Modulation of cellular proliferation with thymidine phosphorylase

Abstract

A method of modulating cellular proliferation by the application of a thymidine phosphorylase to an organism. In a further aspect of the subject method, the thymidine phosphorylase is a conjugate which includes a targeting portion adapted to target the conjugate to a specific cell type or anatomical location. The thymidine phosphorylase has a thymidine phosphorylase activity of at least about 5%, preferably at least about 50% and, most preferably, at least about 90%, of the native E. coli enzyme.

Description

The present invention relates to the medical use of thymidine phosphorylase in the modulation of cellular proliferation.

BACKGROUND OF THE INVENTION

Thymidine phosphorylase (thymine: orthophosphate deoxyribosyl transferase, EC 2.4.2.4) is a cytosolic enzyme which catalyses the reversible phosphorolysis of thymidine and other pyrimidine 2′-deoxyribosides, except for 4-amino substituted compounds such as 2′-deoxycytidine, as follows:

PyrdR+P i ←→Pyr+dR−1−P

The phosphorolytic and synthetic reactions may also be used to transfer a deoxyribose moiety of one deoxynucleoside to form a second deoxynucleoside in a nucleoside deoxyribosyl transferase reaction (Schwartz, M. (1971) Eur. J. Biochem. 21, 191-198).

Thymidine phosphorylase has been purified and characterized from a number of micro-organisms (Schwartz, M. (1971) Eur. J. Biochem. 21, 191-198, Schwartz, M. (1978) Methods Enzymol. 51, 442-445; Avraham, Y. et al (1990) Biochim. Biophys. Acta 1040, 287-293; Hoffee, P. A. et al (1978) Methods Enzymol. 51, 437-442) and from human tissues (Desgranges, C. et al (1981) Biochim. Biophys. Acta. 654, 211-218; Kubilus, J. et al (1978) Biochim. Biophys. Acta 527, 221-228; Yoshimura et al (1990) Biochim. Biophys. Acta 1034, 107-113). Escherichia coli thymidine phosphorylase is a dimer of 90 kD composed of two identical subunits with a molecular weight of 45 kD (Schwartz, M. (1978) Methods Enzymol. 51, 442-445; Walter, M. et al (1990) J. Biol. Chem. 265, 14016-14022). The three-dimensional crystal structure of Escherichia coli has been determined to resolution of 2.8 Å (Walter, M. et al (1990) J. Biol. Chem. 265, 14016-14022). The monomer subunit consists of a small α-helical domain and a large α/β domain. The active site, which binds both thymidine and phosphate, has been located in a cleft between the two domains.

Human thymidine phosphorylase has been identified in many tissues including lymphocytes, heart, spleen, lung and placenta (Yoshimura, A. et al (1990) Biochim. Biophys. Acta. 1034, 107-113) and purified from both placenta (Yoshimura, A. et al (1990) Biochim. Biophys. Acta. 1034, 107-113; Kubilus, J. (1978) Biochem. Biophys. Acta. 527, 221-228) and platelets (Desgranges, C. (1981) Biochim. Biophys. Acta 654, 211-218). It has been suggested that thymidine phosphorylase plays an essential role in maintaining intracellular thymidine homeostasis (Shaw, T. et al (1988) Mutant Res. 200, 99-116). The thymidine salvage pathway requires the action of a permease to transport thymidine across the lipid bilayer. Intracellular thymidine is then phosphorylated by thymidine kinase, generating thymidine monophosphate (TMP) which is further phosphorylated to generate thymidine triphosphate.

Thymidine triphosphate not only regulates the thymidine salvage pathway by inhibition of thymidine kinase, but also inhibits the production of other deoxyribonucleotides by allosteric effects on ribonucleotide diphosphate reductase. The action of thymidine phosphorylase may therefore be to regulate the size of the intracellular thymidine nucleotide pool and hence the size of the other nucleotide pools via ribonucleotide diphosphate reductase as suggested by Shaw, T. et al, Mutant Res. 200, 99-116 (1988), and hence regulate DNA synthesis.

There has been no suggestion that thymidine phosphorylase could be used as an extracellular growth factor in medicine.

SUMMARY OF THE INVENTION

In accordance with the present invention, there is provided a method of modulating cellular proliferation by the application of a thymidine phosphorylase to an organism. In a further aspect of the subject method, the thymidine phosphorylase is a conjugate which includes a targeting portion adapted to target the conjugate to a specific cell type or anatomical location.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows the amino acid sequence of E. coli thymidine phosphorylase;

FIGS. 2 , 2 A and 2 B show the nucleotide sequences of Oligonucleotides 1 to 40;

FIG. 3 shows the construction of plasmid pDTP1;

FIGS. 4 , 4 A, 4 B, 4 C, 4 D and 4 E show the nucleotide sequence ligated into pUC19 to form pDTP1 (FIG. 3 );

FIGS. 5 to 17 show the construction of plasmids pDTP2, pDTP3, pDTP4, pDTP5, pAYE333, pAYE334, pAYE328, pAYE335, pDBP5, pDBP6, pDTP6, pDTP7 and pDTP8, respectively;

FIGS. 18A , 18 B and 18 C show the amino acids in various regions of E. coli thymidine phosphorylase, together with corresponding PCR primers;

FIG. 19 shows the effect of thymidine phosphorylase on HUVE cell mitogenesis (represents CPAE cells grown in the absence of thymidine phosphorylase; ▪ represents cells grown in presence of 100 ng/ml thymidine phosphorylase);

FIG. 20 shows the effect of thymidine phosphorylase on CPAE cell proliferation; and

FIGS. 21A and 21B show the terminal structure of the LewisX and sialylated LewisX counter receptor.

DETAILED DESCRIPTION OF THE INVENTION

Cellular proliferation as utilized herein is defined as being any increase in cell number in the assay using CPAE cells described in Example 5.

The thymidine phosphorylase utilized in the method of the invention may be native thymidine phosphorylase from a prokaryotic or eukaryotic source, or a fragment of the native enzyme having thymidine phosphorylase enzymatic activity, or any other polypeptide having thymidine phosphorylase activity. Thymidine phosphorylase enzymatic activity is defined herein as phosphorolysis and/or nucleoside deoxyribosyl transferase activity. Whole thymidine phosphorylase compounds or fragments may be generated by proteolytic cleavage or by recombinant DNA techniques. The thymidine phosphorylase useful in the method of the invention has a thymidine phosphorylase activity of at least about 5%, preferably at least about 50%, most preferably at least about 90%, of the activity of the native E. coli enzyme.

Thymidine phosphorylase activity is determined spectrophotometrically, relying on the thousand-fold difference in the molar extinction co-efficients of thymidine and thymine. In carrying out the determination, saturating amounts of thymidine (1 mM) and KH 2 PO 4 (0.2M), pH 7.4 are combined in a suitable 1 ml reaction tube with either 0.1-5 μg/ml E. coli thymidine phosphorylase or 1-50 μg/ml human thymidine phosphorylase, and the reaction at 25° C. is followed by monitoring the decrease in absorbance at 290 nm. A decrease in absorbance of 1 corresponds to the conversion of a 1 mM solution of thymidine to thymine, thus the decrease in absorbance can be used to calculate the V max of the enzyme (μmoles min −1 mg −1 ). The thymidine phosphorylase utilized in the method of the invention does not include nature-identical human thymidine phosphorylase since it possesses, at best, only about 5% of the activity of the native E. coli enzyme.

Domains of thymidine phosphorylase can also be expressed. Preferred regions include amino acids 80-130, and analogues thereof (although such analogues preferably retain one or more of amino acids Lys 84 , His 85 , Ser 86 , Ser 95 , Ser 113 and Thr 123 ); a region including amino acids 165-220, or an analogue thereof (preferably retaining one or more of Arg 171 , Ser 186 , and Lys 190 ); or both regions 80-130 and 165-220, see FIG. 1 . Other preferred regions include amino acids 1-241, 79-241 and 79-440. These regions can be expressed, as individual polypeptides or as parts of larger polypeptides, as recombinant proteins in any one of a number of host expression systems including mammalian cells, Escherichia coli and Saccharomyces cerevisiae when appropriately adapted by provision of translation/transcription initiation and termination sequences.

The techniques for preparing the polypeptides possessing thymidine phosphorylase activity described above are within the skill of the art. Preferred fragments are as discussed above. However, it is likewise within the skill of the art to determine whether a given fragment or analogue meets the criteria of the present invention since the identification and characterization of thymidine phosphorylase were known prior to the subject invention. Hence, it is within the skill of the art utilizing known techniques to determine whether a given fragment possesses at least 5% of the thymidine phosphorylase activity of the native E. coli enzyme.

The thymidine phosphorylase utilized in the subject invention may be produced by recombinant DNA techniques in heterologous hosts. Preferably, the host lacks an endogenous thymidine phosphorylase which will facilitate purification of the heterologous thymidine phosphorylase. Saccharomyces cerevisiae may lack an endogenous thymidine phosphorylase. Hosts possessing thymidine phosphorylase can have the endogenous thymidine phosphorylase gene deleted, for example, by site-directed mutagenesis.

