WO1999059613A1 - BINDING MOIETIES FOR UROKINASE PLASMINOGEN ACTIVATOR (uPA) - Google Patents

Abstract

Novel binding moieties for urokinase plasminogen activator (uPA) are disclosed. Such binding moieties are useful, for example, in detecting or isolating uPA from solutions or fluids containing it. Particular polypeptide binding moieties are shown to be useful as affinity ligands for purification of uPA from various sources.

Description

Binding Moieties for Urokinase Plasminogen Activator (uPA)

Field of the Invention The present invention relates to urokinase plasminogen activator and particularly its detection, isolation and purification. Specifically, the present invention relates to discovery of and isolation of new polypeptides capable of binding to urokinase plasminogen activator, or uPA, and to the use of such polypeptides in methods of detection and methods of purification.

Background of the Invention

Human urokinase plasminogen activator, or uPA, is a proteolytic enzyme. uPA serves to convert plasminogen, an inactive proenzyme, into plasmin, a thrombolytic enzyme. uPA has this plasminogen activation function in common with tissue plasminogen activator (tPA), although uPA and tPA have distinct protein structures, tissue specific expression, and biological activities. uPA is believed to be primarily involved in cell-mediated tissue degradation and proteolysis by activation of latent matrix-degrading proteinases or growth factors. uPA acts as a physiological trigger for the fibrinolytic system. Inactivation of uPA may result in fibrin deposition, shown to affect growth, fertility, and survival in mice. Since uPA binds to fibrin in thrombi and activates plasminogen there to dissolve clots, uPA, like tPA, is an important drug for use as a thrombolytic agent. Although tPA remains among the most-prescribed thrombolytic agents, the cost of production and purification is greater than other effective thrombolytic agents, such as streptokinase and urokinase. Specifically, the cost of uPA is much less than that of tPA, allowing for greater distribution to patients in need of thrombolytic treatment at reduced economic expense. uPA may have additional therapeutic uses based on its apparent role in a variety of physiological and pathological processes involving cell migration and tissue remodeling. uPA is produced and secreted by multiple vascular cell types, thus influencing the processes and the extent to which the vasculature is remodeled during the development of the intima or a neointima and during hypertrophy and angiogenesis. (Xing et al. 1(1997); Tkachuk et al. (1996).) uPA has been shown to be an important angiogenic protease in vitro and in vivo. Modified uPA has been shown to induce liver regeneration without bleeding. uPA has further been implicated in various physiological processes such as macrophage invasion, wound- healing, sperm migration, ovulation, fertilization, embryo implantation, and embryogenesis.

Current scientific evidence thus indicates the important role of uPA in a variety of physiological processes. uPA is currently an important therapeutic in the treatment of thrombosis, and intensive research is under way for other potential therapeutic uses. In order to maximize the availability of uPA for treatment and further research, ways in which uPA can be produced more efficiently or at lower cost must be explored.

Effective means for eliminating impurities such as cell debris, pathogens, undesired human proteins, etc. from a production feed stream is also important in the production of uPA, as it is with any protein product intended ultimately for therapeutic administration to human patients.

Thus, there is a continuing need for the development of improved reagents, materials and techniques for the isolation of uPA on a more efficient and cost-effective basis.

Affinity chromatography is a very powerful technique for achieving dramatic single- step increases in purity. Narayanan (1994), for instance, reported a 3000-fold increase in purity through a single affinity chromatography step.

Affinity chromatography is not, however, a commonly used technique in large-scale production of biomolecules such as uPA. The ideal affinity chromatography ligand must, at acceptable cost, (1) capture the target biomolecule with high affinity, high capacity, high specificity, and high selectivity; (2) either not capture or allow differential elution of other species (impurities); (3) allow controlled release of the target under conditions that preserve (i.e., do not degrade or denature) the target; (4) permit sanitization and reuse of the chromatography matrix; and (5) permit elimination or inactivation of any pathogens. However, finding high-affinity ligands of acceptable cost that can tolerate the cleaning and sanitization protocols required in pharmaceutical manufacturing has proved difficult (see, Knight, 1990). Murine monoclonal antibodies (MAbs) have been used effectively as affinity ligands. Monoclonal antibodies, on the other hand, are expensive to produce, and they are prone to leaching and degradation under the cleaning and sanitization procedures associated with purification of biomolecules, leading MAb-based affinity matrices to lose activity quickly (see, Narayanan, 1994; Boschetti, 1994). In addition, although MAbs can be highly specific for a target, the specificity is often not sufficient to avoid capture of impurities that are closely related to the target. Moreover, the binding characteristics of MAbs are determined by the immunoglobulin repertoire of the immunized animal, and therefore practitioners must settle for the binding characteristics they are dealt by the animal's immune system, i.e., there is little opportunity to optimize or select for particular binding or elution characteristics using only MAb technology. Finally, the molecular mass per binding site (25 kDa to 75 kDa) of MAbs and even MAb fragments is quite high.

Up until now, there have been no known affinity ligands suitable for the purification of uPA that approach the characteristics of the ideal affinity ligand described above, that not only bind to the target uPA molecule with high affinity but also release the uPA under desirable or selected conditions, that are able to discriminate between the uPA and other components of the solution in which the uPA is presented, and/or that are able to endure cleaning and sanitization procedures to provide regenerable, reusable chromatographic matrices.

We have now surprisingly discovered a series of polypeptides capable of binding to uPA with high affinity. Such polypeptides are non-naturally occurring polypeptides which may be easily synthesized and adopted to multiple uses where binding to uPA is advantageous. Preferred embodiments described herein bind to the target uPA under solution conditions useful in purification and also release uPA under specific solution conditions that are not harmful to commercially available chromatographic matrices. The uPA binding polypeptides according to the present invention can be utilized in any method where detection of the presence, isolation, labeling, or purification or removal of uPA from a solution containing it is desirable.

Summary of the Invention

Accordingly, it is an object of the present invention to provide novel binding polypeptides for urokinase proteins. Preferred binding molecules of the present invention exhibit not only distinct characteristics for binding of the target uPA but also specific and desirable characteristics for release (elution) of the target uPA. Especially preferred binding molecules according to the invention are short polypeptide sequences, characterized by a stable loop structure. Several binding polypeptides exhibiting high affinity for uPA proteins from various sources have been identified and isolated. Such binding polypeptides are useful for identifying, isolating and purifying uPA and uPA-like polypeptides in a solution containing them.

The most preferred binding polypeptides specific for uPA isolated according to the present invention are polypeptides characterized by a loop structure formed as a result of a disulfide bond between two cysteine residues located at the positions disclosed in SEQ ID NO: 1. Specific polypeptide uPA binding moieties according to the present invention include polypeptides comprising an amino acid sequence of the following general formula: X_rCys-X₂-Trp-Trp-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X_t is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.

Preferred polypeptides according to the invention comprise an amino acid sequence: Ala-Glu-Gly-X_rCys-X₂-Trp-Trp-X₃-Cys-Gly-Ser (SEQ ID NO: 2), wherein X is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His; and especially preferred polypeptides will include additional C-terminal sequences selected from Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 3) or Glu-Gly-Gly-Gly-Ser (SEQ ID NO:4), thus providing preferred embodiments having the following amino acid sequences:

Ala-Glu-Gly-X_rCys-X₂-T -Trp-X₃-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 5) and Ala-Glu-Gly-X_rCys-X₂-Tφ-Trp-X₃-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 6), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; X₃ is Asp or His. Particularly preferred uPA binding polypeptides described herein are:

Ala-Glu-Gly-Val-Cys-Ser-Trp-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Tyr-Cys-Asp-Trp-Trp-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 8); Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 9); and Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10). Solutions from which uPA and uPA-like polypeptides may be isolated and purified include, but are not limited to, blood, blood fractions, urine, urine fractions, and recombinant cell culture supematants containing uPA or a uPA-like polypeptide produced and secreted by the recombinant host cell. In a further embodiment, the present invention provides a method for identifying and isolating uPA binding moieties via phage display technology. In addition, uPA binding moieties having specific and predetermined binding and elution characteristics for a particular form of uPA may be selected from a binding moiety library, such as a phage display library, by a method comprising: (a) preselecting a first solution condition (i.e., the binding conditions) at which it is desired that a binding moiety should exhibit an affinity for uPA, forming an affinity complex;

(b) preselecting a second solution condition (i.e., the release conditions) at which it is desired that the binding moiety will dissociate from the uPA, wherein the second solution condition is different in some respect (e.g., temperature, pH, solvent concentration, etc.) from the first solution condition;

(c) providing a library of analogues of a parental binding domain template, wherein each analogue differs from said parental binding domain by variation of the amino acid sequence at one or more amino acid positions within the domain; (d) contacting said library of analogues with a uPA target at the first solution condition, under conditions suitable for formation of a complex between the binding moiety and a uPA target;

(e) removing from the solution the unbound members (analogues) of the library;

(f) subjecting the uPA complexes that remain from step (e) to the second solution condition for dissociation of at least one of the binding moiety/uPA target complexes;

(g) recovering the binding analogue(s) released under the second solution condition, wherein the recovered analogues identify isolated uPA binding moieties.

