. This tumor (LLC-High) forms more than 30 visible lung metastases whether or not the primary tumor is present.

ES 2 292 174 T3

Example 3

Size of metastases and tumor cell proliferation rate in it. Effect of the primary tumor that inhibits metastasis (LLC-Low)

C57BI6 / J mice were used in all experiments. Mice were inoculated subcutaneously with LLCLow cells, and 14 days later the primary tumor was excised in half of the mice. Mice were sacrificed on days 5, 10, and 15 after the tumor had been removed. Histological sections of lung metastases were obtained. Mice with an intact primary tumor had lung micrometastases that were neovascularized. These metastases were restricted to a diameter of 12-15 cell layers and did not show a significant increase in size even 15 days after tumor removal. In contrast, animals from which the primary tumor was excised revealed large vascularized metastases as early as 5 days postoperatively. These metastases underwent an additional 4-fold increase in volume on day 15 after tumor removal (as reflected by lung histology and weight). Approximately 50% of the animals from which a primary tumor had been excised died of lung metastasis before the end of the experiment. All animals with an intact primary tumor survived to the end of the experiment.

The replication rate in the metastases was determined by counting the BrdU-stained nuclei that had been previously injected into the mice. The high percentage of tumor cells incorporating BrdU in small avascular metastases from animals with an intact primary tumor was equivalent to the incorporation of BrdU from tumor cells in large vascularized metastases from mice from which the primary tumor had been removed (Figure 3 ). This discovery suggests that the presence of a primary tumor has no direct effect on the replication rate of tumor cells in metastasis.

In Figure 3, the left panel shows the BrdU labeling index of tumor cells in the lung in the presence or absence of a primary tumor. Prior to immunohistochemical staining, sections were permeabilized with 0.2 M HCl for 10 minutes and digested with 1 mg / ml proteinase K (Boehringer Mannheim GmbH, Mannheim, Germany) in 0.2 M Tris-HCl, CaCl₂ 2 mM at 37 ° C for 15 minutes. The labeling index was estimated by counting the percentage of positive nuclei at a power of 250. The right panel of Figure 3 represents an analysis of the total lung weight of tumors with intact or excised primary tumors 5, 10 and 15 days after the operation. Animals were sacrificed 6 hours after intraperitoneal injection of BrdU (0.75 mg / mouse).

Example 4

Inhibition of angiogenesis in lung metastases in the presence of an intact primary tumor

To measure the degree of vascularization in lung metastases, tissues were stained with antibodies against von Willebrand factor (a specific endothelial marker, available from Dako Inc., Carpenteria, CA). Metastases from animals with intact tumors formed a thin cuff (8-12 layers of tumor cells) around the existing pulmonary vessels. Except for the endothelial cells of the inner lining of the vessel, few or none of the cells were positive for von Willebrand factor. In contrast, animal lung metastases 5 days after removal of the primary tumor were not only larger but were also infiltrated with capillary buds containing endothelial cells that strongly stained for von Willebrand factor.

In immunohistochemical analysis for the presence of endothelial cells in lung metastases, a lung metastasis with the primary lung tumor intact 19 days after inoculation, had a cuff of tumor cells around a pre-existing microvessel in the lung. Metastasis was limited to 8 to 12 cell layers. There was no evidence of neovascularization around the microvessel, and it did not contain any new microvessels. This was typical of the maximum size of an avascular preangiogenic metastasis.

In an immunohistochemical analysis of tissue removed five days after resection of the primary tumor (19 days after inoculation of the primary tumor), the metastasis surrounded a pre-existing vessel in the lung. In contrast, in the sample in which the primary tumor was not resected, the tumor was neovascularized. Thus, an intact primary tumor inhibits the formation of new capillary blood vessels in the metastasis, but the proliferation of tumor cells in a metastasis is not affected by the primary tumor.

Example 5

A primary tumor inhibits the angiogenesis of a second tumor implanted in the cornea of ​​the mouse. The growth of this second tumor is inhibited

A 0.25 to 0.5 mm Lewis lung tumor (LLC-Low) was implanted² in the cornea of ​​a mouse on day 0. (Muthukkaruppan Vr., et al., Angiogenesis in the mouse cornea. Science 205: 1416-1418, 1979) A primary tumor was formed by inoculating 1 x 10⁶ LLC-Low cells subcutaneously in the back, or 4 or 7 days before implantation in the cornea; or the day of the implant in the cornea; or 4 or 7 days after the implant in the cornea. Control mice received the implant in the cornea but not the subcutaneous tumor. Other control mice received corneal implantation and an LLC-High tumor cell inoculation on the back 4 days prior to corneal implantation. The corneas were evaluated daily by stereomicroscopy for the study of the cornea (slit-lamp) to determine the

ES 2 292 174 T3 corneal tumor growth (measured by an ocular micrometer) and to determine the growth of new capillaries from the limbus edge of the cornea.

In control mice that did not carry a primary subcutaneous tumor, a large part of the corneas (6/8) developed neovascularization beginning on day 6 to 7 days after corneal implantation and continuing until day 10. On day 10, the vascularized corneal tumors had reached about a quarter of the volume of the entire eye. In the presence of the primary subcutaneous LLC-Low tumor, corneal implants did not become vascularized if the primary tumor was in place for at least 4 days or more prior to corneal implantation (Table 1). In the absence of neovascularization, corneal tumors grew slowly as thin, white, avascular discs on the cornea.

However, if the primary tumor did not implant until 4 days after implantation in the cornea, the corneas became vascularized and 3/3 of the corneal tumors grew at similar rates as the tumor-free controls. In the presence of the primary subcutaneous LLC-High tumor, the majority of the corneas (2/3) developed neovascularization beginning on day 7 after implantation in the cornea and continuing until day 10. By day 10, the vascularized corneal tumors had again reached about a quarter of the volume of the entire eye.

TABLE 1

<td colspan="2">Inhibition of tumor angiogenesis in the cornea by a primary subcutaneous tumor. [All primary tumors are LLC-Low except (*) which is LLC-High].</td>

<td>Day of implant in the eye</td><td> 0 0 0 0 0 0 0</td>

<td>Day of implant of the tumor primary</td><td>-7 -4 -4 * 0 none +4 +7</td>

<td>Number of mice with new glasses in the cornea the day 10</td><td> 2/10 0/9 2/3 2/3 6/8 3/3 2/3</td>

0/10 corneas would be expected to show neovascularization when the primary LLC-Low subcutaneous tumor was implanted 7 days prior to tumor implantation in the eye (ie -7). However, 2 of the tumors (2/10) had become necrotic due to being too large (&gt; 3 cm³).

ES 2 292 174 T3

Example 6

Intact primary tumor inhibits secondary subcutaneous implantation-induced angiogenesis of a basic growth factor (bFGF)

Although the experiments described in Examples 4 and 5 show that a primary tumor inhibits angiogenesis in a secondary metastasis, these studies do not reveal whether the primary tumor: (i) inhibits endothelial proliferation (or angiogenesis) directly, or (ii) indirectly down-regulating the angiogenic activity of metastatic tumor cells. To distinguish between these two possibilities, a subcutaneous angiogenesis focus was induced by a matrigel implant containing basic fibroblast growth factor (bFGF). (Passaniti A, et al., A simple, quantitative method for assessing angiogenesis and anti-angiogenic agents using reconstituted basement membrane, heparin and fibroblast growth factor. Lab. Invest. 67: 519, 1992).

Matrigel (a basement membrane protein extract), containing either 25 or 50 ng / ml of bFGF in the presence of heparin, was injected tumor-bearing (LLC-Low). The mice were sacrificed 4 days later and the concentration of hemoglobin in the gel was measured to quantify the formation of blood vessels. The number of new vessels entering the matrigel has previously been shown to correlate with the hemoglobin concentration. (Folkman J., Angiogenesis and its inhibitors in "Important Advances in Oncology 1985", VT DeVita, S. Hellman and S. Rosenberg, editors, JB Lippincott, Philadelphia 1985). Some gels were also prepared for histological examination. In normal mice, matrigel granules containing 50 ng / ml bFGF were completely red. They were heavily invaded by new capillaries, and contained 2.4 g / dl of hemoglobin. The matrigel, which lacked bFGF, was translucent and gray and contained only 0.4 g / dl of hemoglobin (a 6-fold difference). In contrast, matrigel from mice with a primary tumor contained only 0.5 g / dl (Figure 4).

The almost complete inhibition of anciogenesis in this experiment suggests that the presence of a primary Lewis lung tumor can directly inhibit bFGF-induced angiogenesis.

Example 7

Transfer of serum from a tumor-bearing animal to an animal from which the primary tumor has been excised inhibits metastases

Mice with Lewis lung carcinoma were implanted as described above. After 15 days, when the tumors were approximately 1500 mm³, the mice were randomly divided into four groups. Three groups underwent complete surgical resection of the primary tumor; in one group the tumors were left in place (after a sham surgical procedure). Mice in the three resection groups then received daily intraperitoneal injections of saline, serum from normal mice bearing no tumor, or serum from mice with 1500 mm Lewis lung carcinomas.³. The group of mice with tumors left intact received injections of intraperitoneal saline. All mice were treated for 21 days, after which the animals were sacrificed and lung metastases were counted (Table 2).

TABLE 2

<td>Tumor Primary excised primary tumor intact</td>

<td>Serum of Serum Injections Treatment Solution mice (mice injections of solution saline what intraperitoneal) normal saline carry tumor</td>

<td>Number of metastasis of 55 ± 5 50 ± 4 7 ± 2 3 ± 1 lung:</td>

ES 2 292 174 T3

These results were confirmed by lung weight. P = &lt;0.0001 for the difference between the two groups [(55 &amp; 50) vs. (7 &amp; 3)]. Similar results have been obtained using angiostatin from the urine of tumor-bearing animals.

Example 8

Bovine Capillary Endothelial Cell Assay (BCE)

BCE cells are used between steps 9 and 14 only. On day 0, BCE cells were seeded on gelatinized 24-well plates (1.5% gelatin in PBS at 37 °, 10% CO₂ for 24 hours and then rinsed with 0.5 ml PBS) at a concentration of 12,500 cells / well. Cell counts were performed using a hemocytometer. The cells were seeded in 500 µl of DMEM with heat inactivated 10% calf serum (56 ° C for 20 minutes) and 1% glutamine-pen-strep (combination of glutamine, penicillin and streptomycin) (GPS).

The BCE cells were exposed as follows: the medium is removed and replaced with 250 µl of DMEM / 5% BCS / 1% GPS. The sample to be tested is then added to the wells. (The amount varies depending on the sample being tested.) Plates are placed at 37 ° C / 10% CO₂ for about 10 minutes. 250 µl of DMEM / 5% BCS / 1% GPS with 2ng / ml bFGF was added to each well. The final medium is 500 µl of DMEM / 5% BCS / 1% GPS / with 1ng / ml bFGF. The plate is returned to the incubator at 37 ° C / 10% CO₂ for 72 hours.

On day 4, cells are counted by removing medium and then trypsinizing all wells (0.5 ml trypsin / EDTA) for 2 to 3 minutes. The suspended cells are then transferred to 9.5 ml Hemetall scintillation vials and counted using a Coulter counter. One unit of activity is the amount of serum containing angiostatin that can produce half the maximal inhibition of capillary endothelial proliferation when endothelial cells are incubated in 1 ng / ml bFGF for 72 hours.

Example 9

Serum from mice bearing low metastatic Lewis lung tumor (LLC-Low) inhibits proliferation of capillary endothelial cells in vitro

Bovine capillary endothelial cells were stimulated by basic fibroblast growth factor (1ng / ml bFGF), in a 72 hour proliferation assay. Serum from tumor-bearing mice added to these cultures inhibited endothelial cell proliferation in a dose-dependent and reversible manner. Normal serum was not inhibitory (Figure 5). Endothelial cell proliferation was inhibited in a similar manner (relative to controls) by serum obtained from tumor bearing nu / nu mice and SCID mice. After removing the primary tumor, angiostatin activity disappeared from the serum within 3-5 days.