Although rapid cellular proliferation is a hallmark of certain disease states it is also a prerequisite of many normal cellular processes, e.g. in response to tissue damage. Localized, rapid tissue growth and tissue remodelling are required in the wound healing process, where the growth of endothelial cells and the regeneration of a vascular network are necessary following surgical intervention, or for the treatment of burns and ulcers. Endothelial cell regeneration is also required following various surgical techniques where damage or injury results to the endothelial cell lining of the vascular network, for example balloon angioplasty or coronary by-pass surgery; or where an endothelial cell lining has to be generated de novo, for example, on synthetic vascular grafts or prostheses.

The activity of a growth- or proliferation-promoting agent may be enhanced by causing it to be accumulated at the site where such activity is required. Further, such targeting may enable the dosage of the agent required to be reduced since, by accumulating the agent at the required site, the efficacy of a given dosage may be significantly enhanced. Thymidine phosphorylase can be localized by conjugating it with a targeting agent by use of cross-linking agents as well as by recombinant DNA techniques whereby the thymidine phosphorylase DNA sequence, or a functional portion of it, is cloned adjacent to the DNA sequence of the agent when the agent is a protein, and the conjugate expressed as a fusion protein. The targeting agent can be any monoclonal antibody, or active portion thereof, eg Fab or F(ab′) 2 fragment, a ligand (natural or synthetic) recognized by an endothelial cell surface receptor or a functional portion thereof, or any other agent which interacts with protein or structures of the endothelial cell.

Suitable active antibody portions, e.g. Fab or F(ab′) 2 fragments of antibodies, will retain antigen/target binding but have low non-specific binding. Fab or F(ab′) 2 fragments may be obtained by protease digestion, for example using immobilized Protein A and pepsin/papain digestion using ImmunoPure Fab and ImmunoPure F(ab′) 2 preparation kits (Pierce). Other active portions of antibodies may be obtained by reduction of the antibodies or antibody fragments into separate heavy and light chains.

Molecules targeted by thymidine phosphorylase/antibody conjugates or gene fusions can be endothelial cell surface molecules, extracellular matrix components, for example collagen, fibronectin or laminin, or other blood vessel wall structures. Examples of monoclonal antibodies raised to endothelial surface antigens are Tük3 (Dako) and QBend10 (Serotec) which recognize CD34, a glycosylated endothelial cell surface transmembrane protein. Other monoclonal antibodies raised to endothelial cell surface antigens include 9G11, JC70, and By126 (British Bio-technology) raised to CD31 (also known as PECAM-1) and ESIVC7 raised to the CD36 antigen, which is the thrombospondin receptor (Kuzu et al (1992) J. Clin. Pathol. 45, 143-148). QBend20, QBend30 and QBend40 (Serotec) are further examples of other monoclonal antibodies which recognize endothelial cell surface antigens.

The endothelial cell surface molecules to which the targeting antibodies are raised can be non-specific and recognize a number of different endothelial cell types from different tissues, or can be specific for certain endothelial cell types. Antibody A10-33/1 (Serotec) recognizes endothelial cells in metastatic melanomas, H4-7/33 (Serotec) recognizes endothelial cells from small capillaries and a wide range of tumor cells, HM15/3 (Serotec) recognizes sinusoidal endothelial cells, and 1F/10 (Serotec) binds to a 250 kD surface protein on continuous endothelium. Antibodies raised to antigens involved in haemostasis and inflammation can also be used. Antibody 4D10 (Serotec) and BB11 (Benjamin et al (1990) Biochem. Biophys. Res. Commun. 171, 348-353) recognizes ELAM-1 present on endothelial cells in acute inflamed tissues. Antibody 4B9 (Carlo, T. and Harlan, J. (1990) Immunol. Rev. 114, 1-24) recognizes the VCAM adhesion protein. Antibody 84H10 (Makgabo, M. et al (1988) Nature 331, 86-88) recognizes the ICAM1 adhesion protein. Antibody EN7/58 (Serotec) recognizes antigens present on inflamed endothelium and on cells adhering to the endothelial cells. Antibody KG7/30 recognizes a FVIII related protein on endothelial surfaces of inflamed tissues and tumors.

The cytokines IL-1 and TNF stimulate cultured endothelial cells to acquire adhesive properties for various peripheral blood leukocytes in vitro (Bevilaqua, M. et al (1985) J. Clin. Invest. 76, 2003; Schleimer, R. et al (1986) J. Immunol. 136, 649; Lamas, A. et al (1988) J. Immunol. 140, 1500; Bochner, B. et al (1988) J. Clin. Invest. 81, 1355). This adhesiveness is associated with the induction on endothelial cells of a number of adhesive molecules, including ICAM-1, ELAM-1, GMP-140 (also known as PADGEM or CD62) and VCAM-1. These adhesive molecules recognize counter receptors on the surface of the target cell. VCAM-1 recognizes an antigen known as VLA-4, also known as CD49d/CD29 and member of the integrin family (Elices, M. et al (1990) Cell 60, 577; Schwartz, B. et al (1990) J. Clin. Invest. 85, 2019). ICAM-1 recognizes an antigen known as LFA-1, also known as CD11a/CD18, another member of the integrin family (Martin, S. et al (1987) Cell 51, 813-819 Fujita, H. et al (1991) Biochem. Biophys. Res. Comm. 177, 664-672). ELAM-1 and GMP-140 (GMP-140 is also known as CD62 or PADGEM), recognize an antigen known as LewisX, also known as CD15, or sialyl-LewisX (Larsen, E. et al (1990) Cell 63, 467-474; McEver, R. (1991) J. Cell. Biochem. 45, 156-161; Shimizu, Y. et al (1991) Nature 349, 799; Picker, L. et al (1991) Nature 349, 796-798; Polley, M. et al (1991) Proc. Natl. Acad. Sci. USA. 88, 6224-6228; Lowe, J. et al (1990) Cell 63, 475-484; Tiemeyer, M. et al (1991) Proc. Natl. Acad. Sci. USA. 88, 1138-1142).

Monoclonal antibodies to either the receptor expressed on the surface of the endothelial cell or counter receptor on the surface of the responding cell have been shown to block interaction of the components necessary for this cell-cell recognition and were instrumental in establishing the mode of recognition (for references see above).

A second aspect of this invention provides a conjugate of the thymidine phosphorylase as defined above and a moiety which specifically binds endothelial cells. Thymidine phosphorylase can be conjugated, by crosslinking or by recombinant DNA techniques, to natural or synthetic ligands which interact with receptors on the endothelial cell surface. Such ligands include growth factors, for example vascular permeability factor (Gitay-Goren, H. et al (1992) J. Biol. Chem. 267, 6093-6098; Bikfalin, A. et al (1991) J. Cell. Phys. 149, 50-59; Tischer, E. et al (1991) 266, 11947-11954; Conn, G. et al (1990) PNAS 87, 2628-2632; Keck, P. et al (1989) Science 246, 1309-1312; Leung, D. W. et al (1989) Science 246, 1306-1309); platelet-derived growth factor (Beitz, J. et al (1991) PNAS 88, 2021-2025); and well as other biomolecules such as transferrin and urokinase (Haddock, R. et al (1991) J. Biol. Chem. 266, 21466-21473). The ligand domain of the conjugates will be recognized by the endothelial cell surface receptor for that ligand and will target the thymidine phosphorylase to the endothelium.

An agent with thymidine phosphorylase activity, for example thymidine phosphorylase or a fragment of thymidine phosphorylase with thymidine phosphorylase activity as defined above, can also be directed toward a specific adhesion molecule by cross-linking the agent to the counter receptor for that adhesion molecule. In the example of ELAM-1 mediated adhesion, the counter receptor is a carbohydrate determinant known as Lewis-X or sialylated Lewis-X, the terminal structure of which is given in FIGS. 21A and 21B . Synthetic carbohydrates with this terminal structure (Kameyama, A. et al (1991) Carbohydrate. Res. 209, C1-C4) or purified from natural sources, for example LNFIII (Calbiochem), are available. The terminal Lewis-X or sialyl Lewis-X determinant can be cross-linked to free sulfydryl groups within the thymidine phosphorylase agent as described in Example 1. This allows specific targeting of the agent to endothelial cells presenting the ELAM-1 adhesion molecule.

This moiety may be a monoclonal antibody to endothelial cell surface receptors such as ICAM-1, ELAM-1, GMP-140 or VCAM-1. Alternatively, this moiety may be the counter receptor itself, or a functional portion thereof. Fusion may be achieved by i) chemical cross linking of the moiety, be it a monoclonal antibody or the counter receptor, by techniques known in the art, or ii) by recombinant DNA technology whereby the moiety, when it is a single polypeptide chain, is expressed as a gene fusion with the agent in a suitable host.