Optionally, the above procedure can include additional release condition steps, i.e., optionally subjecting the uPA complexes that remain from step (f) to a third solution condition to dissociate other remaining complexes, which may be collected in a separate fraction from the uPA binding molecules released under the second solution conditions. Such a step, if the conditions are stringent enough to dissociate all of the complexes formed in step (d), will identify solution conditions suitable for regeneration of binding matrices utilizing the binding molecules isolated according to this process. Also included in the present invention are non-peptide binding molecules and modified polypeptides that bind uPA and/or uPA-like polypeptides. An example of these modifications is a constrained-loop peptide having paired cysteine residues that form disulfide bonds, modified at the cysteine residues by substitution of one of the cysteines with non-natural amino acids capable of condensing with the other cysteine side-chain to form a stable thioether bridge. Such cyclic thioether analogues of synthetic peptides are described in PCT publication WO 97/46251 , incoφorated herein by reference. Other specifically contemplated modifications include specific amino acid substitutions to lend stability or other properties without significantly affecting uPA binding, e.g., substitution of Glu-Pro for Asp-Pro to reduce acid lability, N- terminal or C-terminal modifications to incoφorate linkers such as poly-glycine segments and alterations to include functional groups, notably hydrazide (-NH-NH₂) functionalities, e.g., to assist in immobilization of binding moieties according to this invention on solid supports.

In a further embodiment, the present invention encompasses a composition of matter comprising isolated nucleic acids, preferably DNA, encoding binding molecules of the present invention.

In another embodiment, the present invention provides a method for detecting a uPA in a solution suspected of containing it, comprising contacting the solution with a binding molecule according to the invention and determining whether a binding complex has formed.

A further embodiment of the present invention is a method for purification of uPA from a solution containing it, comprising the steps:

(a) contacting a solution containing uPA with a binding molecule according to this invention under solution conditions conducive to forming a binding complex comprised of uPA and the binding molecule;

(b) separating the complexes from the non-binding components of the solution; (c) dissociating the uPA from the binding molecule; and

(d) collecting the dissociated, purified uPA.

Also envisioned by the present invention is a method for isolating uPA comprising:

(a) immobilizing a binding molecule according to the invention on a solid support,

(b) contacting a uPA-containing solution solution with the solid support, (c) removing the non-binding components from the solution, and

(d) eluting the uPA from the solid support.

Definitions

As used herein, the term "recombinant" is used to describe non-naturally altered or manipulated nucleic acids, host cells transfected with exogenous nucleic acids, or polypeptides expressed non-naturally, through manipulation of isolated DNA and transformation of host cells.

Recombinant is a term that specifically encompasses DNA molecules which have been constructed in vitro using genetic engineering techniques, and use of the term "recombinant" as an adjective to describe a molecule, construct, vector, cell, polypeptide or polynucleotide specifically excludes naturally occurring such molecules, constructs, vectors, cells, polypeptides or polynucleotides. The term "bacteriophage" is defined as a bacterial virus containing a DNA core and a protective shell built up by the aggregation of a number of different protein molecules. The terms

"bacteriophage" and "phage" are used herein interchangeably. The term "uPA-like polypeptide" is used herein to refer to a modified or truncated form of natural uPA or full-length recombinant uPA, which uPA-like polypeptide retains the thrombolytic properties of uPA and is structurally sufficiently similar to natural or full-length uPA to also be recognized (bound) by polypeptide binding moieties according to this invention. The uPA binding moieties of this invention will be described in the text that follows with reference to uPA, either as a naturally occurring protein or a recombinantly produced form of uPA; and it will be understood by those of skill in the art that the binding moieties described will alternatively also bind to many structurally similar uPA forms, whether trucations, active fragments, cleaved subcomponents, uPA polypeptides modified by substitution, addition or subtraction of amino acids, or any other modification that is conventional in the art. Such obvious alternate uPA forms are encompassed by the term "uPA-like polypeptides"; and the disclosure herein of moieties as binding to uPA or of methods of using such moieties to detect, isolate or purify a uPA protein will also be applicable to binding to uPA-like polypeptides and to methods of detecting, isolating or purifying uPA-like polypeptides. A determination that a binding moiety disclosed herein as recognizing a full-length uPA also recognizes a particular uPA-like polypeptide may be made by simple methods known in the art and illustrated below, and such determinations are within the skill in the art. The term "uPA target" is used herein to refer collectively to uPA and/or uPA-like polypeptides contained in a solution or production feed stream.

The term "binding moiety" as used herein refers to any molecule, polypeptide, peptidomimetic or transformed cell ("transformant") capable of forming a binding complex with another molecule, polypeptide, peptidomimetic or transformant. A "uPA binding moiety" is a binding molecule that forms a complex with uPA. Specific examples of uPA binding moieties are the polypeptides described herein (e.g., SEQ ID NOs: 1, 2 and 5-10) and bacteriophage displaying any of such polypeptides. Also included within the definition of uPA binding moieties are polypeptides derived from or including a polypeptide having an amino acid sequence according to SEQ ID NOs: 1 or 2, above, and such polypeptides which have been modified for particular results. Specific examples of modifications contemplated are C-terminal or N-terminal amino acid substitutions or polypeptide chain elongations for the piupose of linking the binding moiety to a chromatographic support or other substrate, and substitutions of pairs of cysteine residues that normally form disulfide links, for example with non-naturally occurring amino acid residues having reactive side chains, for the puφose of forming a more stable bond between those amino acid positions than the former disulfide bond. All such modified binding molecules are also considered binding moieties according to this invention so long as they retain the ability to bind uPA and/or uPA-like polypeptides. The term "binding" refers to the determination by standard assays that a binding moiety recognizes and forms a binding complex with a given target, in this case uPA. Preferably the association of the two molecules to form a complex is reversible. Such standard assays include equilibrium dialysis, gel filtration, and the monitoring of spectroscopic changes that result from binding.

The term "specificity" refers to a binding moiety having a higher binding affinity for one target over another. The phrase "specificity for uPA" refers to a uPA binding moiety having a higher affinity for uPA than for a structurally non-related protein, for instance bovine serum albumin (BSA). Many of the urokinase binding polypeptides tested herein showed preferential binding to uPA over BSA.

Brief Description of the Drawings

Figure 1 shows the results of an ELISA testing the binding of the TN-6/I phage isolates having urokinase binding affinity from four rounds of screening: TU33, TU34, TU36, TU37, TU38 and TU42 against immobilized urokinase at pH 7, immobilized urokinase at pH 2, immobilized tPA at pH 7, immobilized tPA at pH 2, immobilized BSA at pH 7, and immobilized BSA at pH 2. The ELISA identifies affinity ligands useful for separation of urokinase from a feed stream such as urine.

Figure 2 shows the results of an ELISA testing the binding of some of the TN-10/VIIIa phage isolates having urokinase binding affinity from four rounds of screening: TU51, TU53, TU56, TU58, TU60 and TU62 against immobilized urokinase at pH 7, immobilized urokinase at pH 2, immobilized tPA at pH 7, immobilized tPA at pH 2, immobilized BSA at pH 7, and immobilized BSA at pH 2. The ELISA identifies affinity ligands useful for separation of urokinase from a feed stream such as urine. Figure 3 A shows the results of an ELISA testing the binding of some of the CMTI phage isolates having urokinase binding affinity from four rounds of screening: CU22, CU29, and CU32 against immobilized urokinase at pH 7, immobilized urokinase at pH 2, immobilized tPA at pH 7, immobilized tPA at pH 2, immobilized BSA at pH 7, and immobilized BSA at pH 2. Figure 3B shows the results of an ELISA testing the binding of some of the CMTI phage isolates having urokinase binding affinity from four rounds of screening: CU25, CU27, CU28, CU 31 and CU32 against immobilized urokinase at pH 7, immobilized urokinase at pH 2, immobilized tPA at pH 7, immobilized tPA at pH 2, immobilized BSA at pH 7, and immobilized BSA at pH 2. None of these isolates were suitable as an affinity ligand according to the criteria of the test. Figure 4A shows the results of an ELISA testing the binding of some of the LACI/F phage isolates having urokinase binding affinity from four rounds of screening: LU2, LU5, LU9 and LU12 against immobilized urokinase at pH 7 and immobilized BSA at pH 7. Figure 4B shows the results of an ELISA testing the binding of some of the LACI/F phage isolates having urokinase binding affinity from four rounds of screening: LU2, LU4, LU10 and LU12 against immobilized urokinase at pH 7 and immobilized BSA at pH 7. None of these isolates were suitable as an affinity ligand according to the criteria of the test.

Figure 5 shows a chromatogram of urokinase plasminogen activator (cone, about 250 μg/ml in PBS) applied to a 350 μl affinity chromatography column having an immobilized uPA affinity ligand (TN-6/I isolate TU33) immobilized on an aldehyde-substituted methacrylate resin. Elution was at pH 2 with 20% acetonitrile (ACN). The peak at 18 minutes is estimated to contain approximately 60.7% of the uPA loaded.

Figure 6 shows a chromatogram of uPA (cone, about 250 μg /ml in cell culture supernatant from a human osteogenic sarcoma cell line (with 10% fetal bovine serum) applied to a 350 μl affinity chromatography column having an immobilized uPA affinity ligand (TN-6/I isolate TU33) immobilized on an aldehyde-substituted methacrylate resin. Elution was elution at pH 2 with 20% ACN. The peak at 18 minutes is estimated to contain approximately 37.5% of the loaded uPA.