Tumor-bearing serum also inhibited bovine aortic endothelial cells and endothelial cells derived from spontaneous mouse hemangioendothelioma, (Obeso, et al., "Methods in Laboratory Investigation, A Hemangioendothelioma derived cell line; Its use as a Model for the Study of Endothelial Cell Biology ”, Lab Invest., 63 (2), pages 259-269, 1990) but did not inhibit Lewis lung tumor cells, 3t3 fibroblasts, aortic smooth muscle cells , mink lung epithelium, or human fetal lung fibroblasts W 138.

Example 10

The serum of mice bearing the Lewis lung tumor (LLC-High) that does not inhibit metastasis, does not inhibit the proliferation of capillary endothelial cells in vitro

Serum from mice bearing a primary LLC-High tumor did not significantly inhibit bFGF-stimulated bovine capillary endothelial cell proliferation relative to controls. Furthermore, when this serum was subjected to the first two purification steps (heparin-Sepharose chromatography and gel filtration), no angiostatin activity was found in any fraction.

Example 11

Lewis lung carcinoma ascites (poorly metastatic) also generates angiostatin serum

Mice received intraperitoneal injections of either LLC-Low or LLC-High tumor cells (10⁶), and a week later, 1-2 ml of hemorrhagic ascites were obtained from each of the 10-20 mice. Mesenteric tumor seeding was observed. The mice were then sacrificed. Serum was obtained by cardiac puncture. Serum was also obtained from normal mice, which did not carry tumor as a control. Serum and ascites were centrifuged to remove cells, and supernatant assayed on bFGF-stimulated bovine capillary endothelial cells (1 ng / ml) (see example 8). Ascites originating from both tumor types stimulated significant capillary endothelial cell proliferation (eg, 100% proliferation) over controls after 72 hours (Figure 6). In contrast, serum from poorly metastatic mice inhibited endothelial cell proliferation (inhibition up to 79% of controls). Serum from the highly metastatic line was 200% stimulatory.

ES 2 292 174 T3

These data show that low metastatic line ascites contains a predominance of endothelial growth stimulator over angiostatin. This state is analogous to a solid primary tumor. Furthermore, angiostatin activity appears in serum, as if it were not opposed to stimulatory activity. This pattern is similar to that of the solid primary tumor (LLC-Low). Highly metastatic tumor ascites (LLC-High) also appears to contain a predominance of endothelial cell stimulator, but serum angiostatin cannot be identified.

Example 12

Separation of Angiostatin from Serum by Column Chromatography and Analysis of Growth Inhibitory Fractions by SDS-PAGE

To purify the angiostatin (s), serum from tumor-bearing mice was pooled. Inhibitory activity, tested according to the in vitro inhibitory activity assay described above, was treated by chromatography sequentially using Heparin-Sepharose, A0.5 mm Agarose Biogel, and several cycles of high-performance liquid chromatography in phase. reverse C4 (HPLC). SDS-PAGE of the HPLC fraction containing endothelial inhibitory activity revealed a discrete band of M_r reduced apparent 38,000 Daltons, which was purified approximately 1 million times (see Table 3) to a specific activity of approximately 2x10⁷. At different stages of purification, the pooled fractions were tested with specific antibodies for the presence of known endothelial inhibitors. No platelet factor-4, thrombospondin, or transforming growth factor beta were found in the purified or partially purified fractions.

TABLE 3

Num €: number of times

Specific activity {units * / mg) of purification

Serum 1.69 1

Heparin Sepharose 14.92 8.8

Bio-gel AO.5 m 69.96 41.4

HPLC / C4 2x10⁷ i, 2xl0⁶ * One unit of activity is the amount of serum containing angiostatin that can produce half the maximum inhibition of capillary endothelial proliferation when enc.othelial cells are incubated in 1 ng / ml bFGF for 72 hours.

Example 13

Separation of angiostatin from urine by column chromatography and analysis of growth inhibitory fractions by SDS-PAGE

Purification of the endothelial cell inhibitor (s) from serum is impeded by the small volume of serum that can be obtained from each mouse and by the large amount of protein in the serum.

The urine of tumor-bearing mice was analyzed and found to contain an inhibitor of endothelial cell proliferation that is absent from the urine of non-tumor-bearing mice and mice with LLC-high tumors. Purification of endothelial cell inhibitory activity was carried out by the same strategy as was used for serum purification (described above) (Figure 7).

Figure 7 shows C4 reverse phase chromatography of partially purified serum or urine from tumor bearing animals. All fractions in bovine capillary endothelial cells were assayed with bFGF in a 72-hour proliferation assay as described in Example 8. A discrete inhibition peak was observed in both cases eluting at 30-35% acetonitrile in fraction 23. SDS-polyacrylamide gel electrophoresis of the inhibitory fraction of the third cycle of C4 reverse phase chromatography of serum from tumor bearing animals showed a single band at approximately 38,000 Daltons.

ES 2 292 174 T3

Example 14

Characterization of circulating angiostatin

Endothelial inhibition was tested according to the procedure described in Example 9. Angiostatin was isolated on a Synchropak HPLC C4 column (Synchrom, Inc. Lafayette, IN). The inhibitor was eluted at a 30-35% acetonitrile gradient. On a sodium dodecyl sulfate polyacrylamide gel electrophoresis (PAGE) gel under reducing conditions (b-mercaptoethanol (5% v / v)), the protein band with activity was eluted at 38 kilodaltons. Under non-reducing conditions, the protein with activity eluted at 28 kilodaltons. Activity is found at similar points, whether the urine or serum sample is isolated. No activity was detected with any other bands.

The activity associated with the bands was lost when heated (100 ° C for 10 minutes) or treated with trypsin. When the activity band was extracted with a water / chloroform mixture (1: 1), the activity was found only in the aqueous phase.

Example 15

Purification of inhibitory fragments from human plasminogen

Plasminogen lysine binding site I was obtained from Sigma Chemical Company. The preparation is purified human plasminogen after digestion with elastase. The lysine-binding site I obtained in this way is a population of proteins containing, in aggregate state, at least the first three triple-loop structures (numbers 1 to 3) in the plasmin A chain (Kringle 1 + 2 + 3). (Sotrrup-Jensen, L., et al. In Progress in Chemical Fibrinolysis and Thrombolysis, Vol. 3, 191, Davidson, JF, et al. eds. Rayen Press, New York 1978 and Wiman, B., et al., Biochemica et Biophysica Acta, 579, 142 (1979)). Plasminogen lysine binding site I (Sigma Chemical Company, St. Louis, MO) was resuspended in water and applied to a C4 reverse phase column that had been equilibrated with HPLC grade water / 0.1 TFA. %. The column was eluted with a gradient of water / 0.1% TFA versus acetonitrile / 0.1% TFA and the fractions were collected in polypropylene tubes. An aliquot of each was evaporated on a centrifugal evaporator (speed vac), resuspended with water, and applied to BCE in a proliferation assay. This procedure was repeated twice for the inhibitory fractions using a similar gradient for elution. Inhibitory activity eluted at 30-35% acetonitrile in the final run from column C4. SDS-PAGE of the inhibitory fraction revealed 3 discrete bands of apparent reduced molecular mass of 40,42.5 and 45 kd. SDS-PAGE under non-reducing conditions revealed three bands of molecular mass 30, 32.5 and 35 kd respectively.

Example 16

Inhibitory Activity Extraction from SDS-PAGE

Inhibitory fractions purified from human plasminogen-based purifications were resolved by SDS-PAGE under non-denaturing conditions. The areas of the gel corresponding to bands observed in neighboring lanes loaded with samples by silver staining were cut out of the gel and incubated in 1 ml of phosphate buffered saline at 4 ° C for 12 hours in polypropylene tubes. The supernatant was removed and dialyzed twice against saline for 6 hours (MWCO = 6-8000) and twice against distilled water for 6 hours. The dialysate was evaporated by vacuum centrifugation. The product was resuspended in saline and applied to bovine capillary endothelial cells stimulated by 1 ng / ml basic fibroblast growth factor in a 72 hour assay. The protein extracted from each of the three bands inhibited capillary endothelial cells.

Example 17

Plasminogen fragment treatment studies

Mice were implanted with Lewis lung carcinomas and underwent resections when tumors were 1500-2000 mm³. On the day of the operation, the mice were randomly divided into 6 groups of 6 mice each. Mice received daily intraperitoneal injections with the three purified inhibitory fragments of human plasminogen, whole human plasminogen, urine from tumor-bearing animals, urine from normal mice, or saline. A group of tumor bearing animals having only a sham procedure was treated with injections of saline. Immediately after removal of the primary tumor, mice receive an intraperitoneal injection of 24 pg (1.2 mg / kg / day / mouse) of the inhibitory plasminogen fragments as a loading dose. They then received a daily intraperitoneal injection of 12 pg of the inhibitory fragment (0.6 mg / kg / day / mouse) for the duration of the experiment. Control mice receive the same dose of complete plasminogen molecule after tumor removal. For urine treatments, urine from normal or tumor-bearing mice was filtered, extensively dialyzed, lyophilized, and then resuspended in sterile water to obtain a 250-fold concentration. Mice were given 0.8 ml of the dialyzed urine concentrate, either tumor-bearing mice or normal mice, in two intraperitoneal injections on the day of removal of the primary tumor as a loading dose. They then received daily intraperitoneal injections of 0.4 ml of the urine subjected to dialysis and concentrated during the course of the experiment. Treatments continued for 13 days, at which time all mice were sacrificed and autopsied.

ES 2 292 174 T3

Figures 8 and 9 show the results of the experiment. Figure 8 shows. Superficial lung metastases after 13-day treatment. Superficial lung metastases refer to the number of metastases observed in the lungs of mice at autopsy. A stereomicroscope was used to count metastases. Figure 8 shows the mean number of superficial lung metastases that were counted and the standard error of the mean. As shown, the group of mice with the primary tumor present did not show metastasis. Mice, in which the primary tumor was resected and treated with saline, showed extensive metastasis. Mice treated with the human-derived plasminogen fragment did not show metastasis. Mice treated with whole plasminogen showed extensive metastasis indicating that the whole plasminogen molecule has no endothelial inhibitory activity. Those mice treated with dialyzed and concentrated urine from tumor-bearing mice showed no metastasis. Mice treated with concentrated urine from normal mice showed extensive metastasis. When lung weight was measured, similar results were obtained (Figure 9).

Example 18

Amino Acid Sequence of Human and Murine Angiostatin

The amino acid sequence of angiostatin isolated from mouse urine and angiostatin isolated from the human lysine-binding site I fragment preparation was determined on an Applied Biosystem model 477A protein sequencer. Phenylthiohydantoin amino acid fractions were identified with an on-line ABI model 120a HPLC. The amino acid sequence determined from the N-terminal sequence and the tryptic digests of human and murine angiostatin indicate that the sequence of angiostatin is similar to the sequence starting at amino acid number 98 of murine plasminogen. Thus, the amino acid sequence of angiostatin is a molecule that comprises a protein having a molecular weight of between about 38 kilodaltons and 45 kilodaltons as determined by reducing polyacrylamide gel electrophoresis and having a substantially similar amino acid sequence. to that of the murine plasminogen fragment starting at amino acid number 98 of an intact murine plasminogen molecule. The initial amino acid sequence of murine angiostatin (SEQ ID NO: 2) is shown in Figure 1. The length of the amino acid sequence can be slightly longer or shorter than that shown in Figure 1.

Analysis of the N-terminal amino acid and tryptic digests of the active fraction of human lysine binding site I (see Example 15) shows that the sequence of the fraction begins at approximately amino acid 97 or 99 of Romanian plasminogen and human angiostatin. it is homologous to murine angiostatin. Figure 2 shows the initial amino acid sequence of human angiostatin (starting at amino acid 98), (SEQ ID NO: 3). The amino acid sequence of human and murine angiostatin is compared in Figure 2 with the corresponding internal amino acid sequences of plasminogen from other species including porcine, bovine and Rhesus monkey plasminogen, indicating the presence of angiostatin in those species.