A number of cell- or stage-specific antibodies have been described. These include, for example antibodies to endothelial cell adhesion molecules, including antibody BB11 (anti-ELAM, Benjamin, C. et al (1990), Biochem. Biophys. Res. Commun. 171, 348-353), antibody 4B9 (anti-VCAM, Carlo, T. and Harlan, J. (1990) Immunol. Rev. 114, 1-24) and antibody 84H10 (anti-ICAM1, Makgobo, M. et al (1988) Nature 331, 86-88). These antibodies, or antibodies like them, can be covalently joined to an agent with structural homology to thymidine phosphorylase, be it full length thymidine phosphorylase or a fragment of it with enzyme activity. This can be achieved by gene fusion whereby the DNA sequence encoding the agent with thymidine phosphorylase activity is spliced into the genes encoding either the heavy or light chain of the antibody. Alternatively, the agent can be covalently cross-linked to the antibody via one of a number of bi-functional cross-linking reagents such as, for example, disuccinimidyl suberate (DSS); bis (sulfosuccinimidyl) suberate (BS 3 ); dimethyl adipimidate-2 HCl (DMA); dimethyl pimelimidate-2 HCl (DMP); dimethyl suberimidate-2 HCl (DMS); bismaleimidohexane (BMH); m-maleimidobenzoyl-N-hydroxysuccinimide ester (MBS); m-maleimido-benzoyl-N-hydroxysulfosuccinimide ester (sulfo-MBS); succinimidyl 4-(p-maleimidophenyl)butyrate (SMPB); sulfosuccinimidyl 4-(p-maleimidophenyl)butyrate (sulfo-SMPB); N-succinimidyl (4-iodoacetyl) aminobenzoate (SIAB); sulfosuccinimidyl (4-iodoacetyl) aminobenzoate (sulfo-SIAB); succinimidyl 4-(N-maleimidomethyl)cyclohexane-1-carboxylate (SMCC); sulfosuccinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate (Sulfo-SMCC) or 1,5-difluoro-2,4-dinitrobenzene (DFDNB), (Pierce).

While appropriate concentrations of thymidine phosphorylase can be demonstrated to stimulate mammalian cell growth, at superoptimal concentrations mitogenesis can be reduced or inhibited (FIG. 19 ). Therefore, by targeting thymidine phosphorylase to certain cell types where rapid or uncontrolled cellular proliferation is associated with certain disease states, proliferation of these cell types can be reduced.

Antibodies recognizing antigens related to malignant transformation and angiogenesis can also be used: for example EN2/3 (Serotec) recognizes an antigen characteristic of malignant transformed endothelial cells; EN7/44 (Serotec) recognizes an angiogenesis related antigen present on proliferating, migrating and budding endothelial cells; and H3-5/47 recognizes endothelial cells in angioblasts, angiomas, angiosarcomas and perivascular cells in psoriasis and arthritic tissues.

Alternatively, the entity which is recognized by the targeting portion may be a suitable entity which is specifically expressed by tumor cells, which entity is not expressed, or at least not with such frequency, in cells into which one does not wish to introduce the thymidine phosphorylase. The entity which is recognized will often be an antigen. Examples of antigens include those listed in Table 1 below. Monoclonal antibodies which will bind specifically to many of these antigens are already known (for example those given in the Table) but in any case, with today's techniques in relation to monoclonal antibody technology, antibodies can be prepared to most antigens. The antigen-specific portion may be an entire antibody (usually, for convenience and specificity, a monoclonal antibody), a part or parts thereof (for example an F ab fragment, F(ab′) 2 , dab or “minimum recognition unit”) or a synthetic antibody or part thereof. A compound comprising only part of an antibody may be advantageous by virtue of being less likely to undergo non-specific binding due to the F c part.

Suitable monoclonal antibodies to selected antigens may be prepared by known techniques, for example those disclosed in “Monoclonal Antibodies: A manual of techniques”, H. Zola (CRC Press, 1988) and in “Monoclonal Hybridoma Antibodies: Techniques and Applications”, J. G. R. Hurrell (CRC Press, 1982). All references mentioned in this specification are incorporated herein by reference. Bispecific antibodies may be prepared by cell fusion, by reassociation of monovalent fragments or by chemical cross-linking of whole antibodies, with one part of the resulting bispecific antibody being directed to the cell-specific antigen and the other to the thymidine phosphorylase. The bispecific antibody can be administered bound to the thymidine phosphorylase or it can be administered first, followed by the thymidine phosphorylase. The former is preferred. Methods for preparing bispecific antibodies are disclosed in Corvalan et al (1987) Cancer Immunol. Immunother. 24, 127-132 and 133-137 and 138-143. Bispecific antibodies, chimaeric antibodies and single chain antibodies are discussed generally by Williams in Tibtech , February 1988, Vol. 6, 36-42, Neuberger et al (8 th International Biotechnology Symposium, 1988, Part 2, 792-799) and Tan and Morrison ( Adv. Drug Delivery Reviews 2, (1988), 129-142). Suitably prepared non-human antibodies can be “humanized” in known ways, for example by inserting the CDR regions of mouse antibodies into the framework of human antibodies. IgG class antibodies are preferred.

TABLE 1
Antigen	Antibody	Existing Uses
1. Tumor Associated Antigens
Carcino-embryonic	{C46 (Amersham)	Imaging & Therapy of
Antigen	{85A12 (Unipath)	colon/rectum tumors.
Placental Alkaline	H17E2 (ICRF, Travers	Imaging & Therapy of
Phosphatase	& Bodmer	testicular and ovarian
		cancers.
Pan carcinoma	NR-LU-10 (NeoRx	Imaging & Therapy of
	Corporation)	various carcinomas incl.
		small cell lung cancer.
Polymorphic	HMFG1 (Taylor-	Imaging & Therapy of
Epithelial	Papadimitriou, ICRF)	ovarian cancer, pleural
Mucin (Human milk		effusions.
fat globule)
β-human Chorionic	W14	Targeting of enzyme
Gonadotropin		(CPG2) to human xenograft
		choriocarcinoma in nude
		mice. (Searle et al (1981)
		Br. J. Cancer 44,
		137-144).
a Carbohydrate on	L6 (IgG2a) 1	Targeting of alkaline
Human Carcinomas		phosphatase. (Senter et al
		(1988) P.N.A.S. 85,
		4842-4846.
CD20 Antigen on B	1F5 (IgG2a) 2	Targeting of alkaline
Lymphoma (normal		phosphatase. (Senter et al
and neoplastic)		(1988) P.N.A.S. 85,
		4842-4846.
2. Immune Cell Antigens
Pan T Lymphocyte	OKT-3 (Ortho)	As anti-rejection therapy
Surface Antigen		for kidney transplants.
(CD3)
B-lymphocyte	RFB4 (Janossy, Royal	Immunotoxin therapy of B
Surface Antigen	Free Hospital)	cell lymphoma.
(CD22)
Pan T lymphocyte	H65 (Bodmer,	Immunotoxin treatment of
Surface Antigen	Knowles ICRF,	Acute Graft versus Host
(CD5)	Licensed to Xoma	disease, Rheumatoid
	Corp., USA)	Arthritis.
1 Hellström et al (1986) Cancer Res. 46, 3917-3923
2 Clarke et al (1985) P.N.A.S. 82, 1766-1770
Other antigens include alphafoetoprotein, Ca-125 and prostate specific antigen.

If applied to the treatment of CML or ALL, the ligand binding molecules can be monoclonal antibodies against leukaemia-associated antigens. Examples of these are: anti-CALLA (common acute lymphoblastic leukaemia-associated antigen), J5, BA-3, RFB-1, BA-2, SJ-9A4 Du-ALL-1, anti-3-3, anti-3-40, SN1 and CALL2, described in Foon, K. A. et al 1986 Blood 68(1), 1-31, “Review: Immunologic Classification of Leukemia and Lymphoma”. The ligand binding molecules can also be antibodies that identify myeloid cell surface antigens, or antibodies that are reactive with B or T lymphocytes, respectively. Examples of such antibodies are those which identify human myeloid cell surface antigens or those which are reactive with human B or T lymphocytes as described in Foon, K. A. Id. Additional examples are antibodies B43, CD22 and CD19 which are reactive with B lymphocytes can also be used.

Alternatively, the entity which is recognized may or may not be antigenic but can be recognized and selectively bound to in some other way. For example, it may be a characteristic cell surface receptor such as the receptor for melanocyte-stimulating hormone (MSH) which is expressed in high numbers in melanoma cells. The targeting portion may then be a compound or part thereof which specifically binds to the entity in a non-immune sense, for example as a substrate or analogue thereof for a cell-surface enzyme or as a messenger. In the case of melanoma cells, the targeting portion may be MSH itself or a part thereof which binds to the MSH receptor. Such MSH peptides are disclosed in, for example, Al-Obeidi et al (1980) J. Med. Chem. 32, 174. The specificity may be indirect: a first cell-specific antibody may be administered, followed by a conjugate of the invention directed against the first antibody. Preferably, the entity which is recognized is not secreted to any relevant extent into body fluids, since otherwise the requisite specificity may not be achieved.

The targeting portion of the conjugate of this embodiment of the invention may be linked to the thymidine phosphorylase by any of the conventional ways of linking compounds, for example by disulfide, amide or thioether bonds, such as those generally described in Goodchild, supra or in Connolly (1985) Nucl. Acids Res. 13(12), 4485-4502 or in PCT/US85/03312.