Figure 7 shows a chromatogram of urokinase plasminogen activator (cone, about 250 μg/ml in PBS) applied to a 350 μl affinity chromatography column having an immobilized uPA affinity ligand (TN-6/I isolate TU42) immobilized on an aldehyde-substituted methacrylate resin. Elution was at pH 2 with 20% acetonitrile (ACN). The peak at 17 minutes is estimated to contain approximately 47.4% of the loaded uPA.

Figure 8 shows a chromatogram of uPA (cone, about 250 μg /ml in cell culture supernatant from a human osteogenic sarcoma cell line (with 10% fetal bovine serum) applied to a 350 μl affinity chromatography column having an immobilized uPA affinity ligand (TN-6/I isolate TU42) immobilized on an aldehyde-substituted methacrylate resin. Elution was elution at pH 2 with 20% ACN. The peak at 18 minutes is estimated to contain approximately 30.1% of the loaded uPA.

Detailed Description of the Preferred Embodiments

The present invention provides novel binding moieties for uPA. Such binding moieties make possible the efficient detection, isolation and purification of uPA or uPA-like peptides in solutions or systems that contain them. In particular, the binding moieties of this invention, when immobilized on a solid matrix, may be most advantageously used as affinity ligands for the separation of uPA target proteins from a solution (e.g., urine) or a feed stream (e.g., cell culture of a uPA-secreting host cell). The preferred binding moieties of the present invention bind uPA and or uPA-derived polypeptides with high affinity, i.e., acting at low, physiologically relevant concentrations, comparable to known anti-uPA antibodies and other uPA-binding proteins.

Specific uPA binding polypeptides according to the present invention were isolated initially by screening of phage display libraries, that is, populations of recombinant bacteriophage transformed to express an exogenous peptide loop on their surface. In order to isolate new polypeptide binding moieties for a particular target, such as uPA, screening of large peptide libraries, for example using phage display techniques, is especially advantageous, in that very large numbers (e.g., 5 x 10⁹) of potential binders can be tested and successful binders isolated in a short period of time.

In order to prepare a phage library of potential polypeptides to screen for binding moieties such as uPA binding peptides, a candidate binding domain is selected to serve as a structural template for the peptides to be displayed in the library. The phage library is made up of analogues of the parental domain or template. The binding domain template may be a naturally occurring or synthetic protein, or a region or domain of a protein. The binding domain template may be selected based on knowledge of a known interaction between the binding domain template and uPA, but this is not critical. In fact, it is not essential that the domain selected to act as a template for the library have any affinity for the target at all: Its puφose is to provide a structure from which a multiplicity (library) of similarly structured polypeptides (analogues) can be generated, which multiplicity of analogues will hopefully include one or more analogues that exhibit the desired binding properties (and any other properties screened for).

In selecting the parental binding domain or template on which to base the variegated amino acid sequences of the library, the most important consideration is how the variegated peptide domains will be presented to the target, i.e., in what conformation the peptide analogues will come into contact with the target. In phage display methodologies, for example, the analogues will be generated by insertion of synthetic DNA encoding the analogues into phage, resulting in display of the analogue on the surfaces of the phage. Such libraries of phage, such as M13 phage, displaying a wide variety of different polypeptides, can be prepared using techniques as described, e.g., in Kay et al., Phage Display of Peptides and Proteins: A Laboratory Manual (Academic Press, Inc., San Diego 1996) and U.S. 5,223,409 (Ladner et al.), mcoφorated herein by reference.

For formation of phage display libraries, it is preferred to use a structured polypeptide as the binding domain template, as opposed to an unstructured, linear peptide. Mutation of surface residues in a protein will usually have little effect on the overall structure or general properties (such as size, stability, and temperature of denaturation) of the protein, while at the same time mutation of surface residues may profoundly affect the binding properties of the protein The more tightly a polypeptide segment is constrained, the less likely it is to bind to any particular target; however if the polypeptide does bind, the binding is likely to be of higher affinity and of greater specificity. Thus, it is preferred to select a parental domain and, in turn, a structure for the potential polypeptide binders, that is constrained withm a framework having some degree of rigidity In isolating the specific polypeptides according to this invention, four different libraries, each based on a different binding domain template, were screened. The libraries were designated CMTI-I (having 9.13 * 10⁶ analogues of a Cucurbida maxima trypsm inhibitor domain), TN-6/I (having 8.55 x 10⁶ analogues of a 13-amino acid synthetic microprotem), TN- 10/VIIIa (having 2.3 x 10⁷ analogues of an 18-amιno acid synthetic microprotem), and LACI/F (having 3 12 x 10⁴ analogues of the first Kunitz domain of hpoprotem associated coagulation inhibitor) Each of these libraries was constructed for expression of diversified polypeptides on Ml 3 phage. Small polypeptide domains or microprotem binding loops offer several advantages as useful binding moieties over larger proteins: First, the mass per binding site is reduced, e.g , such highly stable and low molecular weight polypeptide domains can show much higher binding per gram than do, e.g., antibodies (150 kDa) or single-chain antibodies (30 kDa). Second, the possibility of non-specific binding is reduced because there is less surface available for binding. Third, small proteins or polypeptides can be engineered to have unique tetheπng sites such as terminal polylysme segments in a way that is impracticable for larger proteins or antibodies Fourth, a constrained polypeptide structure is more likely to retain its functionality when transferred with the structural domain intact from one framework to another, that is, the binding domain structure is likely to be transferable from the framework used for presentation in a library (e.g., displayed on a phage) to an isolated protein removed from the presentation framework or immobilized on a chromatographic substrate.

The TN-6/I library, for example, was created by making a designed seπes of mutations or variations within a coding sequence for the microprotem template, each mutant sequence encoding a binding loop analogue corresponding in overall structure to the template except having one or more amino acid variations in the sequence of the template. The novel variegated (mutated) DNA provides sequence diversity, and each transformant phage displays one variant of the initial template amino acid sequence encoded by the DNA, leading to a phage population (library) displaying a vast number of different but structurally related amino acid sequences. The amino acid variations are expected to alter the binding properties of the binding loop or domain without significantly altering its structure, at least for most substitutions. It is preferred that the amino acid positions that are selected for variation (variable amino acid positions) will be surface amino acid positions, that is, positions in the amino acid sequence of the domains which, when the domain is in its most stable conformation, appear on the outer surface of the domain (i.e., the surface exposed to solution). Most preferably the amino acid positions to be varied will be adjacent or close together, so as to maximize the effect of substitutions.

As indicated previously, the techniques discussed in Kay et al., Phage Display of Peptides and Proteins: A Laboratory Manual (Academic Press, Inc., San Diego 1996) and U.S. 5,223,409 are particularly useful in preparing a library of potential binders corresponding to the selected parental template. The TN-6/I library and the other phage display libraries mentioned above were prepared according to such techniques, and they were screened for uPA binding polypeptides against an immobilized uPA target.

In a typical screen, a phage library is contacted with and allowed to bind the target, in this case uPA. To facilitate separation of binders and non-binders, it is convenient to immobilize the target on a solid support. Phage bearing a target-binding moiety form a complex with the target on the solid support whereas non-binding phage remain in solution and may be washed away with excess buffer. Bound phage are then liberated from the target by changing the buffer to appropriate elution conditions, which might include changing to an extreme pH (pH 2 or pH 10), changing the ionic strength of the buffer, adding denaturants, or other known means. The recovered phage may then be amplified through infection of bacterial cells and the screening process repeated with the new pool that is now depleted in non-binders and enriched in binders. The recovery of even a few binding phage is sufficient to carry the process to completion. After a few rounds of selection, the gene sequences encoding the binding moieties derived from selected phage clones in the binding pool are determined by conventional methods, described below, revealing the peptide sequence that imparts binding affinity of the phage to the target. When the selection process works, the sequence diversity of the population falls with each round of selection until only good binders remain. The sequences converge on a small number of related binders, typically 10-50 out of the more than 10 million original candidates. An increase in the number of phage recovered at each round of selection, and of course, the recovery of closely related sequences are good indications that convergence of the library has occurred in a screen. After a set of binding polypeptides is identified, the sequence information may be used to design other secondary phage libraries, biased for members having additional desired properties. After analysis of the sequences isolated from the library screening, a family of particular uPA binders was defined. In addition, important consensus motifs were observed. The following sequences conforming to the TN-6/I template were found to bind a uPA target: Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Ile-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO:20); Ala-Glu-Gly-Tyr-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 21); and

Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 22). This series of uPA binders defines a family of polypeptides including the amino acid sequence: X_rCys-X₂-Tφ-Tφ-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His; and wherein the polypeptide has the ability to bind to uPA.