Example 19

Expression of human angiostatin in E. coli

The vector pTrcHisA (Invitrogen) (Figure 10) was used to obtain high-level regulated transcription of the trc promoter for increased translation efficiency of eukaryotic genes in E. coli. Angiostatin is expressed fused to an N-terminal nickel-binding polyhistidine tail for one-step purification using metal affinity resins. The enterokinase cleavage recognition site in the fusion protein allows the subsequent removal of the N-terminal histidine fusion protein from the purified recombinant protein. Recombinant human angiostatin protein was found to bind lysine; It cross-reacts with monoclonal antibodies specific for kringle regions 1, 2, and 3, and inhibits bFGF-driven endothelial cell proliferation in vitro.

To construct the insert, the gene fragment encoding human angiostatin is obtained from human liver mRNA which is reverse transcribed and amplified using polymerase chain reaction (PCR) and specific primers. The 1,131 base pair product codes for amino acids 93 to 470 of human plasminogen. The amplified fragment was cloned into the XhoI / KpnI site of pTrcHisA, and the resulting construct was transformed into E. coli XL-1B host cells (available from Stratagene). A control clone containing the plasmid vector pTrcHisA was also transformed. only in E. coli XL-1B host cells. This clone is referred to as the vector control clone. Both clones were purified identically as described below.

Expression colonies were selected as follows. Colony lifts of E. coli transformed with the gene encoding angiostatin were grown on IPTG-impregnated nitrocellulose filters and coated onto an LB agar plate. Following IPTG induction of expression, colonies were lysed on nitrocellulose filters. Nitrocellulose isolates were blocked, cleared, and gavaged with two separate monoclonal antibodies (mAb Dcd and Vap; gift from SG Mc Cance and FJ Castellino, University of Notre Dame) that recognize specific angiostatin conformations. Strongly expressing colonies were selected by the mAbs.

To identify the optimal time for maximum expression, cells were harvested at various times before and after can IPTG induction and exposed to repeated freeze-thaw cycles, followed by cor analysis. SDS-PAGE, immunoblotting and probe treatment with Mab Dcd and Vap.

ES 2 292 174 T3

From these, the clone pTrcHisA / HAsH4 was selected. Induction with IPTG was for 4 hours after which the cell pellet was collected and resuspended in 50 mM Tris pH 8.0, 2 mM EDTA, 5% glycerol and 200 mg / ml lysozyme and stirred for 30 min. at 4 ° C. The suspension was centrifuged at 14,000 rpm for 25 min. and the pellet was resuspended in 50 mM Tris pH 8.0, 2 mM EDTA, 5% glycerol and 0.1% DOC. This suspension was stirred for 1 hr. at 4 ° C, and then centrifuged at 14,000 rpm for 25 min. The supernatant fraction at this stage contains expressed angiostatin. Human angiostatin expressed by E. coli was found to possess the physical property of native angiostatin, which is the ability to bind lysine. Thus angiostatin expressed by E. coli was purified on a lysine-sepharose column (Pharmacia or Sigma) in a single step. Elution of angiostatin from the column was with 0.2 M epsilon-amino-n-caproic acid pH 7.5.

Subsequent to these experiments, a 10 l fermentation batch of clone pTrcHisA / HAsH4 was scaled up. The cells obtained from this scaling up induction were pelleted and resuspended in 50 mM Tris pH 7.5, disrupted at 10,000 psi by cooling three times at 10 ° C between passages. The lysate obtained was clarified by centrifugation at 10,000 rpm for 30 min at 4 ° C, and angiostatin expressed on lysine-sepharose was isolated (Figure 11).

Purified E. coli expressed human angiostatin was extensively dialyzed against water and lyophilized. Expressed human angiostatin was resuspended in medium (DMEM, 5% BCS, 1% gentamicin / penicillin / streptomycin) to an estimated concentration of 3 jug / ml, and was used in in vitro bovine capillary endothelial cell (BCE) assays, as described in Example 8, page 39. Similarly, the control clone containing only the vector was treated identically to the clone pTrcHisA / HAsH4. It was induced with IPTG in an identical manner, and the bacterial lysate was used to bind to lysine, eluted with 0.2M aminocaproic acid, extensively dialyzed, and lyophilized. This control preparation was also resuspended in medium at an estimated concentration of 3 pg / ml. Recombinant angiostatin samples and controls were obtained from different induction and fermentation batches as well as separate purification runs, all coded from EntreMed, Maryland. BCE trials were performed with these coded samples in a blinded fashion at Children's Hospital, Boston.

The results of the recombinant human angiostatin BCE assays showed that human angiostatin expressed in E. coli inhibited the proliferation of BCE cells due to bFGF (used at 1 ng / ml) (Figure 12). Recombinant angiostatin stock was used in medium (at approximately 3 pg / ml) at a dilution of 1: 5, 1:10 and 1:20. The percentage of inhibition was calculated as follows:

number of cells with angiostatin - number of cells on day 0 number of cells with bFGF alone - number of cells on day 0

The percent inhibition of BCE cell proliferation was comparable to or greater than that of plasminogen-derived angiostatin at similar concentrations. Figure 13 depicts the results of a repeat series of the BCE assay, in which at a 1: 5 dilution of the recombinant reserve angiostatin they gave percentages of inhibition similar to those obtained with plasminogen-derived angiostatin. Figure 13 shows the surprising result that human recombinant angiostatin protein inhibits more than 60%, and as much as more than 75% of BCE proliferation in culture.

Example 20

Angiostatin maintains micrometastasis inactivity by increasing the apoptosis rate

After subcutaneous inoculation of C57 BL6 / J mice with Lewis lung carcinoma cells (1 x 10⁶), primary tumors of approximately 1.5 cm developed³. Animals were subjected to either surgical removal of the primary tumor or sham surgery. On days 5, 10, and 15 after surgery, the mice were sacrificed and their lungs prepared for histological examination. Animals with resected primary tumors showed massive proliferation of micrimetastases compared to sham controls (Figure 14). These changes were accompanied by a significant increase in lung weight.

Analysis of tumor cell proliferation, as measured by bromo-deoxyuridine (BrdU) uptake, showed no differences between animals with intact primary tumors or tumors resected on days 5, 9, and 13, indicating that the increase in tumor mass could not be explained by increased proliferation (figure 15). Therefore, cell death was examined in these animals. Apoptosis, a cell death process that depends on changes in gene expression and involves the elimination of cells during development and in rapidly proliferating tissues such as the small intestine, was examined by immunohistochemically labeling fragmented DNA with the deoxynucleotidyl transferase technique. terminal (TdT). The apoptotic index was determined at each time of sacrifice. Removal of primary tumors caused a statistically significant (approximately 3-4 fold) increase in apoptotic index at all time points examined (Figure 15).

Supporting evidence was obtained by treating mice with excised primary tumors with an exogenous suppressor of angiogenesis. This substance, TNP-1470 (O-chloroacetylcarbamoylfumagilol, formerly called AGM-1470) is a fumagillin analog with reported anti-angiogenic activity. Subcutaneous injection of TNP-1470 (30 mg / kg

ES 2 292 174 T3 every other day) produced results that were strikingly similar to those described above for animals having intact primary tumors. These animals showed lower lung weight, equivalent proliferative index, and increased apoptotic index compared to saline-injected controls (Figure 16).

These data indicate that metastases remain inactive when tumor cell proliferation is balanced by an equivalent rate of cell death. Removal of the primary tumor causes a rapid increase in the growth of metastases, probably due to the removal of angiogenesis inhibitors (angiostatin) that control metastatic growth by increasing apoptosis in tumor cells. These effects are similar to those observed after removal of primary tumors and administration of an exogenous angiogenesis inhibitor. Taken together, these data suggest that the primary tumor releases angiostatin which maintains the quiescence of the micrometastases.

Example 21

Treatment of primary tumors with angiostatin in vivo

Angiostatin was purified from human plasminogen by limited elastase digestion as described in Example 15 above. Angiostatin was resuspended in phosphate buffered saline for administration to six week old male C57BI6 / J mice. Animals were implanted subcutaneously 1 X 10⁶ tumor cells of either Lewis lung carcinoma or T241 fibrosarcoma. Angiostatin treatment was started after four days when the tumors were 80-160 mm in size.³. Mice received angiostatin injections either as a single 40 mg / kg injection or two 80 mg / kg injections intraperitoneally (ip) or subcutaneously (sc). Animals were sacrificed at various times after treatment was extended up to 19 days.

Angiostatin, administered at a daily dose of 40 mg / kg ip, produced a highly significant inhibition of the growth of primary T241 tumors (Figure 17). This inhibitory effect on growth was visibly evident within 2 days and increased in magnitude throughout the time course of the study. On day 18, angiostatin-treated mice had tumors that were approximately 38% of the volume of saline-injected controls. This difference was statistically significant (p &lt;0.001, Student's t test).

Angiostatin treatment (total dose 80 mg / kg / day, administered twice daily at 40 mg / kg ip or sc) also significantly reduced the growth rate of CLL-LM primary tumors (Figure 17). This inhibitory effect was evident at 4 days and increased in magnitude at all subsequent times examined. On the last day of the experiment (day 19), angiostatin-treated mice had a mean tumor volume that was only 20% from saline-injected controls that was significantly different (p &lt;0.001 Student's t test).

In another series of experiments, angiostatin (50 mg / kg every 12 h) was administered to mice implanted with T241 fibrosarcoma, Lewis lung carcinoma (LM), or cellular reticulum sarcoma cells. For each type of tumor cell, the mice that received angiostatin had a substantially reduced tumor size. Figure 19 demonstrates that for T241 fibrosarcoma, angiostatin-treated mice had mean tumor volumes that were only 15% of untreated mice on day 24. Figure 20 demonstrates that for Lewis lung carcinoma (LM), the Angiostatin-treated mice had mean tumor volumes that were only 13% of untreated mice on day 24. Figure 21 demonstrates that for reticulum sarcoma, angiostatin-treated mice had mean tumor volumes that were only 19% of untreated mice on day 24. Data represent the average of 4 mice at each time point.

These results demonstrate that angiostatin is an extremely potent inhibitor of the growth of three different primary tumors in vivo.

Example 22

Treatment of primary tumors in human cell-derived mice with angiostatin in vivo

The effect of angiostatin on two human tumor cell lines, PC-3 human prostate carcinoma and MDA-MB human breast carcinoma, was studied. Immunodeficient SCID mice were implanted with human tumor cells, and mice were treated with angiostatin 50 mg / kg every 12 hours essentially as described in Example 21. The results demonstrate that the angiostatin protein of the present invention is a potent inhibitor. of human tumor cell growth. Figure 22 shows that for PC-3 human prostate carcinoma, angiostatin-treated mice had only 2% mean tumor volume compared to untreated control mice on day 24. Figure 23 shows that for angiostatin carcinoma MDA-MB human breast, angiostatin-treated mice had only 8% mean tumor volume compared to untreated control mice on day 24.

ES 2 292 174 T3

Example 23

Gene therapy - Effect of angiostatin gene transfection on tumor volume

A 1380 base pair DNA sequence for angiostatin derived from mouse plasminogen cDNA (obtained from the American Type Culture Collection (ATCC)), encoding amino acids 1-460 of mouse plasminogen, was generated using PCR, and was inserted into an expression vector. The expression vector was transfected into T241 fibrosarcoma cells and the transfected cells were implanted into mice. Control mice received either untransfected T241 cells, or T241 cells transfected with the vector only (ie transfected cells that do not express angiostatin). Three angiostatin expressing transfected cell clones were used in the experiment. Mean tumor volume was determined over time. The results show the surprising and drastic reduction in mean tumor volume in mice for the angiostatin expressing cell clones compared to the non-expressing and non-transfected control cells.

The mouse DNA sequence encoding mouse angiostatin protein is derived from mouse plasminogen cDNA. Mouse angiostatin spans the kringle 1-4 regions of mouse plasminogen. A scheme for the construction of this clone is shown in figure 24.