When applied topically in accordance with the present invention, the thymidine phosphorylase can be incorporated into an inert cream or base so that it is stabilized. Such a vehicle facilitates even application of the thymidine phosphorylase and maintains the agent at the site of application. This can also be achieved by incorporation of the thymidine phosphorylase, optionally as a component of an inert cream or base, into a bandage or dressing. Alternatively, the thymidine phosphorylase can be applied topically in the form of an aerosol where the thymidine phosphorylase has been previously dissolved or resuspended in, for example, a saline solution. Such formulations can be used either to promote wound healing or to combat tumors. The pharmaceutical carrier may also take the form of a graft, to which the thymidine phosphorylase is cross-linked. More specifically in the case of combatting tumors, injectable formulations of the thymidine phosphorylase can be prepared.

The anti-tumor conjugates of the invention may be administered in any suitable way, usually parenterally, for example intravenously, intraperitoneally or, preferably (for bladder cancers), intra-vesically (ie into the bladder), or directly into the tumor, in standard sterile, non-pyrogenic formulations of diluents and carriers, for example isotonic saline (when administered intravenously). If needed, because the compound of the invention may be immunogenic, cyclosporin or some other immunosuppressant can be administered to provide a longer period for treatment but usually this will not be necessary.

Particular tumors suitable for treatment in accordance with the invention include cancers of the uterine cervix, head, neck, brain gliomas, breast, colon, oesophagus, stomach, liver, pancreas and metastatic forms of any of these.

The invention is further illustrated by the following examples, it being understood that the details given therein are not intended to limit the invention in any manner.

EXAMPLE 1

Expression of an Agent with Thymidine Phosphorylase Activity

Unless otherwise stated, all procedures were carried out as described by Maniatis, Fritsch and Sambrook, “Molecular Cloning: A Laboratory Handbook”, Cold Spring Harbor Laboratory (1982).

The primary amino acid sequence of Escherichia coli thymidine phosphorylase, FIG. 1 , has been described (Walter, M. et al (1990) J. Biol. Chem. 265, 14016-14022). A double stranded cDNA was prepared by annealing 40 overlapping single-stranded oligonucleotides. These were designed to encode the amino acid sequence of thymidine phosphorylase, as described by Walter, M. et al (1990), see above. The DNA sequence (5′→3′) of oligonucleotides 1 to 40 is given in FIGS. 2 to 2 B. Oligonucleotides 1 to 8 were prepared on an Applied Biosystems 380B DNA synthesiser. The oligonucleotides were phosphorylated by T4 polynucleotide kinase. Phosphorylated oligonucleotides 1 to 8 (25 pmol each) were annealed in water for 5 min at 60° C., following which the solution was allowed to cool to 15° C. over the next 60 min. The mixture was made up to 50 mM Tris/HCl pH7.5, 10 mM MgCl 2 , 10 mM dithiothreitol, 1 mM ATP (DNA ligase buffer) and T4 DNA ligase added. The ligation was incubated at 15° C. for 18 hours. The ligated product (oligonucleotide duplex I) was purified from a 10% polyacrylamide gel, electrophoresed under nondenaturing conditions. Oligonucleotides 9 to 20 were phosphorylated and ligated in a similar way to generate oligonucleotide duplex II, while oligonucleotides 21 to 40, when phosphorylated and ligated, generated oligonucleotide duplex III. Oligonucleotide duplexes I, II and III were ligated into the SmaI site of pUC19, generating plasmid pDTP1 (FIG. 3 ). The sequence of the DNA insert is shown in FIGS. 4 to 4 E. Two double-stranded oligonucleotide adapters, oligonucleotide duplexes IV and V, were prepared by annealing oligonucleotides 41 and 42, and oligonucleotides 43 and 44 respectively.

Oligonucleotide Duplex IV

5′-CATGTTATTC -3′ (oligo 41, SEQ42)
3′-  AATAAGAAT-5′ oligonucleotide 42

Oligonucleotide Duplex V

5′-ATCTGAATAAC    -3′ (oligo 43, SEQ43)
3′-TAGACTTATTGGTAC-5′ (oligo 44, SEQ44)

These two adaptors, along with the 1.35 kbp DdeI-EcoRV thymidine phosphorylase DNA sequence from plasmid pDTP1, were ligated into the Escherichia coli expression vector pTrc99A (Pharmacia), digested with NcoI to generate plasmid pDTP2 (FIG. 5 ). The 1.34 kbp NcoI DNA insert encodes Escherichia coil thymidine phosphorylase with a translation initiation codon, 5′-ATG-3′, inserted 5′ to the leucine codon at amino acid position 1 (see FIG. 1 ) and a translation termination codon, 5′-TAA-3′, inserted 3′ to the glutamic acid codon at amino acid position 440 (see FIG. 1 ). Following the introduction of plasmid pDTP2 into Escherichia coli and selection for ampicillin-resistant transformants, thymidine phosphorylase can be expressed from the strong trc promoter.

By introducing EcoRI restriction recognition sites at each end of the synthetic thymidine phosphorylase DNA sequence, thymidine phosphorylase can be expressed from the IPTG inducible tac promoter system present in plasmid pKK223-3 (Pharmacia). This was achieved by the same methodology as described above for the trc promoter system. Two double-stranded oligonucleotide adaptors, oligonucleotide duplex VI and VII, were prepared by annealing oligonucleotides 45 and 46 and oligonucleotides 47 and 48, respectively.

Oligonucleotide Duplex VI

5′-AATTCATGTTATTC   -3′ (oligo 46, SEQ46)
3′-    GTACAATAAGAAT-5′ (oligo 46, SEQ46)

Oligonucleotide Duplex VII

5′-ATCTGAATAAG    -3′ (oligo 47, SEQ47)
3′-TAGACTTATTCTTAA-5′ (oligo 48, SEQ48)

The two adaptors, along with the 1.315 kbp DdeI-EcoRV thymidine phosphorylase DNA sequence from plasmid pDTP1, were ligated into the Escherichia coli expression vector pKK223-3 (Pharmacia) digested with EcoRI, to generate plasmid pDTP3 (FIG. 6 ).

To facilitate expression of recombinant thymidine phosphorylase in mammalian cells, the 1.34 kbp EcoRI synthetic thymidine phosphorylase DNA sequence from plasmid pDTP3 was purified and ligated into the EcoRI site of the mammalian expression vector pcDNAI (Invitrogen Corporation), linearised with EcoRI. The resultant plasmid, pDTP4 (FIG. 7 ), once transfected into mammalian cells, for example CHO or COS-7, directs the expression of recombinant thymidine phosphorylase from the CMV promoter.

EXAMPLE 2

Expression in Yeast

Saccharomyces cerevisiae can be used as an alternative expression host. This host may offer distinct advantages over mammalian- or Escherichia coli -based heterologous expression systems in that Saccharomyces cerevisiae lacks any thymidine kinase activity (Grivell, A. R. and Jackson, J. F. (1968), J. Gen. Microbiol. 54, 307-317).

Saccharomyces cerevisiae is therefore unable to salvage thymidine and may also lack any thymidine phosphorylase activity. Saccharomyces cerevisiae is a preferred organism for the heterologous expression of thymidine phosphorylase since the heterologous protein would not be contaminated with endogenously produced thymidine phosphorylase. To facilitate the expression of thymidine phosphorylase in the yeast Saccharomyces cerevisiae , HindIII sites were incorporated at the 5′ and 3′ ends of the synthetic thymidine phosphorylase DNA. Two double-stranded oligonucleotide adaptors, oligonucleotide duplexes VIII and IX, were prepared by annealing oligonucleotides 49 and 50, and oligonucleotides 51 and 52 respectively:

Oligonucleotide Duplex VIII

5′-AGCTTAACCTAATTCTAACAAGCAAAGATGTTATTC   -3′ (oligo 49, SEQ49)
3′-    ATTGGATTAAGATTGTTCGTTTCTACAATAAGAAT-5′ (oligo 50, SEQ50)

Oligonucleotide Duplex IX

5′-ATCTGAATAAA    -3′ (oligo 51, SEQ51)
3′-TAGACTTATTTTCGA-5′ oligo 52, SEQ52)

These two adaptors and the 1.315 kbp DdeI-EcoRV synthetic thymidine phosphorylase DNA sequence from plasmid pDTPI were ligated into the Saccharomyces cerevisiae expression vector pDBP6 (EP 424 117), linearised with HindIII, to generate plasmid pDTP5, FIG. 8 . The expression vector pAYE335 was constructed as follows. A 1.434 kbp HindIII-EcoRI DNA fragment containing the protease B promoter was cloned into the polylinker of the M13 bacteriophage mp18 (Yanish-Perron et al (1985) Gene 33, 103-119), generating plasmid pAYE333 FIG. 9 . Plasmid pAYE333 was linearised by partial digestion with SnaBI and the double stranded oligonucleotide duplex X inserted by ligation at the SnaBI site within the PRB1 promoter.