The cysteine residues of the microprotein are believed to form a disulfide bond, which causes the microprotein to form a stable loop structure under non-reducing conditions. Thus, the invention relates to the discovery of a uPA binding loop comprising a polypeptide of the formula of SEQ ID NO: 1, wherein Xj is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His. Following the procedures outlined above, additional binding molecules for uPA may be isolated from the phage display libraries described herein or other phage display libraries or collections of potential binding molecules (e.g., combinatorial libraries of organic compounds, random peptide libraries, etc.). Once isolated, the sequence of any individual binding peptide or the structure of any binding molecule can be analyzed, and the binder may be produced in any desired quantity using known methods. For example, the polypeptide binding moieties described herein, since their sequences are now known, may advantageously be produced by chemical synthesis followed by treatment under oxidizing conditions appropriate to obtain the native conformation, i.e., the correct disulfide bond linkages. Synthesis may be carried out by methodologies well known to those skilled in the art (see, Kelley et al. in Genetic Engineering Principles and Methods, (Setlow, J.K., ed.), Plenum Press, NY., (1990) vol. 12, pp. 1-19; Stewart et al., Solid-Phase Peptide Synthesis (1989), W. H. Freeman Co., San Francisco). The binding molecules of the present invention can be made either by chemical synthesis or by semisynthesis. The chemical synthesis or semisynthesis methods allow the possibility of non- natural amino acid residues to be incoφorated.

Polypeptide binding molecules of the present invention are preferably prepared using solid phase peptide synthesis (Merrifield, J. Am. Chem. Soc, 85: 2149 (1963); Houghten, Proc. Natl. Acad. Sci. USA, 82: 5132 (1985)). Solid phase synthesis begins at the carboxy-terminus of the putative polypeptide by coupling a protected amino acid to a suitable resin, which reacts with the carboxy group of the C-terminal amino acid to form a bond that is readily cleaved later, such as a halomethyl resin, e.g., chloromethyl resin and bromomethyl resin, hydroxymethyl resin, aminomethyl resin, benzhydrylamine resin, or t-alkyloxycarbonyl-hydrazide resin. After removal of the α-amino protecting group with, for example, trifluoroacetic acid (TFA) in methylene chloride and neutralizing in, for example, TEA, the next cycle in the synthesis is ready to proceed. The remaining α-amino and, if necessary, side-chain-protected amino acids are then coupled sequentially in the desired order by condensation to obtain an intermediate compound connected to the resin. Alternatively, some amino acids may be coupled to one another forming an oligopeptide prior to addition of the oligopeptide to the growing solid phase polypeptide chain.

The condensation between two amino acids, or an amino acid and a peptide, or a peptide and a peptide can be carried out according to the usual condensation methods such as azide method, mixed acid anhydride method, DCC (dicyclohexylcarbodiimide) method, active ester method (p-nitrophenyl ester method, BOP [benzotriazole-1-yl-oxy-tris (dimethylamino) phosphonium hexafluorophosphate] method, N-hydroxysuccinic acid imido ester method), and Woodward reagent K method.

Common to chemical synthesis of peptides is the protection of the reactive side-chain groups of the various amino acid moieties with suitable protecting groups at that site until the group is ultimately removed after the chain has been completely assembled. Also common is the protection of the α-amino group on an amino acid or a fragment while that entity reacts at the carboxyl group followed by the selective removal of the α-amino-protecting group to allow subsequent reaction to take place at that location. Accordingly, it is common that, as a step in the synthesis, an intermediate compound is produced which includes each of the amino acid residues located in the desired sequence in the polypeptide chain with various of these residues having side-chain protecting groups. These protecting groups are then commonly removed substantially at the same time so as to produce the desired resultant product following purification. The typical protective groups for protecting the α- and ε-amino side chain groups are exemplified by benzyloxycarbonyl (Z), isonicotinyloxycarbonyl (iNOC), O- chlorobenzyloxycarbonyl [Z(NO₂)], p-methoxybenzyloxycarbonyl [Z(OMe)], t-butoxycarbonyl (Boc), t-amyloxycarbonyl (Aoc), isobornyloxycarbonyl, adamatyloxycarbonyl, 2-(4-biphenyl)- 2-propyloxycarbonyl (Bpoc), 9-fluorenylmethoxycarbonyl (Fmoc), methylsulfonylethoxycarbonyl (Msc), trifluoroacetyl, phthalyl, formyl, 2-nitrophenylsulphenyl (NPS), diphenylphosphinothioyl (Ppt), dimethylophosphinothioyl (Mpt), and the like.

As protective groups for the carboxy group there can be exemplified, for example, benzyl ester (OBzl), cyclohexyl ester (Chx), 4-nitrobenzyl ester (ONb), t-butyl ester (Obut), 4- pyridylmethyl ester (OPic), and the like. It is desirable that specific amino acids such as arginine, cysteine, and serine possessing a functional group other than amino and carboxyl groups are protected by a suitable protective group as occasion demands. For example, the guanidino group in arginine may be protected with nitro, p-toluenesulfonyl, benzyloxycarbonyl, adamantyloxycarbonyl, p-methoxybenzenesulfonyl, 4-methoxy-2,6-dimethylbenzenesulfonyl (Mds), 1,3,5-trimethylphenysulfonyl (Mts), and the like. The thiol group in cysteine may be protected with p-methoxybenzyl, triphenylmethyl, acetylaminomethyl ethylcarbamoyl, 4- methylbenzyl, 2,4,6-trimethy-benzyl (Tmb), etc., and the hydroxyl group in the serine can be protected with benzyl, t-butyl, acetyl, tetrahydropyranyl, etc.

After the desired amino acid sequence has been completed, the intermediate polypeptide is removed from the resin support by treatment with a reagent, such as liquid HF and one or more thio-containing scavengers, which not only cleaves the polypeptide from the resin, but also cleaves all the remaining side-chain protecting groups. Following HF cleavage, the protein sequence is washed with ether, transferred to a large volume of dilute acetic acid, and stirred at pH adjusted to about 8.0 with ammonium hydroxide. Upon pH adjustment, the polypeptide takes its desired conformational arrangement.

Polypeptides according to the invention may also be prepared commercially by companies providing peptide synthesis as a service (e.g., BACHEM Bioscience, Inc., King of Prussia, PA; Quality Controlled Biochemicals, Inc., Hopkinton, MA).

The new class of uPA binding polypeptides is designed to be conformationally restrained by disulfide linkages between the two cysteine residues in their sequence. This conformational restraint ensures that the peptides have a stable binding structure that contributes to the peptides' affinity for uPA and their specificity for uPA over non uPA proteins. Other methods for constraining peptides which would retain a similar conformation and uPA specificity for the peptide have been described in the art and are contemplated herein, including the substitution of one or more of the cysteine residues with non-naturally occurring amino acids or peptidomimetics for the puφose of forming a more stable or conformationally preferred linkage between the two positions on the peptide. All such modified uPA binding moieties are also considered uPA binding moieties according to this invention so long as they retain the ability to bind uPA or uPA-like polypeptides. Non-cyclized, or linear, versions of the peptides may also retain moderate binding ability and specificity for uPA and could also be employed in the present invention.

Homologues of the uPA binding polypeptides described herein, as well as homologues to any subsequently discovered uPA binding polypeptides, may be formed by substitution, addition or deletion of one or more amino acids employing methods well known in the art and for particular puφoses known in the art, such as addition of a polyhistidine "tail" in order to assist in purification or substitution of one up to several amino acids, e.g., in order to obliterate an enzyme cleavage site. Other specifically contemplated homologues include polypeptides having N-terminal or C-terminal modifications or linkers, such as polyglycine or polylysine segments, and alterations to include functional groups, notably hydrazide (-NH-NH₂) functionalities, to assist in immobilization of binding peptides according to this invention on solid supports.

Such homologous polypeptides will be understood to fall within the scope of the present invention so long as the substitution, addition or deletion of amino acids does not eliminate its ability to bind uPA. The term "homologous", as used herein, refers to the degree of sequence similarity between two polymers (i.e., polypeptide molecules or nucleic acid molecules). When the same nucleotide or amino acid residue occupies a sequence position in the two polymers under comparison, then the polymers are homologous at that position. The percent homology between two polymers is the mathematical relationship of the number of homologous positions shared by the two polymers divided by the total number of positions compared, the product multiplied by 100. For example, if the amino acid residues at 60 of 100 amino acid positions in two polypeptide sequences match or are homologous, then the two sequences are 60% homologous. The homology percentage figures referred to herein reflect the maximal homology possible between the two polymers, i.e., the percent homology when the two polymers are so aligned as to have the greatest number of matched (homologous) positions. Polypeptide homologues within the scope of the present invention will be at least 88% and preferably at least 91% homologous to at least one of the uPA binding sequences disclosed herein. uPA binding polypeptides according to the present invention also may be produced using recombinant DNA techniques, utilizing nucleic acids (polynucleotides) encoding the polypeptides according to this invention and then expressing them recombinantly, i.e., by manipulating host cells by introduction of exogenous nucleic acid molecules in known ways to cause such host cells to produce the desired uPA binding polypeptides. Recombinant production of short peptides such as those described herein may not be practical in comparison to direct synthesis, however recombinant means of production may be very advantageous where a uPA binding motif of this invention is desired to be incoφorated in a hybrid polypeptide or fusion protein.

The polynucleotides of the present invention may be in the form of RNA or in the form of DNA, which DNA includes cDNA and synthetic DNA. The DNA may be double-stranded or single-stranded, and if single stranded may be the coding strand or non-coding (anti-sense) strand. The coding sequences for uPA binding polypeptides according to the present invention may be manipulated or varied in known ways to yield alternative coding sequences that, as a result of the redundancy or degeneracy of the genetic code, encode the same polypeptide.

The polypeptides and polynucleotides of the present invention are preferably provided in an isolated form, and preferably are purified to homogeneity.