Mouse angiostatin protein clones were transfected into T241 fibrosarcoma cells using the LIPOFECTIN ™ transfection system (available from Life Technologies, Gaithersburg, MD). LIPOFECTIN ™ reagent is a 1: 1 (w / w) liposome formulation of the cationic lipid N- [1- (2,3-dioleyloxy) propyl] -n, n, n-trimethylammonium (DOTMA), and diolecoyl chloride. Phosphotidylethanolamine (DOPE) in membrane filtered water.

The procedure for transient transfection of cells is as follows:

1. T241 cells are grown in 60 cm tissue culture plates², seeding = 1-2 x 10⁵ cells in 2 ml of the appropriate growth medium supplemented with serum.

2. Incubate cells at 37 ° C in a CO incubator.₂ until cells are 40-70% confluent. Typically it will take 18-24 hours, but the time will vary between cell types. The confluence of the T241 tumor cells was approximately 70%.

3. Prepare the following solutions in sterile 12 x 75 mm tubes:

Solution A: For each transfection, dilute 5 µg of DNA in 100 µl of serum-free OPTI-MEM I reduced serum medium (available from Life Technologies) (tissue culture grade deionized water can also be used).

Solution B: For each transfection, dilute 30 µg of LIPOFECTIN in 100 µl of OPTI-MEM medium.

Four. Combine the two solutions, mix gently and incubate at room temperature for 10-15 min.

5. Wash cells twice with serum-free medium.

6. For each transfection, add 0.8 ml of serum-free medium to each tube containing the LIPOFECTIN reagent complexes.^tm-DNA. Mix gently and coat the complex on the cells.

7. Incubate cells for approximately 12 h at 37 ° C in a CO incubator.₂.

8. Replace the DNA-containing medium with 1 mg / ml selection medium containing serum and incubate the cells at 37 ° C in a CO incubator.₂ for a total of 48-72 h.

9. Assay cell extracts for gene activity 48-72 h after transfection.

Transfected cells can be assayed for angiostatin protein expression using angiostatin-specific antibodies. Alternatively, after approximately 10-14 days, G418 resistant colonies appeared on T241 cells transfected with CMV angiostatin. Furthermore, several clones were observed in the vector-only transfected clones but not in the non-transfected clones. G418 resistant clones were selected for their angiostatin expression, using an immunofluorescence method.

Interestingly, in vitro cell growth of angiostatin-transfected T241 cells and Lewis lung cells was either not inhibited or adversely affected, as shown in Figures 25 and 26.

Figure 27 represents the results of the transfection experiment. The three angiostatin-expressing T241 transfected clones produced mean tumor volumes in mice that were substantially reduced relative to tumor volume in control mice. The mean tumor volume of the mice implanted with clone 37 was only 13% of the control, while the tumor volumes of clone 31 and clone 25 were only 21% and 34% of the control tumor volumes, respectively. . These results demonstrate that the DNA sequences that

ES 2 292 174 T3 encoding angiostatin can be transfected into cells, that transfected DNA sequences can express angiostatin protein by implanted cells, and that expressed angiostatin functions in vivo to reduce tumor growth.

Example 24

Location of the site of expression of angiostatin in vivo

To localize the site of expression of angiostatin protein in vivo, total RNA from various cell types, Lewis lung carcinoma cells (mouse), T241 fibrosarcoma (mouse), and Burkitt lymphoma cells (human) were analyzed. ), either from a new tumor or from a cell culture after several steps to determine the presence of angiostatin transcripts. Northern blot analysis of the samples showed an absence of any signal hybridization to the thn sequence of all samples except that of mouse liver RNA which showed a unique signal of approximately 2.4 kb corresponding to mouse plasminogen. Northern blot analysis of human samples shows an absence of any signal hybridization to human angiostatin sequence from all samples except that of normal human liver RNA which shows a unique signal of approximately 2.4 kb corresponding to human plasminogen.

Reverse transcription polymerase chain reaction (RT-PCR) analysis showed an absence of any product from all samples probed with mouse angiostatin sequences except that of normal mouse liver. RT-PCR analysis showed an absence of any product from all human samples probed with human angiostatin sequences except normal human liver (expected size 1050 bp for mouse and 1134 bp for human).

Thus, it appears that mouse angiostatin transcripts (assuming identity with amino acids 97 to 450 of mouse plasminogen) are not produced by all of the above mouse samples and human angiostatin transcripts (assuming identity with amino acids 93 to 470 plasminogen) are not produced by the above human samples. Positive signals obtained in normal human / mouse liver come from hybridization with plasminogen.

Example 25

Angiostatin expression in yeast

The gene fragment encoding amino acids 93 to 470 of human plasminogen was cloned into the XhoI / EcoRI site of pHIL-SI (Invitrogen) which allows secreted expression of proteins using the secretion signal PHO1 in the yeast Pichia pastoris. Similarly, the gene fragment encoding amino acids 93 to 470 of human plasminogen was cloned into the SnaBI / EcoRI site of pPIC9 (Invitrogen) allowing secreted expression of proteins using the factor-a secretion signal in Pichia yeast pastoris. Human angiostatin proteins expressed in these systems will have many advantages over those expressed in E. coli such as protein processing, protein folding, and post-translational modification including glycosylation.

Gene expression in P. pastoris: described: in) Sreekrishna, K. et al. (1988) High level expression of heterologous proteins in methylotropic yeast Pichia pastoris. J. Basic Microbiol. 29 (4): 265-278, and Clare, JJ et al. (1991) Production of epidermal growth factor in yeast: High-level secretion using Pichia pastoris strains containing multiple gene copies, Gene 105: 205-212, both incorporated herein by reference.

Example 26

Angiostatin protein expression in transgenic plants and animals

Transgenic animals such as those of the bovine or porcine family that express the angiostatin gene transcript are created. The transgenic animal expresses the angiostatin protein for example in the milk of these animals. Additionally edible transgenic plants expressing the angiostatin gene transcript are constructed.

The construction of transgenic animals expressing foreign DNA is described in Smith H. Phytochrome transgenics: functional, ecological and biotechnical applications, Semin. Cell. Biol. 1994 5 (5): 315-325, which is incorporated herein by reference.

Example 27

Characterization of angiostatin fragments that inhibit entothelial cell proliferation

The following example characterizes the activity of individual and combined angiostatin fragments. The data suggest that there is a functional difference between individual kringle structures and a potent anti-endothelial and therefore anti-angiogenic activity can be obtained from such angiostatin protein fragments.

ES 2 292 174 T3

As used herein, "angiostatin fragment" means a protein derivative of angiostatin, or plasminogen, which has an endothelial cell proliferation inhibiting activity. Angiostatin fragments are useful for treating angiogenesis-mediated conditions or diseases. For example, angiostatin fragments can be used to inhibit or suppress tumor growth. The amino acid sequences of such an angiostatin fragment, for example, can be selected from a part of murine plasminogen (SEQ ID NO: 1), murine angiostatin (SEQ ID NO: 2); Human angiostatin (SEQ ID NO: 3), Rhesus monkey angiostatin (SEQ ID NO: 4), porcine angiostatin (SEQ ID NO: 5) and bovine angiostatin (SEQ ID NO: 6), unless otherwise indicated by the context in which it is used.

As used herein, "kringle 1" means a protein derivative of plasminogen having anti-angiogenic activity or endothelial cell inhibitory activity, and having an amino acid sequence comprising a sequence homologous to kringle 1, exemplified using, but not limited to, murine kringle 1 (SEQ ID NO: 7), human kringle 1 (SEQ ID NO: 8), Rhesus monkey kringle 1 (SEQ ID NO: 9), porcine kringle 1 (SEQ ID NO : 10), and bovine kringle 1 (SEQ ID NO: 11), unless otherwise indicated by the context in which it is used. Murine Kringle 1 (SEQ ID NO: 7) corresponds to amino acid positions 103 to 181 (inclusive) of the murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 6 to 84 (inclusive) of murine angiostatin of SEQ ID NO: 2. Human kringle 1 (SEQ ID NO: 8), Rhesus monkey kringle 1 SEQ ID NO: 9), porcine kringle 1 (SEQ ID NO: 10) and bovine kringle 1 (SEQ ID NO: 11) correspond to the positions of amino acids 6 to 84 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 2" means a protein derivative of plasminogen having anti-angiogenic activity or endothelial cell inhibitory activity, and having an amino acid sequence comprising a sequence homologous to kringle 2, exemplified by, but not limited to, murine kringle 2 (SeQ ID NO: 12), human kringle 2 (SEQ ID NO: 13), Rhesus monkey kringle 2 (SEQ ID NO: 14), porcine kringle (SEQ ID NO : 15) and bovine kringle 2 (SEQ ID NO: 16), unless otherwise indicated by the context in which it is used. Murine Kringle 2 (SEQ ID NO: 12) corresponds to amino acid positions 185 to 262 (inclusive) of murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 88 to 165 (inclusive) of murine angiostatin of SEQ ID NO: 2. Human kringle 2 (SeQ ID NO: 13), Rhesus monkey kringle 2 (SEQ ID NO: 14), porcine kringle 2 (SEQ ID NO: 151 and bovine kringle 2 (SEQ ID NO: 16) correspond to the amino acid positions 88 to 165 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 3" means a protein derivative of plasminogen which has anti-angiogenic activity or endothelial cell inhibitory activity, and which has an amino acid sequence comprising a sequence homologous to kringle 3, exemplified using, but not limited to, murine kringle 3 (SeQ ID NO: 17), human kringle 3 (SEQ ID NO: 18), Rhesus monkey Kringle 3 (SEQ ID NO: 19), porcine kringle (SEQ ID NO : 20) and bovine kringle 3 (SEQ ID NO: 21). Murine Kringle 3 (SEQ ID NO: 17) corresponds to amino acid positions 275 to 352 (inclusive) of the murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 178 to 255 (inclusive) of murine angiostatin of SEQ ID NO: 2. Human kringle 3 (SEQ ID NO: 18), Rhesus monkey kringle 3 (SEQ ID NO: 1 9), porcine kringle 3 (SEQ ID NO: 20) and bovine kringle 3 (SEQ ID NO: 21) correspond to the positions amino acids 178 to 255 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 4" means a protein derivative of plasminogen having anti-angiogenic activity or endothelial cell inhibitory activity, and having an amino acid sequence comprising a sequence homologous to kringle 4, exemplified by, but not limited to, murine kringle 4 (SEQ ID NO: 22) and human kringle 4 (SEQ ID NO: 23), unless otherwise indicated by the context in which it is used. Murine Kringle 4 (SEQ ID NO: 22) corresponds to amino acid positions 377 to 454 (inclusive) of the murine plasminogen of SEQ ID NO: 1.