Oligonucleotide Duplex X

5′-GCGGCCGC-3′ oligonucleotide 56
3′-CGCCGGCG-5′ oligonucleotide 57

Oligonucleotide sequences 56 and 57 are the same. This generates a NotI restriction site at the 5′ end of the protease B promoter. The promoter element was further modified by site directed mutagenesis (oligonucleotide direct in vitro mutagenesis system-Version 2, Amersham) according to the manufacturer's instructions. Mutagenesis with the 31-mer oligonucleotide

5′-CGCCAATAAAAAAACAAGCTTAACCTAATTC-3′ (oligo 58,
                                 SEQ53)

introduces a HindIII restriction site close to the ATG translation initiation codon:

CGCCAATAAAAAAACAAACTAAACCTAATTCTAACAAGCAAAGATG
unmodified | |         Met
(oligo 59, SEQ54)| |
* *
CGCCAATAAAAAAACAAGCTTAACCTAATTCTAACAAGCAAAGATG
modified  |                                                                                                                                                |          Met
(oligo 60, SEQ55) HindIII

Plasmid pAAH5 (Goodey et al 1987: In Yeast Biotechnology, 401-429, Edited by Berry, D. R., Russell, I, and Stewart, G. G. Published by Allen and Unwin) was linearised by partially digesting with BamHI. The 5′ protruding ends were blunt-ended with T4 DNA polymerase and dNTPs and ligated with the double-stranded oligonucleotide duplex X. A recombinant plasmid pAYE334 ( FIG. 10 ) was selected in which a NotI restriction site had replaced the BamHI site at the 3′ end of the ADHI terminator.

Plasmid pAT153 (Twigg & Sherratt (1980) Nature 283, 216-218) was digested with EcoRI/BamHI and the larger 3.36 kbp DNA fragment purified. The 5′ protruding ends were blunt-ended with T4 DNA polymerase and dNTPs and recircularised with the double-stranded oligonucleotide duplex X, generating plasmid pAYE328 (FIG. 11 ).

The 0.8 kbp NotI-HindIII modified protease B promoter sequence was placed upstream of the 0.45 kbp HindIII-NotI ADHI transcription terminator on the pAT153 based plasmid pAYE328 to generate pAYE335 (FIG. 12 ).

The large 6.38 kbp HindIII-BamHI fragment from the yeast E. coli shuttle vector pJDB207 (Beggs, J. D. 1981 Molecular Genetics in Yeast , Alfred Benzon Symposium 16, 383-395) was treated with the Klenow fragment of E. coli DNA polymerase to create flush ends and ligated with the double stranded oligonucleotide duplex X to generate plasmid pDBP5 (FIG. 13 ). The 1.25 kbp NotI Protease B promoter/ADH1 terminator cassette from plasmid pAYE335 ( FIG. 12 ) was introduced into the unique NotI site of plasmid pDBP5 generating pDBP6 (FIG. 14 ).

Transcription initiation and termination sequences are provided by the PRB1 promoter and ADH1 terminator respectively. The thymidine phosphorylase expression plasmid was introduced into the Saccharomyces cerevisiae strain DS569 (MATa, leu2) pSAC3 (EP 424 117), by the method described by Beggs, J. D. (1978) Nature 275, 104-109. Transformants were selected on a minimal medium lacking leucine (0.15% (w/v) yeast nitrogen base without amino acids and ammonium sulfate (Difco), 5% (w/v) ammonium sulfate, 0.1M citric acid/Na 2 HP 4 , 12H 2 O pH6.5, 2% (w/v) sucrose). Transformants were grown for 72 hours at 30° C., 200 rpm in 2000 ml flasks containing either 1000 ml of complex (YEP, 1% (w/v) yeast extract, 2% (w/v) bactopeptone and 2% (w/v) sucrose), or defined (0.15% (w/v) yeast nitrogen base without amino acids and ammonium sulfate (Difco), 0.5% (w/v) ammonium sulfate, 0.1M citric acid/Na 2 HPO 4 12H 2 O pH6.5, 2% (w/v) sucrose) liquid medium.

Thymidine phosphorylase was purified from Escherichia coli as described by Cook, W. et al (1987) J. Biol. Chem. 262, 3788-3789. Thymidine phosphorylase was purified from mammalian cells by disrupting the cells as described previously (Desgranges, C. et al (1981) Biochim. Biophys. Acta. 654, 211-218), and the enzyme purified as described by Cook et al (1987) J. Biol. Chem. 262, 3788-3789. Thymidine phosphorylase was purified from Saccharomyces cerevisiae by centrifuging the culture and resuspending the culture in an equal volume per weight of 50 mM Tris/HCl pH7.6. Cells were lysed by vortexing after the addition of glass beads (40 mesh). The soluble proteins were harvested by centrifugation and the thymidine phosphorylase purified as described by Cook et al (1987) J. Biol. Chem. 262, 3788-3789.

The thymidine phosphorylase activity of the purified E. coli enzyme was determined spectrophotometrically, relying on the thousand-fold difference in the molar extinction co-efficients of thymidine and thymine. Saturating amounts of thymidine (1 mM) and KH 2 PO 4 (0.2M), pH 7.4 were combined in a suitable 1 ml reaction tube with 0.1-5 μg/ml E. coli thymidine phosphorylase or 1-50 μg/ml human thymidine phosphorylase (purified according to Desgranges et al., 1981), respectively, and the reaction at 25° C. was followed by monitoring the decrease in absorbance at 290 nm. A decrease in absorbance of 1 corresponds to the conversion of a 1 mM solution of thymidine to thymine, thus the decrease in absorbance can be used to calculate the V max of the enzyme (μmoles min −1 mg −1 ). The results are given in the following table:

Enzyme                           V                                                                        max                                                                     (μmoles min                                                                        −1                                                                     mg                                                                        −1                                                                    )
E. coli                                                                     thymidine phosphorylase 210
Human thymidine phosphorylase    10

Thus, it can be seen that the E. coli thymidine phosphorylase possesses a V max at least 20 fold higher than the human exzyme.

EXAMPLE 3

Expression of Three Thymidine Phosphorylase Domains

Two double-stranded oligonucleotide adaptors, oligonucleotide duplexes XI and XII, were prepared by annealing oligonucleotides 61 and 62, and oligonucleotides 63 and 64 respectively:

Oligonucleotide Duplex XI

Oligonucleotide duplex XI:
5′-AGCTTAACCTAATTCTAACAAGCAAAGATGGGTCCAATTG     -3′ (Oligo 61, SEQ56)
3′-     ATTGGATTAAGATTGTTCGTTTCTACCCAGGTTAACAGCT-5′ (Oligo 62, SEQ57)
Oligonucleotide duplex XII:
5′-  CGTTGGTGTTGCTAACGGTGCTGGTGTTAGAACTACTGCTTTATTAACTGATTAAA    -3′ (Oligo 63, SEQ58)
3′-TAGCAACCACAACGATTGCCACGACCACAATCTTGATGACGAAATAATTGACTAATTTTCGA-5′ (Oligo 54, SEQ59)

To facilitate expression of a domain comprising amino acids 1-241, oligonucleotide duplexes VIII and XII and the 0.66 kbp DdeI-PvuI synthetic thymidine phosphorylase DNA sequence from plasmid pDTP1 were ligated into the Saccharomyces cerevisiae expression vector pDBP6 (EP 424 117), linearised with HindIII, to generate plasmid pDTP6 (FIG. 15 ). To facilitate expression of a domain comprising amino acids 79-241 oligonucleotide duplexes XI and XII, and the 0.42 kbp SalI-PvuI synthetic thymidine phosphorylase DNA sequence from plasmid pDTP1 were likewise ligated into the Saccharomyces cerevisiae expression vector pDB6 to generate plasmid pDTP7 (FIG. 16 ). To facilitate expression of a domain comprising amino acids 79-440, oligonucleotide duplexes XI and IX and the 1.07 kbp SalI-EcoRV synthetic thymidine phosphorylase DNA sequence from plasmid pDTP1 were likewise ligated into vector pDB6 to generate plasmid pDTP8 (FIG. 17 ). Transcription initiation and termination sequences are provided by the PRB1 promoter and ADH1 terminator respectively. Plasmids pDTP6, pDTP7 and pDTP8 were independently introduced into the Saccharomyces cerevisiae strain DS569 and cultured as described above.

EXAMPLE 4

Cloning of Thymidine Phosphorylase From Other Organisms

This can be achieved by designing degenerate single-stranded oligonucleotides based upon the Escherichia coli thymidine phosphorylase amino acid sequence. The preferred regions are amino acids 82-91; 110-133 and 171-196, based on the amino acid sequence described in FIG. 1 . Within these regions, degenerate DNA sequences can be generated that will encode the amino acids from this region ( FIGS. 18A to 18 C). From these six degenerate DNA sequences, oligonucleotides can be designed that are greater than 10 base pairs in length and have the lowest degeneracy, for example:

From domain 82-92
5′-GTNGAYAARCAYWS-3′ SEQ 60
3′-CANCTRTTYGTRWS-5′ SEQ 61
From domain 110-133
5′-CCNATGATHWSNGG-3′ SEQ 62
3′-GGNTACTADWSNCC-5′ SEQ 63
From domain 171-196
5′-MGNGAYATHACNGC-3′ SEQ 64
3′-KCNCTRTADTGNCG-5′ SEQ 65

These single-stranded DNA primers can be used to amplify regions of the thymidine phosphorylase gene from a number of different species by the polymerase chain reaction (PCR). If the organism is prokaryotic, the PCR amplification can be performed on the genomic DNA according to the manufacturer's instruction (GeneAmp, Perkin Elmer Cetus). However, if the organism is eukaryotic, then the PCR amplification should preferably be performed on total RNA or mRNA from the desired species. This can be achieved by reverse transcribing the mRNA into cDNA/RNA hybrid. This can then be used as a template for PCR amplification according to the manufacturer's instructions (RNA PCR, Perkin Elmer Cetus).