Where recombinant production of uPA binding polypeptides is desired, the present invention also contemplates vectors that include polynucleotides of the present invention, host cells that are genetically engineered with vectors of the invention, and recombinant polypeptides produced by culturing such genetically engineered host cells. Host cells are genetically engineered (transduced or transformed or transfected) with the vectors of this invention which may be, for example, a cloning vector or an expression vector. The vector may be, for example, in the form of a plasmid, a viral particle, a phage, etc. The engineered host cells can be cultured in conventional nutrient media modified as appropriate for activating promoters, selecting transformants or amplifying the uPA binder-encoding polynucleotides. The culture conditions, such as temperature, pH and the like, are those suitable for use with the host cell selected for expression and will be apparent to the skilled practitioner in this field. The polynucleotide may be included in any one of a variety of expression vectors for expressing a polypeptide. Such vectors include chromosomal, nonchromosomal and synthetic DNA sequences, e.g., derivatives of SV40; bacterial plasmids; phage DNA; baculovirus; yeast plasmids; vectors derived from combinations of plasmids and phage DNA, viral DNA such as vaccinia, adenovirus, fowl pox virus, and pseudorabies. However, any other vector may be used as long as it is replicable and viable in the host. The appropriate DNA sequence may be inserted into the vector by a variety of procedures. In general, the DNA sequence is inserted into an appropriate restriction endonuclease site(s) by procedures known in the art. Such procedures and others are within the capability of those skilled in the art.

The DNA sequence in the expression vector is operatively linked to an appropriate expression control sequence(s) (promoter) to direct mRNA synthesis. As representative examples of such promoters, there may be mentioned LTR or SV40 promoter, the E. coli lac or tφ, the phage lambda P_L promoter and other promoters known to control expression of genes in prokaryotic or eukaryotic cells or their viruses. The expression vector also contains a ribosome binding site for translation initiation and a transcription terminator. The vector may also include appropriate sequences for amplifying expression. In addition, expression vectors preferably will contain one or more selectable marker genes to provide a phenotypic trait for selection of transformed host cells, such as dihydrofolate reductase or neomycin resistance for eukaryotic cell culture, or such as tetracycline or ampicillin resistance for bacterial cell cultures such as E. coli.

The vector containing the appropriate DNA sequence as hereinabove described, as well as an appropriate promoter or control sequence, may be employed to transform an appropriate host to permit the host to express the protein. As representative examples of appropriate host cells, there may be mentioned bacterial cells, such as E. coli, Streptomyces, Salmonella typhimurium; fungal cells, such as yeast; insect cells such as Drosophila and Sf9; animal cells such as CHO, COS or Bowes melanoma; plant cells, etc. The selection of an appropriate host for this type of uPA binder production is also within the capability of those skilled in the art from the teachings herein. Many suitable vectors and promoters useful in expression of proteins according to this invention are known to those of skill in the art, and many are commercially available. The following vectors are provided by way of example. Bacterial: pQE70, pQE60, pQE-9 (Qiagen), pbs, pDIO, phagescript, psiX174, pbluescript SK, pbsks, pNH8A, pNH16a, pNHl 8A, pNH46A (Stratagene); ptrc99a, pKK223-3, pKK233-3, pDR540, pRIT5 (Pharmacia). Eukaryotic: pWLNEO, pSV2CAT, pOG44, pXTl, pSG (Stratagene) pSVK3, pBPV, pMSG, pSVL (Pharmacia). Any other plasmid or vector may be used as long as it is replicable and viable in the selected host cell.

Introduction of the vectors into the host cell can be effected by any known method, including calcium phosphate transfection, DEAE-Dextran mediated transfection, or electroporation (see Davis et al., Basic Methods in Molecular Biology. (1986)).

In the practice of the present invention, a determination of the affinity of the uPA binding moiety for uPA is a useful measure, and is referred to as specificity for uPA. Standard assays for quantitating binding and determining affinity include equilibrium dialysis, equilibrium binding, gel filtration, or the monitoring of numerous spectroscopic changes (such as fluorescence) that may result from the interaction of the binding moiety and its target. These techniques measure the concentration of bound and free ligand as a function of ligand (or protein) concentration. The concentration of bound polypeptide ([Bound]) is related to the concentration of free polypeptide ([Free]) and the concentration of binding sites for the polypeptide, i.e., on uPA, (N), as described in the following equation:

[Bound] = N x [Free]/((l/K_a)+[Free]). A solution of this equation yields the association constant, K_a, a quantitative measure of the binding affinity. The association constant, K_a is the reciprocal of the dissociation constant, K_d. The K_d is more frequently reported in measurements of affinity. In general, preferred binding moieties of the present invention will have a dissociation constant of less than about 5 x 10^"6 M, preferably less than about 10^"8 M and most preferably less than 10^"9 M. In comparative terms, a binding polypeptide having a K_d that is at least 2 times higher for a non-uPA protein such as BSA than for uPA would be considered as a weak uPA binder. A peptide having a K _d 10 times greater for BSA than uPA would be a moderate uPA binder, and a peptide having a K _d 100 times or more greater for BSA than for uPA would be termed highly specific for uPA. Preferably the binding moieties of the present invention have a K_d at least 2 times higher for BSA than for uPA, more preferably at least 10 times higher, and most preferably at least 100 times higher. The foregoing assay of uPA affinity can be adapted to a microtiter plate format for evaluating large numbers of polypeptides. Single point concentrations can be used to quickly differentiate molecules of high uPA specificity or binding affinity from those with low uPA specificity or binding affinity.

Uses for uPA Binding Polypeptides

The present invention makes possible the efficient detection of uPA targets in a solution or purification of uPA by affinity chromatography.

The uPA may be produced in any known way, including chemical synthesis; production in transformed host cells; secretion into culture medium by naturally occurring cells or recombinantly transformed bacteria, yeasts, fungi, insect cells, plant cells and mammalian cells; secretion from genetically engineered organisms (e.g., transgenic mammals); or in biological fluids or tissues such as urine, blood, milk, etc. The solution that contains the crude uPA as it is initially produced (i.e., the production solution) will sometimes be referred to as the "feed stream". Each method of producing uPA yields uPA in a feed stream that additionally contains a number of impurities (with respect to uPA). One puφose of the present invention is to produce affinity ligands and preparations (such as chromatography media) comprising such ligands that allow rapid and highly specific purification of uPA from any feed stream. When utilized in purification processes, the binding moieties described herein are most advantageously used in affinity chromatography processes. Any conventional method of chromatography may be employed. Preferably, a polypeptide uPA binder of the invention will be immobilized on a solid support suitable, e.g., for packing a chromatography column. The immobilized polypeptide uPA affinity ligand can then be loaded or contacted with a feed stream under conditions favorable to formation of ligand/uPA complexes, non-binding materials can be washed away, then the uPA can be eluted under conditions favoring release of the uPA molecule from a ligand/uPA complex. Alternatively, bulk chromatography can be carried out by adding a feed stream and an appropriately tagged affinity ligand together in a reaction vessel, then isolating complexes of the uPA and ligand by making use of the tag (e.g., a polyHis affinity tag, which can by used to bind the ligand after complexes have formed), and finally releasing the uPA from the complex after unbound materials have been eliminated.

For detection of uPA and/or uPA-like polypeptides in a solution such as urine or conditioned media suspected of containing it, a binding molecule according to the invention can be detectably labeled, e.g., radiolabeled or enzymatically labeled, then contacted with the solution, and thereafter formation of a complex between the binding molecule and the uPA target can be detected. A phage binding molecule according to the invention, i.e., a recombinant phage displaying a uPA binder polypeptide on its surface, may form a complex with a uPA target protein that is detectable as a sediment in a reaction tube, which can be detected visually after settling or centrifugation. Alternatively, a sandwich-type assay may be used, wherein a uPA binding moiety is immobilized on a solid support such as a plastic tube or well, or a chromatographic matrix such as sepharose beads, then the solution suspected of containing the uPA target is contacted with the immobilized binding moiety, non-binding materials are washed away, and complexed uPA is detected using a suitable detection reagent, such as a monoclonal antibody recognizing the uPA target, which reagent is detectable by some conventional means known in the art, including being detectably labeled, e.g., radiolabeled or labeled enzymatically, as with horseradish peroxidase, and the like.

Isolation of uPA binding moieties in accordance with this invention will be further illustrated below. The specific parameters included in the following examples are intended to illustrate the practice of the invention, and they are not presented to in any way limit the scope of the invention.

Example I: The Isolation of uPA Binding Polypeptides The techniques described above were employed to isolate affinity ligands for natural urokinase plasminogen activator (uPA). The process of creating uPA affinity ligands involved three general steps: (1) screening of approximately 40 million variants of stable parental protein domains for binding to uPA, (2) producing small quantities of the most promising binders, and (3) chromatographic testing of one binding moiety bound to activated beads for the affinity purification of uPA from a spiked cell culture sample.