As used herein, "kringle 2-3" means a protein derivative of plasminogen which has anti-angiogenic activity or endothelial cell inhibitory activity, and which has an amino acid sequence comprising a sequence homologous to kringle 2. -3, exemplified by, but not limited to, murine kringle 23 (SEQ ID NO: 24), human kringle 2-3 (SEQ ID NO: 25), Rhesus monkey kringle 2-3 (SEQ ID NO: 26) , kringle 2-3 porcine (SEQ ID NO: 27) and bovine kringle 2-3 (SEQ ID NO: 28), unless otherwise indicated by the context in which it is used. Murine Kringle 2-3 (SEQ ID NO: 24) corresponds to amino acid positions 185 to 352 (inclusive) of the murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 88 to 255 (inclusive) of angiostatin murine of SEQ ID NO: 2. Human kringle 2-3 (SEQ ID NO: 25), Rhesus monkey kringle 2-3 (SEQ ID NO: 26), porcine kringle 2-3 (SEQ ID NO: 27) and bovine kringle 2-3 (SEQ ID NO : 28) correspond to amino acid positions 88 to 255 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 1-3" means a protein derivative of plasminogen which has anti-angiogenic activity or endothelial cell inhibitory activity, and which has an amino acid sequence comprising a sequence homologous to kringle 1 -3, exemplified by, but not limited to, that of murine kringle 13; SEQ ID NO: 29), human kringle 1 (SEQ ID NO: 30), Rhesus monkey kringle 1-3 (SEQ ID NO: 31), porcine kringle 1-3 (SEQ ID NO: 32) and kringle 1-3 bovine (SEQ ID NO: 33), unless otherwise indicated by

ES 2 292 174 T3 the context in which it is used. Murine Kringle 1-3 (SEQ ID NO: 29) corresponds to amino acid positions 103 to 352 (inclusive) of the murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 6 to 255 (inclusive) of angiostatin murine of SEQ ID NO: 2. Human kringle 1-3 (SEQ ID NO: 30), Rhesus monkey kringle 1-3 (SEQ ID NO: 31), porcine kringle 1-3 (SEQ ID NO: 32) and bovine kringle 1-3 (SEQ ID NO : 33) correspond to amino acid positions 6 to 255 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 1-2" means a protein derivative of plasminogen which has anti-angiogenic activity or endothelial cell inhibitory activity, and which has an amino acid sequence comprising a sequence homologous to kringle 1 -2, exemplified by, but not limited to, murine kringle 12 (SEQ ID NO: 34), human kringle 1-2 (SEQ ID NO: 35), Rhesus monkey kringle 1-2 (SEQ ID NO: 36 ), kringle 1-2 porcine (SEQ ID NO: 37) and bovine kringle 1-2 (SEQ ID NO: 38), unless otherwise indicated by the context in which it is used. Murine Kringle 1-2 (SEQ ID NO: 34) corresponds to amino acid positions 103 to 262 (inclusive) of the murine plasminogen of SEQ ID NO: 1, and corresponds to amino acid positions 6 to 165 (inclusive) of angiostatin murine of SEQ ID NO: 2. Human kringle 1-2 (SEQ ID NO: 35), Rhesus monkey kringle 1-2 (SEQ ID NO: 36), porcine kringle 1-2 (SEQ ID NO: 37) and bovine kringle 1-2 (SEQ ID NO : 38) correspond to amino acid positions 6 to 165 (inclusive) of angiostatin of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, respectively.

As used herein, "kringle 1-4" means a protein derivative of plasminogen which has anti-angiogenic activity or endothelial cell inhibitory activity, and which has an amino acid sequence comprising a sequence homologous to kringle 1 -4, exemplified by, but not limited to, that of murine kringle 1-4 (SEQ ID NO: 39) and human kringle 1-4 (SEQ ID NO: 40), unless otherwise indicated by the context in the one used. Murine Kringle 1-4 (SEQ ID NO: 39) corresponds to amino acid positions 103 to 454 (inclusive) of the murine plasminogen of SEQ ID NO: 1.

The amino acid sequences of kringle 1, kringle 2, kringle 3, kringle 4, kringle 2-3, kringle 1-3, kringle 12 and kringle 1-4, are homologous, respectively, to the specific kringle sequences identified above. Preferably, the amino acid sequences have a degree of homology to the disclosed sequences of at least 60%, more preferably at least 70%, and more preferably at least 80%. It should be understood that a variety of amino acid substitutions, additions, deletions or other modifications can be made to the fragments listed above to enhance or modify the anti-angiogenic activity or endothelial cell proliferation inhibitory activity of the angiostatin fragments. The term "functional homologue" means that altered sequences can be used according to the invention that includes deletions, additions or substitutions of different residues, resulting in a sequence that codes for the same gene product or a functionally equivalent one. The gene product itself may contain deletions, additions, or substitutions of amino acid residues within an angiostatin sequence, resulting in a silent change, thus producing a functionally equivalent angiostatin protein. Such amino acid substitutions can be made on the basis of similarity in polarity, charge, solubility, hydrophobicity, hydrophilicity, and / or the amphipathic nature of the residues involved. For example, negatively charged amino acids include aspartic acid and glutamic acid; positively charged amino acids include lysine, histidine, and arginine; amino acids with uncharged polar head groups having similar hydrophilicity values ​​include the following: glycine, asparagine, glutamine, serine, threonine, tyrosine; and amino acids with nonpolar head groups include alanine, valine, isoleucine, leucine, phenylalanine, proline, methionine, tryptophan. Such modifications are not intended to exceed the scope and spirit of the claims. For example, to avoid homodimerization by disulfide bond formation between kringles, cysteine ​​residues C4 in recombinant human kringle 2 (SEQ ID NO: 13) and C42 in recombinant kringle 3 (SEQ ID NO: 18) were mutated to serines. Furthermore, it is understood that a variety of amino acid substitutions, additions, deletions, or other modifications can be made to the angiostatin fragments identified above, which do not significantly alter the endothelial cell proliferation inhibitory activity of the fragments, and therefore, They are not intended to exceed the scope of the claims. By "do not significantly alter" is meant that the angiostatin fragment has at least 60%, more preferably at least 70% and more preferably at least 80% of the endothelial cell proliferation inhibitory activity compared to that of the closest homologous angiostatin fragment disclosed herein.

Gene expression and construction

A PCR-based method was used to generate the cDNA fragments encoding kringle 1 (K1), kringle 2 (K2), kringle 3 (K3), kringle 4 (K4) and kringle 2-3 (K2-3) of the human plasminogen (HPg). Recombinant kringle 1 (K1r), kringle2 (K2r), kringle 3 (K3r), kringle 4 (K4r), and kringle 2-3 (K2-3r) were expressed in E. coli, as previously described (Menhart, N. , Shel, LC, Kelly, RF, and Castellino, FJ (1991) Biochem. 30, 1948-1957; Marti, D., Schaller, J., Ochensberger, B., and Rickli, EE (1994) Eur. J. Biochem. 219,455-462; Sohndel, S., Hu, C.-K., Marti, D., Affolter, M., Schaller, J., Llinas, M., and Rickli, EE (1996) Biochem. in press; Rejante, MR, Byeon, LJ. L., and Llinas, M. (1991) Biochem. 30, 11081-11092). To avoid homodimerization through the formation of disulfide bridges between kringle as shown in figure 32B, the cysteary residues C169 in K2r and C297 in K3r were mutated to serines, as observed in SEQ ID NOS 13 and 18, at positions 4 and 42, respectively. (Sohndel, S., Hu, C.-K., Marti, D., Affolter, M., Schaller, J., Llinas, M., and Rickli, EE (1996) Biochem. In press). The K3r and K2-3r contained a hexa-histidine tail at the N-terminus that was used for protein purification (not shown).

ES 2 292 174 T3

Proteolytic digestion

K1-3, K1-4, and K4 fragments were prepared by digestion of Lys-HPg (Abbott Labs) with porcine elastase (Sigma), as previously described (Powell, JR, and Castellino, FJ (1983) Biochem. 22, 923927). Briefly, 1.5 mg of elastase was incubated at room temperature with 200 mg of human plasminogen in 50 mM Tris-HCl pH 8.0 overnight with shaking. The reaction was terminated by the addition of diisopropyl fluorophosphate (DFF) (Sigma) to a final concentration of 1 mM. The mixture was shaken for an additional 30 minutes at room temperature and dialyzed overnight against 50 mM Tris-HCl, pH 8.0. Protein purification

Recombinant K1 was expressed in E. coli DH5a bacterial cells using a plasmid vector pSTII. This protein was purified to homogeneity by chromatography using Lysine-Sepharose 4B (Pharmacia) and Mono Q (BioRad) columns. Bacterial cells of E. coli (strain HB101) expressing K2r and K3r were grown at OD₆₀₀ at about 0.8 at 3 ° C in 2 x YT medium containing 100 mg / ml ampicillin and 25 mg / ml kanamycin. IPTG (isopropyl-bD-thiogalactopyranoside) was added to a final concentration of 1 mM and cells were grown for an additional 4.5 hours at 37 ° C to induce recombinant protein production. Cells were harvested by centrifugation and pellets were stored at -80 ° C. The thawed cell lysates were resuspended in the extraction buffer (6M guanidine hydrochloride in 0.1M sodium phosphate, pH 8.0). The suspension was centrifuged at 15,000 xg for 30 minutes and b-mercaptoethanol was added to the supernatant to a final concentration of 10 mM. The supernatant was then loaded onto a Ni agarose column.²⁺-MrA (1.5 cm x 5 cm) pre-equilibrated with the extraction buffer. The column was washed successively with extraction buffer at pH 8.0 and pH 6.3, respectively. Recombinant K2 and K3 were eluted with extraction buffer at pH 50.

The proteolytically cleaved K1-3, K1-4 and K4 fragments were purified using a column of lysine Sepharose 4B (2.5 cm x 15 cm) equilibrated with 50 mM Tris-HC1, pH 8.0 until absorbance was reached at 180 nm of 0.005. The absorbed kringle fragments were eluted with Tris buffer containing 200 mM ε-aminocrapoic acid, pH 8.0. Eluted samples were dialyzed overnight against 20 mM Tris-HCl, pH 5.0, and applied to a BioRad Mono-S column equilibrated with the same buffer. K4, K1-3 and K1-4 fragments were eluted with step gradients of 0-20%, 20-50% and 50-70% 20 mM phosphate / 1M KCl, pH 5.0. Most of the K1-3 and K1-4 fragments were eluted from the column with 0.5 M KCl as determined by SDS-PACE. All fractions were dialyzed overnight against 20 mM Tris-HCl, pH 8.0. After dialysis, the K1-3 and K1-4 fragments were further purified using a Heparin-Sepharose column (5 cm x 10 cm) (Sigma) pre-equilibrated with 20 mM Tris-HCl buffer, pH 8.0. The K1-3 fragment was eluted with 350 mM KCl and K1-4 was recovered from the flow-through fraction. Purified kringle fragments were analyzed on SDS gels followed by silver staining, by Western blot analysis with polyclonal anti-human K4 and K1-3 antibodies, and by amino-terminal end sequencing analysis. In vitro refolding

Refolding of K2r, K3r and K2-3r was carried out according to a standard protocol (Cleary, S., Mulkerrin, MG, and Kelley, RR (1989) Biochem. 28, 1884-1891). The purified proteins were adjusted to pH 8.0 and dithiothreitol (DTT) was added to a final concentration of 5 mM. Following an overnight incubation, the solution was diluted with 4 volumes of 50 mM Tris-HCl, pH 8.0, containing 1.25 mM reduced glutathione. After 1 hour of incubation, oxidized glutathione was added to a final concentration of 1.25 mM and incubated for 6 hours at 4 ° C. The re-naturalized protein was dialyzed initially against H2O for 2 days and for an additional two days against 50 mM phosphate buffered saline, pH 8.0. The solution was then loaded onto a Lysine-Bio-Gel column (2 cm x 13 cm) equilibrated with the same phosphate buffered saline. The column was washed with phosphate buffered saline and the protein was eluted with a phosphate buffer containing 50 mM 6-AHA (6-aminohexanoic acid). Reversed phase HPLC was carried out on an Aquapore Butyl column (2.1 x 100 mm, 30 nm widepore, 7 mm, Applied Biosystems) and a Hewlett Packard liquid chromatography with acetonitrile gradients was used. .

Reduction and alkylation

Reduction and alkylation of kringle fragments were carried out according to a standard protocol (Cao, Y., and Pettersson, RF, (1990) Growth Factors 3, 1013). Approximately 20-80 mg of the purified proteins in 300-500 ml of DME medium in the absence of serum were incubated at room temperature with 15 ml of 0.5 M DTT for 15 minutes. After incubation, 30 ml of 0.5 M iodoacetamide was added to the reaction. The protein solution was dialyzed at 4 ° C overnight initially against 20 volumes of DMEM. The solution was further dialyzed at 4 ° C for an additional 4 hours against 20 volumes of fresh DMEM. After dialysis, the samples were analyzed on an SDS gel and tested for their inhibitory activities on endothelial cell proliferation.