The primers (SEQ60-SEQ65) should be used in pairs, SEQ60 with SEQ63 or 65; SEQ62 with SEQ61 or 65; SEQ64 with SEQ61 or 63; SEQ61 with SEQ62 or 64; SEQ63 with SEQ60 or 64; and SEQ65 with SEQ60 or 62. The PCR amplification cycle conditions should be set to maximise the amplification of the desired product. For example, 94° C. for 1 min; 37° C. for 2 min, 74° C. for 3 min, 40 cycles of amplification, extending the 74° C. incubation by 10 seconds every cycle. Alternatively, to reduce spurious priming during gene amplification, touchdown PCR methodologies can be employed, as described in Don et al (1991) Nucleic Acids Research 19, 4008. Briefly, this procedure employs one relatively stringent annealing temperature (eg 55° C.) which is reduced by 1° C. per cycle for the first 10 cycles of amplification. The final annealing temperature would then be 45° C. This annealing temperature is then maintained for another 20 to 25 cycles while the amplification portion of the cycle (72° C. for 3 min) is extended by 10 seconds per cycle.

The PCR amplification product can be blunt-ended with the Klenow fragment of Escherichia coli DNA polymerase and cloned into the SmaI site of pUC19. E. coli can be transformed to ampicillin resistance with the PCR/pUC19 ligation and transformants containing the thymidine phosphorylase DNA or cDNA sequence identified by DNA sequencing. Such a sequence can be used as a hybridization probe to identify and clone the full length thymidine phosphorylase gene and/or cDNA from appropriate libraries. The full length thymidine phosphorylase gene or cDNA can then be tailored for expression from any one of the expression vectors described in Example 1.

A full length thymidine phosphorylase DNA clone (cDNA or genomic) can also be identified by standard hybridization techniques using 5′ 32 P end labelled oligonucleotides. These oligonucleotides can be the sequences described earlier, namely SEQ60-65. These are firstly end labelled with [γ- 32 P]ATP and T4 polynucleotide kinase and then used as probes to identify potential thymidine phosphorylase genomic or cDNA clones from within appropriate genomic/cDNA libraries prepared in phage or plasmid vectors. Once identified, the thymidine phosphorylase gene or cDNA can be tailored for expression as before.

EXAMPLE 5

Thymidine Phosphorylase Stimulates Endothelial Cell Proliferation

HUVE cells were supplied by the American Type Culture Collection (ATCC number CRL1730). Reagents were supplied by Sigma Chemical Co Ltd unless otherwise stated. Routine HUVE cell cultivation was performed in M199 containing Earle's salts and sodium bicarbonate, 20% (v/v) Foetal Bovine Serum (FBS), 2 mM L-glutamine, 90 μg/ml endothelial cell growth supplement and 100 μg/ml heparin. HUVE cells were maintained in tissue culture flasks (Falcon) coated with type I collagen from calf skin, cross-linked with 1-cyclohexyl-3-(2-morpholinoethyl)-carbodiimide-metho-p-toluene sulfonate (Macklis et al (1985) In Vitro 21, 189-194). HUVE cells were grown at 36.5° C., 5% CO 2 in a LEEC humidified incubator.

HUVE cells were subcultured at a ratio of 1:3 whenever confluent, cells being disassociated using a 0.5 gl −1 trypsin, 0.2 gl −1 EDTA solution in Ca 2+ /Mg 2+ free Dulbecco's phosphate buffered saline (DPBS) (0.2 gl −1 KCl, 0.2 gl −1 KH 2 PO 4 , 8 gl −1 NaCl, 1.15 gl −1 Na 2 HPO 4 ). There was no evidence of transformation or loss of the endothelial morphology under these conditions.

A confluent HUVE cell monolayer was trypsinized, washed and resuspended in M199 (including 2 mM L-glutamine and 20% (v/v) FBS). Duplicate haemocytometer counts were made and the cells were diluted to 4×10 4 cells/ml M199 (containing 2 mM L-glutamine and 20% (v/v) FBS). A collagen-coated cluster plate (24 flat bottomed 16 mm diameter wells, Costar) was seeded with approximately 2×10 4 cells/well (0.5 ml/well) and incubated for 8 hours at 36.5° C. in an atmosphere which included 5% CO 2 to allow cell attachment and growth factor depletion. Escherichia coli thymidine phosphorylase was added to the wells and cells were incubated for another 18 hours. Sterile [6- 3 H]-thymidine (1.0 μCi/well, 29 Ci/mmol, Amersham International plc) diluted to 10 μl with DPBS was added and the cells were incubated for a further 4 hours. The medium was discarded, the cells washed gently with 3 ml DPBS and the DNA was fixed with 0.5 ml 5% w/v trichloroacetic acid (20 min, ice cold). The trichloroacetic acid was discarded, the wells were washed with 3 ml Milli-Q water and the DNA was solubilised with 0.3 ml 1M sodium hydroxide (20 min, room temperature, gentle agitation), neutralised with 0.3 ml 1M hydrochloric acid and then transferred to vials containing 10 ml scintillation fluid (Aqualuma, Lumac). The tritium radioactivity was measured by liquid scintillation counting (10 min/vial) using a Packard Tri-Carb 1500 liquid scintillation analyser (FIG. 19 ).

Thymidine phosphorylase stimulates [ 3 H]-thymidine incorporation by HUVE cells. A bell-shaped dose-response curve was derived, with maximal stimulation being observed around 40 ng/ml. Thymidine phosphorylase is therefore mitogenic towards endothelial cells.

The effect of E. coli thymidine phosphorylase addition on endothelial cell proliferation was determined over an eight day period by the acid phosphatase assay for endothelial cell number, Connolly et al (1986) Anal. Biochem. 152, 136-140. CPAE cells were maintained as previously described by Finnis et al (1992) Yeast 8, 57-60. Cluster plates (24 flat bottomed, 16 mm diameter wells, Costar) were seeded with CPAE cells (10 4 per well) in 0.5 ml minimal essential medium containing 2% heat inactivated (65° C., 30 min) dialysed (1 kDa cut-off) FBS, 100 μM thymidine, non-essential amino acids and antibiotics. E. coli thymidine phosphorylase in 10 μl DPBS was added to each well to achieve a final concentration of 100 ng/ml. Cells were grown at 36.5° C. in an atmosphere which included 5% CO 2 in a humidified incubator. To determine endothelial cell number, the culture medium was discarded and the cells were washed with 1 ml DPBS. Cells were then incubated for 2 hours with a medium containing 0.5 ml 10 mM p-nitrophenyl phosphate (SIGMA 104), 0.1% (v/v) Triton X-100 and 0.1M sodium acetate pH5.5, and the reaction was stopped by the addition of 50 μl 1M sodium hydroxide. The absorbance at 405 nm minus the absorbance at 620 nm (A 405 -A 620 ) was determined for each well in the absence of cells and subtracted from the A 405 -A 620 in the presence of cells (FIG. 20 ). In the absence of thymidine phosphorylase, cell number, as indicated by the corrected absorbance, declines. However, in the presence of 100 ng/ml thymidine phosphorylase, the cell number increases.

EXAMPLE 6

Expression of a Fusion of Thymidine Phosphorylase and Antibody Region

Thymidine phosphorylase, or a functional portion thereof, can be expressed as a fusion protein with a functional variable-region of a monoclonal antibody which will direct the thymidine phosphorylase to the chosen location. Total RNA is prepared from 1×10 8 hybridoma cells expressing the desired monoclonal antibody. Total RNA was used for first strand cDNA synthesis using Moloney murine leukaemia virus reverse transcriptase and random primers at 37° C. for 1 hour as described by Chaudhary, V. et al (1990) PNAS 87, 1066-1070. The variable light (V L ) and (V H ) regions of the antibody cDNA were amplified by polymerase chain reaction (PCR) using primers (V L -5′, V L -3′, V H -5′ and V H -3′) and conditions described by Chaudhary, V. et al (1990, see above). The V L and V H cDNA segments, when ligated in the order 5′-V L -V H -3′, are linked by a short peptide linker region. By using appropriate double-stranded oligonucleotide linkers, this functional antibody variable region can be cloned upstream of the synthetic thymidine phosphorylase described in FIGS. 1 and 4 to 4 E. The DNA insert encoding the V L -V H -thymidine fusion protein was inserted into the HindIII site of pDBP6 ( FIG. 14 ) with the aid of suitable double-stranded oligonucleotide linkers. The V L -V H -thymidine phosphorylase fusion protein can then be expressed from the S. cerevisiae PRB1 promoter unless the expression vector is used to transform a suitable host strain, such as DS569 (MATa, leu2) pSAC3 (EP 424 117). Transformants were selected and cultured as described in Example 2.