High molecular weight urokinase derived from human urine was immobilized on Reacti-Gel™ agarose beads (Pierce Chemical Co.) at 4° C essentially as described in Markland et al. (1996). Briefly, the agarose beads (500 μl) were suspended in acetone, then transferred to a spin column to drain. The beads were washed with two 1 ml volumes of ice cold water, then washed twice with ice cold 0.1 M boric acid (pH 8.5). The bottom of the spin column was capped and a solution of 100 μg of uPA (2-chain natural uPA isolated from human urine) dissolved in 400 μl of 0.1 M boric acid (pH 8.5) was added. The column was capped and the bead slurry was incubated on an end-over-end rotator for 30 hours at 4° C. After incubation the column was unsealed and the solution was drained by gravity flow. The column contents were washed with two 1 ml volumes of cold 1 M NaCl followed by two 1 ml volumes of cold 1 M

Tris HCl (pH 8.5). The column was capped and the beads were resuspended in 1 ml of cold 1 M Tris HCl (pH 8.5), then the column was resealed and the slurry mixed using an end-over-end rotator for 4 hours at 4° C. After washing with two 1 ml volumes of cold water, the beads were drained, suspended in an equal volume (~400μl) of cold glycerol, transferred to a VΛ ml microfuge tube and stored at -20° C until use.

The screening was carried out using four phage display libraries: CMTI, TN-6/I, TN- 10/Vπia, and LACI/F (Tables 1-4, respectively).

The construction of the CMTI library is given in Table 1. Table 1 shows the planned amino acid variability of the sequences as a function of permitted codon variability. The CMTI library was constructed by introducing combinatorial sequence diversity into codons specifying a surface-exposed loop formed between cysteines 3 and 10 of the parental CMTI protein. The cysteines were not varied because they form an important part of the structure. Table 1 shows the DNA sequence of the CMTI library. Residues F.₅ and Y_₄ correspond to residues 14 and 15 in the signal sequence of M13mpl8 from which the recipient phage was engineered. Cleavage by Signal Peptidase I (SP-I) is assumed to occur between A, and R,. Residues designated 100- 113 make up a linker between the CMTI variants and mature III, which begins with residue A₂₀₁. The amino acid sequence Y₁₀₄IEGRIV should allow specific cleavage of the linker with bovine Factor X_a between R₁₀₈ and I₁₀₉. The M13-related phage in which this library was constructed carries an ampicillin-resistance gene (Ap^κ) so that cells infected by library phage become ampicillin-resistant. At each variable amino acid position, the wild-type amino acid residue is shown underscored. The amino acid sequence shown in Table 1 is designated SEQ ID NO: 11; the nucleotide sequence shown in Table 1 is designated SEQ ID NO: 12.

The peptide construction of TN-6/I, TN-10/VIIIa, and LACI/F is given below (Tables 2- 4, respectively). In these tables, the encoded amino acids of the variegated, phage-displayed polypeptide domain are shown. DNA encoding the polypeptides was inserted into Ml 3 gene III in a similar manner as described above with respect to the CMTI library.

-23-

-24-

BSTΓΓUTE S RULE 26

This library design gives 8.55 x 10 protein sequences and 17 x 10⁶ DNA sequences.

-25-

-26-

BSTΓΓUTE SHEET RULE 26

-27-

SUBSTΓΓUTE SHEET RULE 26

Prior to screening, the immobilized urokinase agarose beads were tested with clonally pure phage preps displaying each of the parental binding domain polypeptides, to ensure that under the screening conditions there was a low background level of phage recovered. For each of the parental polypeptides, the fraction of input phage recovered was less than or equal to 1 x

10 ^s, which was an acceptably low background level of binding.

Four rounds of screening were performed. Each round consisted of a binding step, a wash procedure, and one or more elution steps. The binding conditions for all rounds were: incubation at 4° C for 20 hours in PBS, 0.1% BSA, 0.01% Tween 80. The wash and elution conditions for each round are summarized in Table 5 below.

After each round, the phage eluted were counted and then amplified by transduction. pH 5 and pH 2 eluates were amplified separately after Round 2 and kept separate during successive screens, so that the pH 2 eluates selected candidates that still had a comparatively high affinity at pH 5. Table 6 below shows the convergence of the screen over the four rounds for each of the libraries.

A convergent screen is one in which the fraction of input increases over successive rounds, indicating that the diversity of the phage library is being reduced. This is a desired result, because it indicates that a ligand candidate for the immobilized target molecule is potentially being selected from the population. Table 6 shows some convergence between

Rounds 2 and 4 for all of the libraries, with the most pronounced results for the TN-6/I (pH 2) and the CMTI (pH 5) elutions.

-28-

E 26

From each of four convergent screens, approximately 12 phage isolates were selected for sequencing. In all of the sequenced isolates, some homology amongst the selectants was seen. The greatest homology was seen in the TN-6/I and CMTI sequences, which is consistent with the finding that these libraries showed the greatest enrichment during screening. For

-29- example, nine of the TN-6/I isolates were found to have the same DNA and amino acid sequence

Candidates were selected for characteπzation of the relative binding affinity, specificity and pH-release characteπstics of the phage-bound proteins for the target uPA, using pH 2 as the release test. The test involved immobilization of urokinase and BSA on Immulon 2 microtiter plates, with detection of relative binding of the phage using a biotinylated sheep antι-M13 antibody ELISA kit from 5 Pπme - 3 Pπme, Inc. (Boulder, Colorado US)

The phage isolates tested had the following designations:

• from TN-6/I, pH 2 release: TU33, TU34, TU36, TU37, TU39, TU42 • from TN-10/Vπia, pH 2 release: TU50, TU51, TU53, TU54, TU55, TU56, TU57, TU58, TU60, TU62, TU63

• from CMTI, pH 5 release: CU22, CU25, CU27, CU28, CU29, CU31, CU32

• from LACI/F, pH 2 release: LU2, LU4, LU5, LU9, LU10, LU12

Individual isolate phage were tested for binding to immobilized urokinase or BSA. Because urokinase has a high sequence homology to tPA, the candidates also were tested for tPA binding. The results are shown m Figures 1, 2, 3 and 4. A potential affinity ligand was identified by having (1) significantly higher binding affinity for the target urokinase than the control phage, (2) a significantly higher binding affinity for the target under binding conditions (pH 7) than at elution conditions (pH 2), and (3) little or no binding to BSA. From the ELISA results, seven isolates (four from TN-6/1 and three from TN-10/VIIIa) were suitable for use as potential affinity ligands: TU33, TU36, TU39, TU42, TU53, TU56 and TU58. All seven isolates exhibited high affinity at pH 7 for urokinase and reduced affinity at pH 2. No ligand candidates were discovered from the CMTI or LACI/F hbraπes that bound with higher affinity to urokinase than the parental display phage. The isolates from TN-10/VIIIa were found to bind sham beads (control beads without uPA), and therefore the TN-6/1 candidates only were investigated further.

Amplification, isolation, and sequencing of the encoded insert DNA of the four analog- beaπng phage of TN-6/1 revealed the specific ammo acid sequences of the binding moieties for urokinase. Table 7 below shows the amino acid sequences of the vaπegated region (amino acid positions 1-17) for the 4 sequenced analogues of TN-6/1.

-30-

26

In the foregoing Table 7, the template sequence is SEQ ID NO: 14 and the TU33, TU36, TU39, and TU42 sequences are SEQ ID NOs: 7, 20, 21, and 22, respectively.

The two TN-6/1 analogues exhibiting the highest affinity for uPA in the phage ELISA were selected for further analysis. (See Fig. 1.)

Ligand Synthesis and Immobilization

Two of the four TN-6/1 derivative ligands, TU33 and TU42, were synthesized using the sequence information determined from the DNA of the phage isolates. Both of the ligands were chemically synthesized with the C-terminal Factor Xa cleavage site removed, and a hydrazide group placed at the C-terminus for easy immobilization to chromatographic supports with an aldehyde functionality.

Each ligand candidate was immobilized on an aldehyde-functional methacrylate resin support (TosoHaas formyl 650-M; Montgomeryville, PA). About 2.7 μmols of the ligand polypeptide were coupled to 1 ml of the activated chromatography support. Binding constants of the two ligands for uPA were determined by small scale equilibrium binding experiments.

The K_jS were calculated to be 392 nM for TU33 and 1.87 μM for TU42.

Column Testing Similar chromatographic protocols were used in testing the uPA isolation performance of the TU33 and TU42 uPA affinity ligands. 3 x 50 mm glass columns (350 μl Omnifit, Toms River, NJ) were packed with the ligand immobilized on the TosoHaas formyl 650-M support described above. Samples containing uPA and buffer solutions were pumped onto the column with a Watson-Marlow lOlu peristaltic pump. Chromatography was performed at 200 μl/min. or at a linear velocity of 170 cm/hr. Detection was made with 6 μl flow cells of either a Waters 996 Photodiode Array detector or a Waters 490e Programmable Multiwavelength UV/VIS detector Chromatograms were analyzed with a workstation utilizing Millennium chromatography manager software.