Endothelial proliferation assay

Bovine capillary endothelial cells (BCE) were isolated as previously described (Folkman, J., Haudenschild, CC, and Zetter, BR (1979) Proc. Natl. Acad. Sci USA. 76, 5217-5121) and maintained at DMEM

ES 2 292 174 T3 supplemented with 10% heat inactivated calf serum (BCS), antibiotics, and 3 ng / ml recombinant human bFCF (Scios Nova, MountainvLew, CA). The BCE cell growth monolayers in 6-well plates were dispersed in a 0.05% trypsin solution. Cells were resuspended with DMEM containing 10% BCS. Approximately 12,500 cells in 0.5 ml were added to each well of the 24-well gelatinized tissue culture plates and incubated at 37 ° C (in CO₂ 10%) for 24 hours. The medium was replaced with 500 ml of fresh DMEM containing 5% BCS and individual or combined kringle fragment samples were added in triplicate to each well. After 30 minutes of incubation, bFGF was added to a final concentration of 1 ng / ml. After 72 hours of incubation, cells were trypsinized, resuspended in Hematall (Fisher Scientific, Pittsburg, PA), and counted with a Coulter counter.

Purification and characterization of the kringle fragment of human plasminogen

The cDNA fragments encoding individual kringles (K1, K2, K3, and K4) and kringles 2-3 (K2-3) were amplified from human plasminogen by a PCR-based method (Figure 28). The PCR amplified cDNA fragments were cloned into a bacterial expression vector. Recombinant proteins expressed from Escherichia coli were refolded in vitro and purified to&gt; 98% homogeneity using HPLC-coupled chromatography (Figure 29). Under reducing conditions, recombinant K2, K3, and K4 migrated with molecular weights of 12-13 kDa (Figure 29A, lanes 2-4), corresponding to the predicted molecular weights of each kringle fragment. Recombinant K1 migrating with a higher molecular weight of 17 kDa was identified by SDS gel electrophoresis. Fragments of K1-4 and K1-3 were obtained by proteolytic digestion of Lys-human plasminogen (Lys-HPg) with elastase, as previously described (Powell, JR, and Castellino, FJ (1983) Biochem. 22, 923-927; Brockway, WJ, and Castellino, FJ (1972) Arch. Biochem Biophys). These two fragments (Figure 29B, lanes 1 and 2) with predicted molecular weights of 43 kDa and 35 kDa, respectively, were also purified to homogeneity. Analysis of the amino acid sequence of the N-terminal end of the purified fragments gave an identical sequence, -YLSE-, followed by SEQ ID NO: 30 and SEQ ID NO: 40, for K1-3 and K1-4, respectively. The N-terminal sequence for K4 produced -VVQD- with approximately 20% -VQD-, followed by SEQ ID NO: 23, each of which is predicted from the expected sequence starting with Valine¹⁷⁶ and Valine¹⁷⁷of human angiostatin (SEQ ID NO: 3).

Anti-proliferative activity in endothelial cells for individual kringles

Individual recombinant kringle fragments of angiostatin were assayed for growth inhibitory activities of bovine capillary endothelial cells (BCE) by bFGF. As shown in Figure 30A, K1r inhibited BCE cell proliferation in a dose-dependent manner. The concentration of K1r required to achieve 50% inhibition (DE₅₀) was approximately 320 nM (Table 4). In contrast, K4r showed little or no inhibitory effect on endothelial cell proliferation. Recombinant K2 and K3r, two fragments of kringles that do not bind lysine, also produced a dose-dependent inhibition of endothelial cell proliferation (Figure 30B). However, the inhibitory potency of K2r was substantially lower than that of K1r and K3r (DE₅₀ = 460) (figure 30 and table 4). No cytotoxicity or well differentiated morphology could be detected with apoptotic endothelial cells such as rounding, shedding and cell fragmentation, even after incubation with a high concentration of these kringle fragments. These data suggest that the anti-endothelial growth activity of angiostatin may be shared by K1, K2 and K3 fragments, and to a lesser extent by K4.

TABLE 4

Inhibitory activity on capillary endothelial cell proliferation

Fragments

OF₅₀ (nM)

Kringle

Kringle

Kringle

Kringle

Kringle

Kringle

Kringle

2-3

1-3

320

460

1-4 (angiostatin) 135

ES 2 292 174 T3

Anti-proliferative activity in endothelial cells of K1-3 and K1-4 fragments

To evaluate the anti-proliferative effect on endothelial cells of pooled kringle fragments, purified proteolytic fragments of human K1-4, K1-3 and K2-3r were assayed in BCE cells. Consistent with previous findings (O'Reilly, MS, Holmgren, L., Shing, Y., Chen, C., Rosenthal, RA, Moses, J., Lane, WS, Cao, Y., Sage, EH, and Folkman, J. (1994) Cell 79, 315-328), the proliferation of BCE cells, as shown in Figure 31, was significantly inhibited by the angiostatin-like fragment of K1-4 (DE₅₀ = 135 nM) (table 4). An increase in endothelial anti-growth activity was obtained with the K1-3 fragment (DE₅₀ = 70 nM) (table 4). Inhibition of endothelial cell proliferation occurred in a dose-dependent manner. These results indicate that removal of K4 from angiostatin enhances anti-endothelial growth activity. Additive inhibition by K2ry K3r

The K2-3r fragment showed only weak inhibitory activity that was similar to that of K2r alone (Figure 31). However, both K2r and K3r inhibited endothelial cell proliferation (Figure 30B). This finding suggests that the inhibitory effect of K3 was hidden in the structure of K2-3. Previous structural studies showed that a disulfide bridge was present between kringles between K2 (cysteine¹⁶⁹) and K3 (cysteine²⁹⁷) of human plasminogen, corresponding to cysteine⁹¹ and cysteine²¹⁹ of SEQ ID NO: 3 (Sohndel, S., Hu, C.-K., Marti, D., Affolter, M., Schaller, J., Llinas, M., and Rickli, EE (1996) Biochem. in press) See Figure 32B. The inhibitory effect of K2r and K3r in combination was tested. Interestingly, additive inhibition was observed when individual K2r and K3r fragments were added together to BCE cells. See Figure 32A. These results imply that it is preferable to open the disulfide bridge between K2 and K3 in order to obtain the maximum inhibitory effect of K2-3.

Proper folding of kringle structures is required for anti-endothelial activity of angiostatin

To study whether folding of kringle structures is required for anti-endothelial proliferation activity, natural angiostatin was reduced with DTT and tested on bovine capillary endothelial cells. After reduction, angiostatin was further alkylated with iodoacetamide and analyzed by SDS gel electrophoresis. As shown in Figure 34A, the DTT-treated protein migrated to a higher position with a molecular weight of approximately 42 kDa (lane 2) compared to wild-type angiostatin with a molecular weight of 33 kDa (lane 1), which suggesting that angiostatin was completely reduced. The anti-proliferation activity of angiostatin was largely suppressed upon reduction (Figure 34B). From these results, it is concluded that the correct folding of angiostatin through the disulfide bridges between kringles is preferable to maintain its potent effect on the inhibition of endothelial cell proliferation.

The alignment of the amino acid sequence of the kringle domains of human plasminogen shows that K1, K2, K3 and K4 show a remarkable sequence homology and an identical macroscopic architecture (56-82% identity) as observed in figures 35 Among these structures, the high affinity lysine-binding kringle, K1, is the most potent inhibitory segment of endothelial cell proliferation. It is of interest that the intermediate affinity lysine-binding fragment, K4, lacks inhibitory activity. These data suggest that the lysine binding site of kringle structures may not be directly involved in inhibitory activity. The conservation of amino acids and the functional divergence of these kringle structures provide an ideal system for studying the role of mutations produced by DNA replication during evolution. Similar divergent activities are also found with respect to the regulation of angiogenesis exhibited by a group of structurally related proteins in the growth hormone families - prolactin and chemokine -CXC- (Maione, TE, Gray, GS, Petro, AJ, Hunt , AL, and Donner, SI (1990) Science 247, 77-79 .; Koch, AE, Polverini, PJ, Kunkel, SL, Harlow, LA, DiPietro, LA, Elner, VM, Elner, SJ, and Strieter, RM (1992) Science 258, 1798-1801 .; Cáo, Y., Chen, C., Weatherbee, JA, Tsang, M., and Folkman, J. (1995) J. Exp. Med. 182, 2069-2077 .; Strieter, RM, Polverini, PJ, Arenberg, DA, and Kunkel, SL (1995) Shock 4, 155-160 .; Jackson, D., Volpert, OV, Bouck, N., and Linzer, DIH (1994 Science 266, 1581-1584).

Further sequence analysis reveals that K4 contains two positively charged lysine residues adjacent to cysteines 22 and 78 (Figure 35). Nuclear magnetic resonance (NMR) analysis of¹H shows that these 4 lysines, together with lysine 57, form the core of a positively charged domain in K4 (Lins M, unpublished data), while other kringle structures lack such a positively charged domain. Whether this lysine-enriched domain contributes to the loss of kringle 4 inhibitory activity from human plasminogen remains to be studied. K4 was previously reported to stimulate the proliferation of other cell types and increase intracellular calcium release (Donate, LE, Gherardi, E., Srinivasan, N., Sowdhamini, R., Aporicio, S. and Blundell, TL (1994 ) Prot. Sci. 3, 2378-2394). The fact that the removal of K4 from angiostatin enhances its inhibitory activity on endothelial cells suggests that this structure may avoid some of the inhibitory effect of K1-3.

The mechanism underlying how angiostatin and its related kringle fragments specifically inhibit endothelial cell growth remains uncharacterized. It is not yet clear whether the inhibition is mediated by a receptor that is specifically expressed on proliferating endothelial cells, or whether angiostatin is internalized by endothelial cells and subsequently inhibits cell proliferation. Alternatively, angiostatin can interact with an endothelial cell adhesion receptor, such as integrin a_vb₃, which blocks integrin-mediated angiogenesis (Brooks, PC, Montgomery, AM, Rosenfeld, M., Reisfeld RA, Hu, T. Klier, G., and Cheresh, DA (1994) Cell 79, 1157-1164). It is of interest that Friedlander et. to the. (Friedlander, M., Brooks, P.

ES 2 292 174 T3

C., Shaffer, RW, Kincaid, CM, Varner, JA and Cheresh, DA (1995) 270, 1502) recently reported that in vivo angiogenesis in chorioallantoic or corneal membrane models (induced by bFGF by tumor necrosis factor ) was dependent on integrin a_vb₃. However, VEGF-stimulated angiogenesis, transforming growth factor a, or phorbol esters was dependent on a_vb₅. Antibodies to individual integrins specifically blocked one of these pathways, and a cyclic protein antagonist of both integrins blocked the angiogenesis induced by each of the cytokines (Friedlander, M., Brooks, PC, Shaffer, RW, Kincaid, CM, Varner, JA and Cheresh, DA (1995) 270, 1502). Since bFGF and VEGF-induced angiogenesis are inhibited by angiostatin, it can block a common pathway for this integrin-mediated angiogenesis.

In recent decades an increasing number of endogenous angiogenesis inhibitors have been identified (Folkman, J. (1995) N. Engl. J. Med. 333, 1757-1763). Of the nine characterized endothelial cell suppressors, some inhibitors are proteolytic fragments. For example, the N-terminal 16 kDa fragment of human prolactin inhibits endothelial cell proliferation and blocks angiogenesis in vivo (Clapp, C., Martial, JA, Guzman, RC, Rentierdelrue, F., and Weiner, RI (1993) Endorinology 133, 1292-1299). In a recent article, D'Angelo et. to the. reported that the 16 kDa anti-angiogenic fragment at the N-terminus inhibited the activation of mitogen-activated protein kinase (MAPK) by VEGF and bFGF in capillary endothelial cells (D'Angelo, G., Struman, I. , Martial, J., and Weiner, R. (1995) Proc. Natl. Acad. Sci. 92, 6374-6378). Similar to angiostatin, the intact parent prolactin molecule does not inhibit endothelial cell proliferation nor is it an inhibitor of angiogenesis. Platelet factor 4 (PF-4) inhibits angiogenesis at high concentrations (Maione, TE, Gray, GS, Petro, AJ, Hunt, AL and Donner, SI (1990) Science 247, 77-79; Cao, Y., Chen, C., Weatherbee, JA, Tsang, M. and Folkman, J. (1995) J. Exp. Med. 182, 2069-2077). However, the proteolytically cleaved PF-4 fragment truncated at the N-terminus shows a 30-50-fold increase in its anti-proliferative activity relative to the intact PF-4 molecule (Gupta, SK, Hasse.l , T. and Singh, JP (1995) Proc. Natl. Acad. Sci. 92, 7799-7803). Smaller protein fragments of fibronectin, murine epidermal growth factor, and thrombospondin have also been shown to specifically inhibit endothelial cell growth (Homandberg, GA, Williams, JE, Grant, D., Schumacher, B., and Eisenstein, R. ( 1985) Am. J. Pathol. 120, 327-332; Nelson, J., Allen, WE, Scott, WN, Bailie, JR, Walker, B., McFerran, NV, and Wilson, DJ (1995) Cancer Res. 55, 3772-3776; Tolsma, SS, Volpert, OV, Good, DJ, Frazer, W. A., Polverini, PJ, and Bouck, N. (1993) J. Cell Biol. 122,497-511). Proteolytic processing of a large protein can change the structure and conformation of the original molecule or expose new epitopes that are anti-angiogenic. Thus, protease (s) may play a critical role in the regulation of angiogenesis. To date, little is known about the regulation of these protease activities in vivo.