EXAMPLE 7

Conjugation

This example illustrates how thymidine phosphorylase, or a functional portion thereof, can be conjugated to antibodies by various coupling/cross-linking reagents. This methodology equally applies to the coupling of thymidine phosphorylase to any other protein which would act as a ligand to target thymidine phosphorylase.

1 mg of antibody in 0.5 ml 30 mM HEPES pH7.4 was reacted with 50 μl sulfo-SMCC at 50 mg/ml in 30 mM HEPES pH7.4 (Pierce) for 30 min at 4° C. The sulfhydryl reactive antibody was isolated by dialysis into 30 mM HEPES pH7.4. 1 mg thymidine phosphorylase in 0.5 ml in degassed 30 mM HEPES pH7.4, 1 mM EDTA, was then reacted with the activated antibody at 4° C. for 5 min. The thymidine phosphorylase antibody conjugate was isolated by gel filtration.

This same procedure can be employed to cross-link thymidine phosphorylase and antibodies if sulfo-MBS is employed as the cross-linking group instead of sulfo-SMCC.

SMCC or MBS (Pierce) can also be used as the cross-linking group instead of sulfo-SMCC or sulfo-MBS, in which case the SMCC or MBS should be dissolved in DMSO before use.

Thus, 1.5 mg MBS or SMCC was dissolved in 50 μl DMSO. 0.5 mg of thymidine phosphorylase and 0.5 mg of the antibody were added to phosphate buffered saline pH6.0 to a final concentration of 1 mg/ml and mixed in a 10 ml conical flask at 22° C. The MBS or SMCC solution was added to the protein solution and shaken for a further 30 minutes at 22° C. The derivatized protein was desalted by FPLC using a G25 superfine column into PBS pH7.4. Peak fractions were pooled.

Where necessary, free sulfhydryl groups on the antibody can be blocked with N-ethyl-maleimide and primary amines on the thymidine phosphorylase moiety can be modified by Trant's Reagent (Pierce) to introduce sulfhydryl groups allowing coupling to the activated antibody. Using the same cross-linking the thymidine phosphorylase moiety first as described above and then cross-link the sulfhydryl groups on the targeting ligand, for example the antibody.

Thymidine phosphorylase or a functional portion thereof can also be conjugated to synthetic carbohydrates with the Lewis-X or sialyl Lewis-X determinant, FIGS. 21A and 21B , or purified natural carbohydrates containing this determinant, eg LNFIII (Calbiochem) by procedures involving chemical spacers like p-aminophenyl, aminophenylethyl and acetyl phenylenediamine as described by Berg, E. et al (1991) J. Biol. Chem. 266, 14869-14872.

EXAMPLE 8

Formulation of an Agent with Thymidine Phosphorylase Activity

Thymidine phosphorylase prepared as above was made up as a 0.01% w/v solution in water-for-injection containing 4.5% w/v human albumin, for injection into a vein adjacent an ulcer or other target site. Alternatively, the thymidine phosphorylase, made up as a 0.01% (w/v) solution in water-for-injection, can be applied to the target site in the form of an aerosol.

EXAMPLE 9

Bioactive Graft

Thymidine phosphorylase was cross-linked to Dacron vascular prosthetic grafts and human-albumin-coated Dacron grafts, (Hake, V. et al (1991) Thorac. Cardiovasc. Surgeon 39, 208-213). This procedure enhances the rate of endothelial cell growth and hence reduces the time taken to form a confluent endothelium. Thymidine phosphorylase at 2 mg/ml in 30 mM HEPES pH7.4 was reacted with sulfo-SMCC 50 mg/ml in 30 mM HEPES pH7.4 for 30 min at 4° C. in the presence of the vascular prostheses. The thymidine phosphorylase conjugated vascular prostheses were then implanted as required.