Two tests of chromatographic performance of each of the uPA affinity ligands (TU33 and TU42) were conducted. One test utilized a feed stream of pure uPA m buffer. The second test utilized a cell culture supernatant (from a human osteogenic sarcoma cell line) with 10% fetal bovine serum spiked with uPA. uPA loads were -250 μg for all sets of expeπments. The equilibration and wash buffer consisted of PBS/0.01% Tween 20. Feed stream linear velocity was 170 cm/hr (0.2 ml/mm). uPA elution was done at pH 2 under conditions of 20% acetomtπle in 30 mM H₃P0₄. Cleaning was performed using 30% and 70% isopropanol sequentially. uPA enzymatic activity was measured usmg a fluorogenic substrate, i.e., Boc-IEGR-AMC

(Novabiochem; LaJolla, CA) m a Tπs NaCl/PEG/Tπton X-100 buffer at pH 8. Enzymatic rate determination was made using a Perkm-Elmer 650-15 fluorescence spectrophotometer and a chart recorder The assay was able to measure pmol amounts of enzyme. The assay allowed determination of uPA m complex mixtures, confirmed structural mtegπty of the plasminogen activator m vaπous elution protocols, and provided confirmation of reversed phase mass determination. Spectral data m the range 200 nm to 350 nm were collected with 3.6 nm resolution. Figures 5, 6, 7, and 8 show the output traces collected at 280 nm.

Figure 5 is the chromatogram of a sample of pure uPA (312 μg uPA m buffer) over an affinity chromatography column having the immobilized TU33 uPA affinity ligand and eluted as descπbed above. The chromatogram shows shaφ elution of about 60.7% of the uPA mateπal after about 18 mm.

Figure 6 shows the results of a 208 μg sample of uPA isolated by contact with immobilized affinity ligand TU33 from a cell culture supernatant mixture as descπbed above The chromatogram shows elution of about 37.5% of the uPA mateπal after about 18 mm Reverse phase analysis of the TU33 column revealed uPA elution puπty of 13.3% (data not shown), suggesting that there is at least one impuπty in the complex mixture that co-elutes at a significant level with uPA for this ligand (and also TU42, as indicated below).

Figure 7 is the chromatogram of a sample of uPA (257 μg uPA in buffer) over an affinity chromatography column having the immobilized TU42 uPA affinity ligand and eluted as descπbed above. The chromatogram shows shaφ elution of about 47.4% of the uPA mateπal after about 17 mm.

Figure 8 shows the uPA isolation by affinity puπfication utilizing peptide ligand TU42 immobilized as descπbed above, from a cell culture supernatant sample spiked with 267 μg uPA as descπbed above. The chromatogram shows elution of about 30.1% of the uPA mateπal after

-32- about 18 min. Reverse phase analysis of the TU42 column revealed uPA elution purity of

12.8%) (data not shown).

From the data above, a new uPA binding domain was defined, comprising amino acid sequences of the formula: X₁-Cys-X₂-Trp-Trp-X₃-Cys-Gly-Ser (SEQ ID NO: 1 ), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp ; X₃ is Asp or His.

The following specific polypeptides corresponding to the TN-6/1 microprotein template of the TN-6/1 library were considered especially useful:

Ala-Glu-Gly-X_rCys-X₂-Tφ-Tφ-X₃-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 5), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; X₃ is Asp or His; and particularly the following: Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7);

Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 8). Further specified with this invention are the above affinity ligand peptides specifically modified for immobilization to chromatographic supports, including polypeptides of the general sequence: Ala-Glu-Gly-XrCys-X Tφ-Tφ-X Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 37), wherem X, is Val, He or Tyr ; X₂ is Ser or Asp ; X₃ is Asp or His; and specifically including the sequences:

Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 9), and Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10).

Example II: Testing of CMTI Analogue tPA Affinity Ligands for uPA Isolation

Twelve phage displaying CMTI peptides previously isolated from a screen against immobilized tissue plasminogen activator (tPA) were collected and screened for comparison against uPA. Although there is some sequence homology between tPA and uPA, it was not predictable that a phage isolate binding to tPA would also bind to uPA. The CMTI peptides displayed by the collected phage had the following sequences:

-33-

BSTΓΓUTE SHEET RULE 26

For uPA screening, the binding step was performed in the presence of PBS, 0.01% Tween 20, followed by washes with PBS. The twelve test phage were screened along with a control phage displaying the wild type CMTI sequence (SEQ ED NO: 13) and a control phage displaying no CMTI protein. Two elution steps were used: first the beads were suspended with 200 mM arginine, 150 mM NaCl, 100 mM ammonium acetate, and 0.01% Tween 20 at pH 4.5 and the beads tumbled for 10 minutes at room temperature; then the beads were pelleted, the eluate collected, and the beads were resuspended in 500 μl of pH 2 elution buffer (150 mM NaCl, 50 mM sodium citrate, pH 2.0), tumbled for 10 minutes at room temperature, pelleted, and the eluate was collected. Both the pH 4.5 and the pH 2 eluates were neutralized by transferring to a tube containing 130 μl of 1M HEPES (pH 8.0).

Titers of the phage appearing in the eluates were determined in order to calculate total number of phage pfu recovered. A fraction of input recovered (FIR) was determined by dividing the number of pfu recovered by the total applied to the beads. The pH 4.5 and pH 2 FIRs were added together to obtain a total FIR, and the phage were ranked according to total FIRs. A strong binder was indicated if the total FIR was at least two-fold higher than the FIR of

-34- the wild type control. Of the tested phage isolates, four were considered strong binders: CMTI- 4, CMTI-9, CMTI-19 and CMTI-II-01 (in order of positive FIR rank). CMTI-4 showed the strongest binding to uPA, however it was observed that the recovery of CMTI-4 phage was fairly evenly distributed between the two eluates, indicating that this isolate retained some affinity for the target uPA at pH 4.5. The next highest affinity binders, CMTI-9 and CMTI-19, showed shaφer release at pH 4.5 (i.e., most of those phage recovered were in the pH 4.5 fraction), and CMTI-9 was further investigated in uPA capture experiments.

Synthetic CMTI-9 peptide (SEQ ED NO: 23) was immobilized on an aldehyde- functional methacrylate resin support (TosoHaas formyl 650-M; Montgomeryville, PA). In addition, a single loop variant of this peptide, designated C9-TN8, having a short linker sequence and a terminal hydrazine group, i.e., RLCPKTDLGCMKDSDGGA-NH-NH2 (SEQ ID NO: 35) was immobilized on controlled pore glass beads (ProSep™ media; Bioprocessing, Ltd., Durham, UK). The ligand densities of the two prepared chromatography supports were 13 mg/ml (4μmol/ml) for CMTI-9/resin and 1.8 mg/ml (1 μmol/ml) for C9-TN8/media. Purified natural urokinase was dissolved at a concentration of 0.25-0.5 mg/ml in phosphate buffered saline (PBS), pH 7 + 0.01% Tween 20. Equilibration of the columns used the same buffer (pH 7 PBS, 0.01% Tween 20). The column volumes were 0.35 ml. Elution from the CMTI-9 column used 30 mM phosphoric acid, 150 mM NaCl, pH 1.3; elution of the C9-TN8 column used 100 mM glycine, 150 mM NaCl, 200 mM arginine, 0.01% Tween 20, pH 2. The operating flow rate was 0.2 ml/min. (or about 170 cm/hr.). uPA was loaded at 2.0 mg/ml gel on the CMTI-9/media column; uPA was loaded at 1.0 mg/ml gel on the C9-TN8/media column. The yield data are shown below:

The foregoing experiments point to additional new uPA binding moieties, comprising polypeptides including the general sequence:

-35- Arg-X,-Cys-X₂-X₃-X₄-X5-X6-X7-Cys-X₈-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val-Cys-Leu-

Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 36), wherein Xj is Leu or Tφ; X₂ is Ser or Pro; X₃ is Lys or Thr; X₄ is Ser, Tyr or Thr; X₅ is Ser, His,

Asp or Thr; X^ is Leu, Lys or Met; X₇ is Gly or Glu; X₈ is Met or Lys. specifically including the sequences:

Arg-Leu-Cys-Pro-Lys-Thr-Asp-Leu-Gly-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-

Val-Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 23),

Arg-Tφ-Cys-Pro-Lys-Thr-His-Lys-Glu-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val-

Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 24), Arg-Tφ-Cys-Pro-Lys-Ser-Thr-Met-Gly-Cys-Lys-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val-

Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 27), and

Arg-Tφ-Cys-Ser-Thr-Tyr-Ser-Leu-Gly-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val-

Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 29).

The foregoing data also reveal that a truncated peptide based on the CMTI domain is a useful urokinase binding moiety:

Arg-Leu-Cys-Pro-Lys-Thr-Asp-Leu-Gly-Cys-Met-Lys-Asp-Ser-Asp-Gly-Gly-Ala (SEQ ED

NO: 35). Based on results with this peptide, it is anticipated that the N-terminal portions of the other positive CMTI uPA affinity ligands, encompassing the first fifteen amino acids of SEQ ID

NO: 36, i.e., Arg-X,-Cys-X₂-X₃-X₄-X₅-X₆-X₇-Cys-X₈-Lys-Asp-Ser-Asp, wherein X, is Leu or Tφ; X₂ is Ser or Pro; X₃ is Lys or Thr; X₄ is Ser, Tyr or Thr; X₅ is Ser, His, Asp or Thr; X_β is

Leu, Lys or Met; X₇ is Gly or Glu; X₈ is Met or Lys, will also be useful for purification of urokinase by affinity chromatography.