The data also show that the disulfide bridge-mediated folding of kringle structures in angiostatin is preferable to maintain its inhibitory activity on endothelial cell growth. Kringle structures analogous to these of plasminogen are also found in a variety of different proteins. For example, apolipoprotein (a) has up to 37 plasminogen kringle 4 repeats (McLean, JW, Tomlinson, JE, Kuang, W.-J., Eaton, DL, Chen, E. Y., Fless, GM, Scanu, AM, and Lawn, RM (1987) Nature 330, 132-137). The amino terminal part of prothrombin also contains two kringles that are homologous to those of plasminogen (Wíilz, DA, Hewett-Emmett, D., and Seegers, W. H, (1977) Proc. Natl. Acad. Sci. 74, 1969-1973). Urokinase has been shown to possess a kringle structure that shares broad homology with plasminogen (Gunzler, WA, J., SG, Otting, F., Kim, S.-MA, Frankus, E., and Flohe, L. (1982) Hoppe-Seyler's A. Physiol. Chem. 363, 11551165). In addition, surfactant protein B and hepatocyte growth factor (HGF) also carry kringle structures (Johansson, J., Curstedt, T., and Jornvall., H. (1991) Biochem. 30, 6917-6921; Lukker , NA, Presta, LG, and Godowski, PJ (1994) Prot. Engin. 7, 895-903).

Example 28

Suppression of endothelial cell proliferation and metastasis by angiostatin fragments

The following example characterizes the activity of additional angiostatin fragments. The data suggest that potent tumor suppressor and anti-endothelial activity can be obtained from such angiostatin protein fragments.

As used herein, "kringle 1-4BKLS" means a protein derivative of plasminogen having endothelial cell inhibitory activity, and having an amino acid sequence comprising a sequence homologous to kringle 1-4BKLS, exemplified by , but not limited to murine kringle 1-4BKLS (SEQ ID NO: 41), and human kringle 1-4BKLS (SEQ ID NO: 42), unless otherwise indicated by the context in which it is used. Murine Kringle 1-4BKLS (SEQ ID NO: 41) corresponds to amino acid positions 93 to 470 (inclusive) of the murine plasminogen of SEQ ID NO: 1. This example demonstrates that an "angiostatin fragment" can be a plasminogen fragment. and which encompasses an amino acid sequence greater than angiostatin presented in SEQ ID NO: 3, for example, and which still has therapeutic anti-angiogenic activity or endothelial cell proliferation inhibitory activity.

An amino acid sequence of kringle 1-4BLKS is homologous to the specific kringle 1-4BLKS sequences identified above. Preferably, the amino acid sequences have a degree of homology to the described sequences of at least 60%, more preferably at least 70%, and more preferably at least 80%. It should be understood that a variety of amino acid substitutions, deletions, and other modifications can be made.

ES 2 292 174 T3 two to the fragments listed above to enhance or modify the endothelial cell inhibitory activity of the fragments. Such modifications are not intended to exceed the scope and spirit of the claims. Furthermore, it is understood that a variety of silent amino acid substitutions, additions, or deletions can be made in the kringle fragments identified above, which do not significantly alter the endothelial cellular inhibitory activity of the fragments, and are therefore not intended to exceed the scope of the claims. Cloning of angiostatin in Pichia pastoris

Angiostatin coding sequences were amplified by PCR using Vent polymerase (New England Biolabs) and primers number 154 (5'-ATCGCTCGAGCGTTATTTGAAAAGAAAGTG-3 ') (SEQ ID NO: 43) and number 151 (5'-ATCGGAATTCAAGCAGGACA-AG 3 ') (SEQ ID NO: 44) containing the XhoI and Eco RI linkers, respectively, and using the plasmid pTrcHis / HAs as a template. This plasmid contained sequences that code for amino acids 93 to 470 of human plasminogen (SEQ ID NO: 42) for its Cloning in the Xho I / ECo RI site of the expression vector pHIL-S1 using the natural secretion signal of P. pastoris , PHO 1. This same sequence was amplified in the same way using primers number 156 (5'-ATCGTACG TATTATTTGAAAAGAAAGTG-3 ') (SEQ ID NO: 45) and number 151 containing the Sna BI and Eco RI linkers, respectively, for their cloning into the Sna BI / ECo RI site of the expression vector pPIC9 with the alpha factor secretory signal. The amplification products were gel purified, the linkers digested with the appropriate enzymes and purified again using Gene-Clean (Bio 101). These gene fragments were ligated into the appropriate vectors. The resulting clones were selected and plasmid preparations of the clones were obtained and linearized to generate the recombinant His' Mut strains.^s and His + Mut + when transformed into the host strain of P. pastoris, GS115. Integration was confirmed by PCR.

Both recombinant His⁺ and His⁺ Mut⁺ were induced with methanol and detected for high expression of angiostatin using SDS-PAGE gels stained with Coomassie blue and immunoblots using a mouse monoclonal antibody against kringles 1 to 3 (Castellino, Enzyme Research Laboratories, Inc., South Bend , IN). From these, a transformed P pastoris GS115 clone, pHIL-S1 / HAs18, was selected and phenotypically characterized as His⁺ Mut^s.

PHIL-S1 / HAs18 expression

Angiostatin expression from pHIL-S1 / HAs18 was typical for a His + Mut clone^s. On induction in baffled shake flasks, 1 μl of cells were cultured at OD₆₀₀ in 150 of buffered methanol complexed medium containing 1% yeast extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base with ammonium sulfate, 0 biotin .00004% and 0.5% methanol in a 11-baffled flask. The cells were constantly shaken at 30 ° C, 250 rpm. Methanol was batch fed at 24 hour intervals by adding absolute methanol to a final concentration of 0.5%. After 120 hours, the cells were centrifuged at 5,000 rpm for 10 minutes, and the supernatants were stored at -70 ° C until used.

Purification of Angiostatin from P. pastoris Fermentation Broth by Lysine Chromatography Sepharose

All procedures were carried out at 4 ° C. The crude fermentation broth, typically 200 ml, containing angiostatin, was clarified by centrifugation at 14,000 xg and concentrated through a 30 kDa molecular weight cutoff membrane, Centriprep 30 (Amicon), to approximately one quarter of the original volume. A volume of 50 mM phosphate buffer, pH 7.5, was added to the concentrated sample which was again concentrated by Centriprep to a quarter of the original sample volume. The sample was diluted again in volume: volume with 50 mM sodium phosphate buffer, pH 7.5. 60 g of lysine-Sepharose 4B (Pharmacia) were resuspended in 500 ml of ice cold 50 mM phosphate buffer, pH 7.5 and they were used to pack a 48 x 100 mm column (packing volume -180 ml). The column was washed overnight with 7.5 column volumes (CV) of 50 mM sodium phosphate buffer, pH 7.5, at a flow rate of 1.5 ml / min. The sample was pumped onto the column at a flow rate of 1.5 ml / min and the column was washed with 1.5 CV of 50 mM sodium phosphate, pH 7.5, at a flow rate of 3 ml / min. min. The column was then washed with 1.5 CV of phosphate buffered saline, pH 7.4, at a flow rate of 3 ml / min: angiostatin was then eluted with 0.2 ε-amino-n-caproic acid. M, pH 7.4 at a flow rate of 3 ml / min. Fractions containing significant absorbance were pooled and dialyzed for 24-48 hours against deionized water and lyophilized. A typical recovery from a 100 mg total protein load is 10 mg of angiostatin. The columns were regenerated using 5 column volumes of 50 mM sodium phosphate / 1 M NaCl, pH 7.5.

Bovine Capillary Endothelial Cell Proliferation Assay

Bovine capillary endothelial cells were obtained as described above. Cells were maintained in DMEM containing 3 mg / ml recombinant human bFGF (Scios Nova, Mountainview, CA), supplemented with 10% heat-inactivated fetal bovine serum, 100 U / ml penicillin, 100 mg / ml streptomycin, and fungizone. 0.25 mg / ml (BioWhittaker) in 75 cm cell culture flasks². The test was carried out as described above.

ES 2 292 174 T3

Animal studies

Six to eight week old male C57BI / 6J mice (Jackson Laboratories) were inoculated subcutaneously with the murine low metastatic Lewis lung carcinoma (LLC-LM) line (1 x 10⁶ cells / injection). Approximately 14 days after implantation, when the primary tumor reached 1.5 cm³, the animals were anesthetized with methoxyflurane and the primary tumors were surgically removed. The incision site was closed with simple interrupted suture. Half of the animals in this group received a loading dose (3 mg / kg subcutaneously) of plasminogen-derived or recombinant angiostatin subcutaneously immediately after surgery, followed by daily inoculations of 1.5 mg / kg during 14 days. A control group of mice received an equal volume of PBS each day for 14 appointments after surgery. All mice were sacrificed 14 days after removal of the primary tumor (28 days after tumor implantation), the lungs were removed and weighed, and surface metastases were counted with a stereomicroscope.

Characteristics of recombinant human angiostatin fragments

A gene fragment encoding human angiostatin including human plasminogen kringles 1 to 4 containing a total of 26 cysteines was expressed in Pichia pastoris, the methylotropic yeast. Angiostatin expressed in P. pastoris binds to lysine-Sepharose and can be specifically eluted by ε-aminocaproic acid. This demonstrates that fully functional epsilon-aminocaproic acid binding kringle (s), which are physical properties of plasminogen kringle 1 and 4 (Sottrup-Jensen, L. et al., Progress in Chemical Fibrinolysis and Thrombolysis, Vol. 3 (1978) Ravens Press, NY p. 191), can be expressed and secreted by P pastoris and purified by techniques that do not require refolding (Figure 36A and B). A conformationally dependent monoclonal antibody to Kringle 1 to 3 recognized angiostatin expressed from P. pastoris, as well as angiostatin purified by elastase cleavage of plasminogen (Castellano, Enzyme Research Laboratories, Inc., SouthBend, IN ) (Figure 36B). This antibody does not recognize the reduced forms of plasminogen or angiostatin.

Angiostatin expressed by P. pastoris is observed as a doublet migrating at 49 kDa and 51.5 kDa on Coomassie-stained denatured, non-reduced SDS-PAGE gels. Proteins expressed by P pastoris are post-translationally modified with most N-glycosylation of the rich mannose type and negligible O-glycosylation. To evaluate the possibility of glycosylation in angiostatin expressed by P. pastoris, recombinant angiostatin was digested with endoglycosidase H specific for mannose-rich structures, causing: The 51.5 kDa band to migrate identically with the 49 kDa band (FIG. 37A and B). O-glycanase digestion with pretreatment with neuraminidase to remove sialic acid residues did not change the doublet migration pattern (data not shown). These results indicate that P. pastoris expressed angiostatin in two ways: (1) with a complex chain linked to N probably from the structure:

(Man) 2-l50Man

Man-GlcNAc-GlcNAc-Asn (Man) i-2-Man and (2) without any glycosylation.