Example III: Isolation of uPA-like Polypeptides fro Different Feed Streams A. Capture of Human uPA from Tobacco Extract

The CMTI-9 polypeptide uPA affinity media prepared as described above was tested in order to simulate separation of uPA as a product produced in recombinant tobacco plants. A tobacco extract was obtained from non-transgenic tobacco leaves shredded to simulate use of commercial mechanical damage promoter machinery and included secreted material at the 2-7 hour period eluted with PBS at pH 5.8. The extract was clarified by centrifugation, filtered through a 0.45 μM filter and adjusted to pH 7. Purified human uPA derived from urine, and virtually all in the high molecular weight, two-chain form, was spiked into extract at 100, 10 and 1 μg/ml to simulate different expression levels. Extract loads were pumped onto a 3 X 50 mm column (350 μL) packed with uPA affinity media (CMTI-9 polypeptide (SEQ ID NO:23) immobilized onto TosoHaas formyl 650- M chromatography media). The operational flow rate was 200 μL/min. or a linear velocity of 170 cm/hr. Loading level was kept constant at 100 μg. The column was washed with PBS and eluted with 100 mM glycine/150 mM NaCl at pH 2. The pH 2 eluate was immediately adjusted to pH 7. The collected fractions were analyzed for uPA activity via a fluorogenic substrate and the eluate further analyzed by reversed phase for purity.

The following is a summary of the recovery of uPA based on activity and purity based on reversed phase analysis: Sample μg uPA % Mass %Mass (normalized) %Purity

100 μg/ml

Load 133 100.0 100.0

Flowthrough 2 1.2 1.6 pH 2 Elution 96 72.2 98.4 83.6

73.4 100.0

10 μg/ml

Load 112 100.0 100.0

Flowthrough 7 6.1 10.6 pH 2 Elution 63 56.2 89.4 75.5

62.9 100.0

1 μg/ml

Load 97 100.0 100.0

Flowthrough 28 28.3 36.7 p pHH 22 EElluutitioonn 4477 4488..77 6633..33 37.0

76.9 100.0 While the purity decreases with decreasing uPA concentration, capture was maintained even at very low levels. The 10 μg/ml load was 10 ml extract over the 350 μL column and the 1 μg/ml load was 100 ml plant extract over the 350 μL column.

B. Capture of Human Single-Chain uPA (sc-uPA)

Single-chain uPA obtained from American Diagnostica (product #107) from a secreting cell line, at 1.1 mg/mL. 50 μg at lOOμg/mL in PBS was loaded onto the same type of 3 x 50 mm affinity column as described above. The operational flow rate was 200 μL/min or a linear velocity of 170 cm/hr. The column was washed with PBS and eluted with 100 mM glycine/150 mM NaCl pH 2. The pH 2 eluate was immediately adjusted to pH 7. The collected fractions were analyzed for uPA activity via a fluorogenic substrate and the eluate further analyzed by reversed phase for recovery. The results showed that >99% of the sc-uPA was found in the elution.

Following the foregoing description, the characteristics important for the separation of uPA from any feed stream according to any desired protocol can be engineered into the binding domains of a designed library, so that the method of this invention invariably leads to the isolation of affinity ligand candidates suitable for separation of the uPA under desirable conditions of binding and release. High yield of the uPA without inactivation or disruption of the product, with high purity, with the elimination of even closely related impurities, at acceptable cost and with re-usable or recyclable materials all can be achieved according to the present invention. Additional embodiments of the invention and alternative methods adapted to a particular uPA form or feed stream will be evident from studying the foregoing description. All such embodiments and obvious alternatives are intended to be within the scope of this invention, as defined by the claims that follow.

References

Boschetti, E., J. Chromatography, A 658: 207-236 (1994). Knight, P., Bio/Technology, 8: 200 (1990).

Ladner, R. C, "Constrained peptides as binding entities," Trends in Biotechnology, 13(10): 426- 430 (1995). Markland, W., Roberts, B.L., Ladner, R.C., "Selection for Protease Inhibitors Using Bacteriophage Display," Methods in Enzymology, 267: 28-51 (1996).

Narayanan, S.R., "Preparative affinity chromatography of proteins," J. Chrom. A, 658: 237-258 (1994).

Tkachuk, V., Stepanova, V., Little, P.J., Bobik, A., "Regulation and role of urokinase plasminogen activator in vascular remodelling," Clin. Exp. Pharmacol. Physiol, 23(9):759-765 (Sep 1996). Vedvick, T., Buckholtz, R.G., Engel, M., Urcam, M., Kinney, S., Provow, S., Siegel, R.S., and Thill, G.P., "High level secretion of biologically active aprotinin from the yeast Pichia pastoris" , J. Industrial Microbiol, 7: 197-202 (1991).

Wagner, S.L., Siegel, R.S., Vedvick, T.S., Raschke, W.C., and Van Nostrand, W.E., "High level expression, purification, and characterization of the Kunitz-type protease inhibitor domain of protease Nexin-2/amyloid ^β-protein precursor," Biochem. Biphys. Res. Comm., 186: 1138-1145 (1992).

Xing, R.H., Mazar, A., Henkin, J., Rabbani, S.A., "Prevention of breast cancer growth, invasion, and metastasis by antiestrogen tamoxifen alone or in combination with urokinase inhibitor B- 428," Cancer Res., 57(16):3585-3593 (Aug. 1997).

Each of the publications mentioned herein is incoφorated by reference.

Claims

CLAIMS:

1. A uPA binding moiety comprising a polypeptide having an amino acid sequence: X,-Cys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.

2. A uPA binding moiety comprising a polypeptide having an amino acid sequence: Ala-Glu-Gly-X,-Cys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser (SEQ ID NO: 2), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.

3. A uPA binding moiety comprising a polypeptide having an amino acid sequence selected from the group consisting of:

Ala-Glu-Gly-X,-Cys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 5) and Ala-Glu-Gly-X,-Cys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 6), wherein X[ is Val, He or Tyr ; X₂ is Ser or Asp; X₃ is Asp or His.

4. A uPa binding moiety comprising a polypeptide selected from the group consisting of: Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 8); Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 9); and Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10).

5. A binding moiety capable of binding to uPA in a solution comprising:

PBS, 0.1% BSA, 0.01% Tween 80 at 4┬░ C and dissociating from said uPA in a solution comprising: 50 mM citrate, 150 mM NaCl, 0.1% BSA, pH 2 at room temerature.

6. A method for detecting uPA or a uPA-like polypeptide in a solution suspected of containing it comprising:

(a) contacting such solution with a polypeptide according to Claim 1, and

(b) determining whether binding has occurred between said polypeptide and said uPA or uPA-like polypeptide.

7. A method for purifying uPA or uPA-like polypeptide comprising:

(a) immobilizing a binding moiety according to Claim 1 on a solid support;

(b) contacting a solution containing uPA or a uPA-like polypeptide with said support; and, thereafter,

(c) separating the solution from said support.

8. A method according to Claim 6 or Claim 7, wherein said binding moiety is a polypeptide including an amino acid sequence selected from the group consisting of: Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 8); Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 9); Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10); Arg-Leu-Cys-Pro-Lys-Thr-Asp-Leu-Gly-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys- Val-Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 23);

Arg-Tφ-Cys-Pro-Lys-Thr-His-Lys-Glu-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val- Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 24);

Arg-Tφ-Cys-Pro-Lys-Ser-Thr-Met-Gly-Cys-Lys-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val- Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 27); and

Arg-Tφ-Cys-Ser-Thr-Tyr-Ser-Leu-Gly-Cys-Met-Lys-Asp-Ser-Asp-Cys-Leu-Ala-Glu-Cys-Val- Cys-Leu-Glu-His-Gly-Tyr-Cys-Gly (SEQ ID NO: 29).

9. A recombinant bacteriophage expressing exogenous DNA encoding a binding peptide capable of binding to uPA, said binding peptide having an amino acid sequence: X_rCys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X- is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.

10. A recombinant bacteriophage expressing exogenous DNA encoding a uPA binding moiety having an amino acid sequence selected from the group consisting of: Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 8); Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 9); and

Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10), wherein said binding moiety is displayed on the surface of said bacteriophage.

11. A method for detecting uPA or a uPA-like polypeptide in a sample, comprising contacting said sample with a bacteriophage according to Claim 9 or Claim 10 and detecting whether binding has occurred between said bacteriophage and uPA or a uPA-like polypeptide.

12. Separation media comprising:

(a) a porous chromatographic matrix material, and, immobilized thereon,

(b) a uPA binding moiety comprising a polypeptide having an amino acid sequence selected from the group consisting of: X_rCys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X[ is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.

13. The separation media according to Claim 15 wherem said polypeptide is selected from the group consisting of:

Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ID NO: 7); Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Tyr-Ile-Glu-Gly-Arg (SEQ ED NO: 8); Ala-Glu-Gly-Val-Cys-Ser-Tφ-Tφ-Asp-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ED NO: 9); and Ala-Glu-Gly-Tyr-Cys-Asp-Tφ-Tφ-His-Cys-Gly-Ser-Glu-Gly-Gly-Gly-Ser (SEQ ID NO: 10).

14. A method for separating uPA or a uPA-like polypeptide from a solution containing it comprising:

(a) contacting said solution with separation media as defined in Claim 12 under binding conditions,

(b) removing unbound material, and

(c) eluting bound uPA or a uPA-like polypeptide from said separation media.

15. A polynucleotide encoding a uPA binding moiety comprising a polypeptide of the formula: X,-Cys-X₂-T╧å-T╧å-X₃-Cys-Gly-Ser (SEQ ID NO: 1), wherein X, is Val, He or Tyr ; X₂ is Ser or Asp; and X₃ is Asp or His.