Inhibition of bovine capillary endothelial cells in vitro

To determine whether recombinantly expressed angiostatin had the potential for anti-angiogenic activity, BCE was cultured in the presence of bFGF to determine whether the addition of purified recombinant angiostatin would inhibit BCE proliferation. Purified angiostatin expressed by P. pastoris inhibited bFGF-directed proliferation of bovine endothelial cells in vitro (Figure 38B) in a dose-dependent manner (Figure 38C). At 1 ug / ml of recombinant angiostatin, the inhibition was 80%. The 50% inhibition was equivalent to that obtained with angiostatin derived from elastase cleavage of human plasminogen.

In vivo metastasis suppression

The transplantable murine LLC (LM) line from which angiostatin was first identified was used. When implanted subcutaneously into syngeneic C57B1 / 6J mice, these tumors grow rapidly, producing tumors&gt; 1.5 cm³ within 14 days. After resection of the primary tumor, micrometastases in the lungs grow exponentially, until they completely cover the surface of the lung. These metastases are highly vascularized by day 14 after resection of the primary tumor. If the primary tumor is left, the micrometastases remain latent and are not macroscopically visible. Recombinant angiostatin was administered systemically to mice after resection of the primary tumor to test for suppression of metastasis growth. Angiostatin expressed by P. pastoris administered systemically at 30 ug / mouse / day inhibited metastasis growth as quantified by noting surface metastases (Figure 39A) and total lung weight (Figure 39B). The lung weights of mice that had resected primary tumors and that had

ES 2 292 174 T3 received daily doses of recombinant angiostatin or angiostatin obtained from elastase cleavage of plasminogen were comparable to those of normal mice (190 to 200 mg). The lungs of mice in which the primary tumors were resected and subsequently treated with daily doses of recombinant angiostatin were pink in color with a minimal number of non-vascularized micrometastases (Figure 40). In contrast, saline-treated mice after resection of the primary tumor had their lungs coated with vascularized metastases (Figure 41). Also of notable importance was an absence of systemic or local toxicity produced by angiostatin expressed by P. pastoris at the dose and regimen used in this study. There was no evidence of inflammation or bleeding in all treated mice.

The angiostatin protein expressed by P. pastoris possesses two important physical characteristics of the natural protein: (1) it is recognized by a conformationally dependent monoclonal antibody that arises against kringle 1 to 3 of human plasminogen (Figure 36B) and (2 ) binds to lysine (Figure 36A and B). These properties indicated that the recombinant angiostatin protein was expressed with a conformation that mimics that of the native molecule. Angiostatin protein expressed by P. pastoris inhibits bFGF-stimulated bovine capillary endothelial cell proliferation in vitro (Figure 38), when administered systemically, recombinant angiostatin maintained the otherwise lethal metastatic Lewis lung carcinoma in a inhibited state (Figure 39A and B and Figure 40).

Preliminary data demonstrate the absence of a detectable transcript for angiostatin in recently resected Lewis lung tumors from mice or in CLL cells after 4 passages in in vitro culture. Plasminogen, produced by the liver, is maintained in circulation at a stable plasma concentration of 1.6 ± 0.2 juM. LLC-LM tumors may produce an enzyme that clears plasminogen, bound or circulating, to produce angiostatin. Alternatively, inflammatory cells attracted to the tumor site could produce such an enzyme.

It is intriguing that both natural human plasminogen and P. pastoris are produced in a glycosylated and non-glycosylated form. In the case of human plasminogen, a single transcript for a single gene can produce both forms. The molecular mechanism of post-translational differential modifications of human plasminogen, as well as those observed in TPA, are unknown.

Angiostatin is highly expressed by P. pastoris. The supernatants contain 100 mg / l of the protein. Therefore, the amounts required for clinical trials must be simple to produce and purify using conventional technology well known to those of skill in the art. The development of this expression system and the demonstration of the in vitro and in vivo activity of purified recombinant angiostatin against metastasis provided the basis for evaluating the ability of these fragments to inhibit tumor growth and prolong life in patients. cancer patients and others with angiogenesis-mediated disease.

Example 29

Production and Administration of Angiostatin Aggregate

This example demonstrates the production and administration of angiostatin aggregate to inhibit endothelial cell proliferation and tumor growth. Typically, it is assumed that it is necessary to solubilize and refold recombinant proteins produced from E. coli (renatured, reduced, and alkylated) to achieve in vivo activity. In the process, a significant amount of the protein is often lost. In this example, the recombinant angiostatin addition occurs after purification and is used directly, without further renaturation, reduction, or alkylation, to inhibit angiogenesis and tumor growth. Thus, this example provides a surprisingly efficient means of producing and using angiostatin. This addition of angiostatin and the method for its administration also provides a means of sustained release of angiostatin, thus optimizing its efficacy. As used herein "angiostatin aggregate" means angiostatin that has been substantially purified, but not manually refolded, as described in more detail below.

Angiostatin expression

The E. coli expression system, which has the advantage of being fast, productive and inexpensive, was used for the expression of mouse angiostatin. Two oligonucleotide primers, flanking a plasminogen kringles 1-4 cDNA sequence (plasminogen cDNA was purchased from ATCC), were designed for a PCR-based strategy to construct the angiostatin expression system. (See, for example, Menhart, N., Shel, LC, Kelly, RF, and Castellino, FJ (1991) Biochem. 30, 1948-1957); Marti, D., Schaller, J., Ochensberger, B., and Rickli, EE (1994) Eur. J. Biochem. 219, 455-462; Sohndel, S., Hu, C.-K., Marti, D., Affolter, M., Schaller, J., Llinas, M., and Rickli, EE (1996) Biochem. in print; Rejante, MR, Byeon, I.-JL and Llinas, M. (1991) Biochem. 30, 11081-11092). The PCR product was inserted into the Nco I and XhoI sites of the vector pET22 (Novegen), which contains the T7 lac promoter and an oligohistidine sequence. The cDNA was then transformed into E. coli (strain BL21 (DE3)), which contains a chromosomal copy of the T7 RNA polymerase gene under the control of lacuV5. The expression of recombinant angiostatin was induced by the addition of IPTG. Expressed angiostatin accumulated as inclusion bodies in host cells and typically consisted of more than 40% of the total cellular protein, as calculated by Coomassie blue staining.

ES 2 292 174 T3

The invention contemplates that a recombinant angiostatin, or a fragment thereof, derived from any appropriate species can be expressed in a variety of vector and host systems, well known to those of skill in the art.

Purification and aggregation of angiostatin

The insoluble fraction of the sonicated host cells containing the inclusion bodies was purified by centrifugation at 9,000 rpm for 25 minutes, and the washing was repeated. The resulting product contained approximately 80% angiostatin, as calculated by Coomassie blue staining. The N-terminal oligohistidine domain of the fusion protein enabled the convenient and inexpensive purification of recombinant angiostatin with chelation affinity chromatography on a Ni column.²+ -NTA-agarose (1.5 cm x 5 cm) under denaturing conditions, dissolving the inclusion bodies in 6 M urea. As seen in Figure 41, lane 7, purified angiostatin shows a single band in SDS -PAGE by Coomassie blue staining, with a migration rate between about 45 kD to 65 kD, and more preferably about 55 kD. This single step chromatography can purify the protein to substantial homogeneity, however, the invention contemplates that a variety of other purification techniques well known to those of skill in the art can be employed to achieve this product.

The protein, in its elution buffer, was then dialyzed (15,000 MWCO (molecular weight cutoff) against phosphate buffered saline (PBS) for 24 hours with 3 dialysate changes at 4 ° C. During the dialysis a white precipitate was observed The sample was then removed from the dialysis bags and the precipitate was removed by centrifugation. The precipitate was then resuspended using a vortex, in PBS to form a fine suspension for use in animal models, or to be stored at -20 ° C. This precipitate provided the addition of angiostatin. Alternatively, 20 mM tris-HCl, pH 7.9 / 150 mM NaCl can be used as a dialysis buffer, e.g. The invention contemplates that more or less dialysis and a variety of other dialysis buffers can be used to give a corresponding amount of addition of angiostatin, within limitations that can be routinely determined by one of ordinary skill in the art, in view of the current disclosure.

In vitro tests of the addition of angiostatin

Bovine capillary endothelial cells (BCE) were used in an assay as explained in Example 8. The cells were exposed to recombinant human angiostatin aggregate prepared as described in this example. The results are shown in Figure 42, which reveals a significant inhibition of endothelial cells exposed to the aggregate of angiostatin.

In vivo testing of the addition of angiostatin

All work with animals was carried out in the animal facilities of the Children's Hospital according to institutional guidelines.

A. The recombinant mouse angiostatin aggregate, prepared as described in this example, was tested with the chicken CAM (chorioallantoic membrane) assay. (O'Reilly, MS, Holmgren, L., Shinc; and., Chen, C., Rosenthal, RA, Moses, M., Lane, WS, Cao, Y., Sage, EH and Folkman, J., Cell (1994) 79: 315-328). At a dose of 25 ug, the inhibition of capillary formation was 100% (the five chicken CAMs tested gave 2-3 + zones of inhibition). At a dose of 100 ug, the inhibition of capillary formation lasted for 96 hours. The effect of other angiogenesis inhibitors and plasminogen-derived angiostatin is known to last approximately 48 hours. This suggests that the addition of angiostatin provides a sustained release advantage.

B. Lewis lung carcinoma was inoculated subcutaneously in the mid-back of C57B16 mice. Mice were implanted 1 x 10⁶ cells in 0.1 ml of PBS prepared as described above and caged in groups of 6 or less. Tumors were measured with a dial caliper and tumor volumes were determined using the formula width x width x length x 0.52, and the ratio of treated tumor volume to control (T / C) was determined for the last point of time.

Once the tumor volume was 100-200 cubic millimeters, which occurred within 3-5 days, the mice were randomized for two separate experiments. In the first experiment, mice received 2-3 mg / kg of the recombinant mouse angiostatin suspension aggregate in PBS injected subcutaneously at a site remote from the tumor every 24 hours (n = 3 mice / group). The control group received comparable injections of saline. In the second separate experiment (n = 6 mice / group), test mice received 10 mg / kg of the recombinant mouse angiostatin suspension aggregate in PBS injected subcutaneously at a site remote from the tumor every 24 hours. The control group received comparable injections of saline.

No toxicity on weight loss was observed in any of the mice treated with the addition of recombinant angiostatin suspension. In mice, the addition of angiostatin was observed as a subcutaneous mass after injection that was resorbed over a period of several hours. This suggests that it was gradually solubilizing, and shows that the addition of angiostatin is an effective means of sustained release. At both doses, there was a

ES 2 292 174 T3 significant inhibition of the growth of primary Lewis lung carcinoma tumors in treated mice. See Figures 43 and 44. The T / C obtained by administering 10 µg / kg / day as a single injection for 19 days (when saline control animals began to die and were sacrificed) is 0.06 , which constitutes a reduction in tumor volume never before achieved according to the inventors' knowledge.

The in vitro and in vivo results of this example provide a reasonable basis to expect success in using the products and procedures described for the inhibition of endothelial cell proliferation, angiogenesis, and tumor growth in humans. Although not wishing to be bound by theory, the recombinant angiostatin aggregate produced by the present method may have a different conformation, and thus different physical properties, from elastase-generated angiostatin or other purified proteins that are normally renatured. Purified recombinant angiostatin was precipitated to form a suspension aggregate by dialyzing the protein solution against a relatively large volume of buffer or water. This is believed to occur because the improperly folded protein becomes insoluble and aggregates. Previously, a denatured protein aggregate would not have been expected to be so remarkably effective in vivo. The potent antitumor activity demonstrated is also believed to be due to a slow but steady dissolution and release of protein aggregate from subcutaneous sites. The invention further contemplates that either naturally occurring or recombinant angiostatin, as well as angiostatin fragments, may be used in the purification methods described to produce either the natural or recombinant angiostatin aggregate, or the angiostatin fragment aggregate, which can be used for the successful inhibition of endothelial cell proliferation and tumor growth, without the need for renaturation.