76 440 amino acids amino acid linear protein NO Escherichia coli K12 Protein 1..440 /note= “Figure 1” 1Leu Phe Leu Ala Gln Glu Ile Ile Arg Lys Lys Arg Asp Gly His Ala1 5 10 15Leu Ser Asp Glu Glu Ile Arg Phe Phe Ile Asn Gly Ile Arg Asp Asn 20 25 30Thr Ile Ser Glu Gly Gln Ile Ala Ala Leu Ala Met Thr Ile Phe Phe 35 40 45His Asp Met Thr Met Pro Glu Arg Val Ser Leu Thr Met Ala Met Arg 50 55 60Asp Ser Gly Thr Val Leu Asp Trp Lys Ser Leu His Leu Asn Gly Pro65 70 75 80Ile Val Asp Lys His Ser Thr Gly Gly Val Gly Asp Val Thr Ser Leu 85 90 95Met Leu Gly Pro Met Val Ala Ala Cys Gly Gly Tyr Ile Pro Met Ile 100 105 110Ser Gly Arg Gly Leu Gly His Thr Gly Gly Thr Leu Asp Lys Leu Glu 115 120 125Ser Ile Pro Gly Phe Asp Ile Phe Pro Asp Asp Asn Arg Phe Arg Glu 130 135 140Ile Ile Lys Asp Val Gly Val Ala Ile Ile Gly Gln Thr Ser Ser Leu145 150 155 160Ala Pro Ala Asp Lys Arg Phe Tyr Ala Thr Arg Asp Ile Thr Ala Thr 165 170 175Val Asp Ser Ile Pro Leu Ile Thr Ala Ser Ile Leu Ala Lys Lys Leu 180 185 190Ala Glu Gly Leu Asp Ala Leu Val Met Asp Val Lys Val Gly Ser Gly 195 200 205Ala Phe Met Pro Thr Tyr Glu Leu Ser Glu Ala Leu Ala Glu Ala Ile 210 215 220Val Gly Val Ala Asn Gly Ala Gly Val Arg Thr Thr Ala Leu Leu Thr225 230 235 240Asp Met Asn Gln Val Leu Ala Ser Ser Ala Gly Asn Ala Val Glu Val 245 250 255Arg Glu Ala Val Gln Phe Leu Thr Gly Glu Tyr Arg Asn Pro Arg Leu 260 265 270Phe Asp Val Thr Met Ala Leu Cys Val Glu Met Leu Ile Ser Gly Lys 275 280 285Leu Ala Lys Asp Asp Ala Glu Ala Arg Ala Lys Leu Gln Ala Val Leu 290 295 300Asp Asn Gly Lys Ala Ala Glu Val Phe Gly Arg Met Val Ala Ala Gln305 310 315 320Lys Gly Pro Thr Asp Phe Val Glu Asn Tyr Ala Lys Tyr Leu Pro Thr 325 330 335Ala Met Leu Thr Lys Ala Val Tyr Ala Asp Thr Glu Gly Phe Val Ser 340 345 350Glu Met Asp Thr Arg Ala Leu Gly Met Ala Val Val Ala Met Gly Gly 355 360 365Gly Arg Arg Gln Ala Ser Asp Thr Ile Asp Tyr Ser Val Gly Phe Thr 370 375 380Asp Met Ala Arg Leu Gly Asp Gln Val Asp Gly Gln Arg Pro Leu Ala385 390 395 400Val Ile His Ala Lys Asp Glu Asn Asn Trp Gln Glu Ala Ala Lys Ala 405 410 415Val Lys Ala Ala Ile Lys Leu Ala Asp Lys Ala Pro Glu Ser Thr Pro 420 425 430Thr Val Tyr Arg Arg Ile Ser Glu 435 440 71 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..71 /function= “oligonucleotide 1” 2CTTAGCTCAA GAAATTATTA GAAAAAAAAG AGATGGTCAT GCTTTATCTG ATGAAGAAAT 60TAGATTCTTC A 71 71 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..71 /function= “oligonucleotide 2” 3TTAACGGTAT TAGAGATAAC ACTATTTCTG AAGGTCAAAT TGCTGCTTTA GCTATGACTA 60TTTTCTTCCA T 71 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 3” 4GATATGACTA TGCCAGAAAG AGTTTCTTTA ACTATGGCTA TGAGAGATTC TGGTACTGTT 60TTAGATTGGA 70 27 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..27 /function= “oligonucleotide 4” 5AATCTTTACA TTTAAACGGT CCAATTG 27 42 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..42 /function= “oligonucleotide 5” 6TCGACAATTG GACCGTTTAA ATGTAAAGAT TTCCAATCTA AA 42 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 6” 7ACAGTACCAG AATCTCTCAT AGCCATAGTT AAAGAAACTC TTTCTGGCAT AGTCATATCA 60TGGAAGAAAA 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 7” 8TAGTCATAGC TAAAGCAGCA ATTTGACCTT CAGAAATAGT GTTATCTCTA ATACCGTTAA 60TGAAGAATCT 70 61 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..61 /function= “oligonucleotide 8” 9AATTTCTTCA TCAGATAAAG CATGACCATC TCTTTTTTTT CTAATAATTT CTTGAGCTAA 60G 61 73 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..73 /function= “oligonucleotide 9” 10TCGACAAACA TTCTACTGGT GGTGTTGGTG ATGTTACTTC TTTAATGTTA GGTCCAATGG 60TTGCTGCTTG TGG 73 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 10” 11TGGTTACATT CCAATGATTT CTGGTAGAGG TTTAGGTCAT ACTGGTGGTA CTTTAGATAA 60ATTAGAATCT 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 11” 12ATTCCAGGTT TCGATATTTT CCCAGATGAT AACAGATTCA GAGAAATTAT TAAAGATGTT 60GGTGTTGCTA 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 12” 13TTATTGGTCA AACTTCTTCT TTAGCTCCAG CTGATAAAAG ATTCTACGCT ACTAGAGATA 60TTACTGCTAC 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 13” 14TGTTGATTCT ATTCCATTAA TTACTGCTTC TATTTTAGCT AAAAAATTAG CTGAAGGTTT 60AGATGCTTTA 70 74 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..74 /function= “oligonucleotide 14” 15GTTATGGATG TTAAAGTTGG TTCTGGTGCT TTCATGCCAA CTTACGAATT ATCTGAAGCC 60TTGGCTGAAG CGAT 74 62 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..62 /function= “oligonucleotide 15” 16CGCTTCAGCC AAGGCTTCAG ATAATTCGTA AGTTGGCATG AAAGCACCAG AACCAACTTT 60AA 62 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 16” 17CATCCATAAC TAAAGCATCT AAACCTTCAG CTAATTTTTT AGCTAAAATA GAAGCAGTAA 60TTAATGGAAT 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 17” 18AGAATCAACA GTAGCAGTAA TATCTCTAGT AGCGTAGAAT CTTTTATCAG CTGGAGCTAA 60AGAAGAAGTT 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 18” 19TGACCAATAA TAGCAACACC AACATCTTTA ATAATTTCTC TGAATCTGTT ATCATCTGGG 60AAAATATCGA 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 19” 20AACCTGGAAT AGATTCTAAT TTATCTAAAG TACCACCAGT ATGACCTAAA CCTCTACCAG 60AAATCATTGG 70 79 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..79 /function= “oligonucleotide 20” 21AATGTAACCA CCACAAGCAG CAACCATTGG ACCTAACATT AAAGAAGTAA CATCACCAAC 60ACCACCAGTA GAATGTTTG 79 66 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..66 /function= “oligonucleotide 21” 22CGTTGGTGTT GCTAACGGTG CTGGTGTTAG AACTACTGCT TTATTAACTG ATATGAACCA 60AGTTTT 66 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 22” 23AGCTTCTTCT GCTGGTAACG CTGTTGAAGT TAGAGAAGCT GTTCAATTCT TAACTGGTGA 60ATACAGAAAC 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 23” 24CCAAGATTAT TCGATGTTAC TATGGCTTTA TGTGTTGAAA TGTTAATTTC TGGTAAATTA 60GCTAAAGATG 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 24” 25ATGCTGAAGC TAGAGCTAAA TTACAAGCTG TTTTAGATAA CGGTAAAGCT GCTGAAGTTT 60TCGGTAGAAT 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 25” 26GGTTGCTGCT CAAAAAGGTC CAACTGATTT CGTTGAAAAC TACGCTAAAT ACTTACCAAC 60TGCTATGTTA 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 26” 27ACTAAAGCTG TTTACGCTGA TACTGAAGGT TTCGTTTCTG AAATGGATAC TAGAGCTTTA 60GGTATGGCTG 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 27” 28TTGTTGCTAT GGGTGGTGGT AGAAGACAAG CCTCTGATAC TATTGATTAC TCTGTTGGTT 60TCACTGATAT 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 28” 29GGCTAGATTA GGTGATCAAG TTGATGGTCA AAGACCATTA GCTGTTATTC ATGCTAAAGA 60TGAAAACAAC 70 60 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..60 /function= “oligonucleotide 29” 30TGGCAAGAAG CTGCTAAAGC TGTTAAAGCT GCTATTAAAT TAGCTGATAA AGCTCCAGAA 60 29 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..29 /function= “oligonucleotide 30” 31TCTACTCCAA CTGTTTACAG AAGGATATC 29 40 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..40 /function= “oligonucleotide 31” 32GATATCCTTC TGTAAACAGT TGGAGTAGAT TCTGGAGCTT 40 60 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..60 /function= “oligonucleotide 32” 33TATCAGCTAA TTTAATAGCA GCTTTAACAG CTTTAGCAGC TTCTTGCCAG TTGTTTTCAT 60 69 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..69 /function= “oligonucleotide 33” 34CTTTAGCATG AATAACAGCT AATGGTCTTT GACCATCAAC TTGATCACCT AATCTAGCCA 60TATCAGTGA 69 71 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..71 /function= “oligonucleotide 34” 35AACCAACAGA GTAATCAATA GTATCAGAGG CTTGTCTTCT ACCACCACCC ATAGCAACAA 60CAGCCATACC T 71 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 35” 36AAAGCTCTAG TATCCATTTC AGAAACGAAA CCTTCAGTAT CAGCGTAAAC AGCTTTAGTT 60AACATAGCAG 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 36” 37TTGGTAAGTA TTTAGCGTAG TTTTCAACGA AATCAGTTGG ACCTTTTTGA GCAGCAACCA 60TTCTACCGAA 70 70 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..70 /function= “oligonucleotide 37” 38AACTTCAGCA GCTTTACCGT TATCTAAAAC AGCTTGTAAT TTAGCTCTAG CTTCAGCATC 60ATCTTTAGCT 70 69 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..69 /function= “oligonucleotide 38” 39AATTTACCAG AAATTAACAT TTCAACACAT AAAGCCATAG TAACATCGAA TAATCTTGGG 60TTTCTGTAT 69 71 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..71 /function= “oligonucleotide 39” 40TCACCAGTTA AGAATTGAAC AGCTTCTCTA ACTTCAACAG CGTTACCAGC AGAAGAAGCT 60AAAACTTGGT T 71 57 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..57 /function= “oligonucleotide 40” 41CATATCAGTT AATAAAGCAG TAGTTCTAAC ACCAGCACCG TTAGCAACAC CAACGAT 57 10 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..10 /function= “oligonucleotide 41” 42CATGTTATTC 10 11 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..11 /function= “oligonucleotide 43” 43ATCTGAATAA C 11 15 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..15 /function= “oligonucleotide 44” 44CATGGTTATT CAGAT 15 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “oligonucleotide 45” 45AATTCATGTT ATTC 14 13 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..13 /function= “oligonucleotide 46” 46TAAGAATAAC ATG 13 11 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..11 /function= “oligonucleotide 47” 47ATCTGAATAA G 11 15 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..15 /function= “oligonucleotide 48” 48AATTCTTATT CAGAT 15 36 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..36 /function= “oligonucleotide 49” 49AGCTTAACCT AATTCTAACA AGCAAAGATG TTATTC 36 35 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..35 /function= “oligonucleotide 50” 50TAAGAATAAC ATCTTTGCTT GTTAGAATTA GGTTA 35 11 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..11 /function= “oligonucleotide 51” 51ATCTGAATAA A 11 15 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..15 /function= “oligonucleotide 52” 52AGCTTTTATT CAGAT 15 31 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..31 /function= “oligonucleotide 58” 53CGCCAATAAA AAAACAAGCT TAACCTAATT C 31 46 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..46 /function= “oligonucleotide 59” 54CGCCAATAAA AAAACAAACT AAACCTAATT CTAACAAGCA AAGATG 46 46 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..46 /function= “oligonucleotide 60” 55CGCCAATAAA AAAACAAGCT TAACCTAATT CTAACAAGCA AAGATG 46 40 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..40 /function= “oligonucleotide 61” 56AGCTTAACCT AATTCTAACA AGCAAAGATG GGTCCAATTG 40 40 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..40 /function= “oligonucleotide 62” 57TCGACAATTG GACCCATCTT TGCTTGTTAG AATTAGGTTA 40 56 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..56 /function= “oligonucleotide 63” 58CGTTGGTGTT GCTAACGGTG CTGGTGTTAG AACTACTGCT TTATTAACTG ATTAAA 56 62 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..62 /function= “oligonucleotide 64” 59AGCTTTTAAT CAGTTAATAA AGCAGTAGTT CTAACACCAG CACCGTTAGC AACACCAACG 60AT 62 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 70” 60GTNGAYAARC AYWS 14 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 71” 61SWRTGYTTRT CNAC 14 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 72” 62CCNATGATHW SNGG 14 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 73” 63CCNSWDATCA TNGG 14 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 74” 64MGNGAYATHA CNGC 14 14 base pairs nucleic acid double linear cDNA NO NO not provided misc_feature 1..14 /function= “SEQ 75” 65GCNGTDATRT CNCK 14 1311 base pairs nucleic acid double linear DNA (genomic) NO NO not provided misc_feature 1..1311 /function= “Figure 4” CDS 2..1311 66C TTA GCT CAA GAA ATT ATT AGA AAA AAA AGA GAT GGT CAT GCT TTA 46 Leu Ala Gln Glu Ile Ile Arg Lys Lys Arg Asp Gly His Ala Leu 1 5 10 15TCT GAT GAA GAA ATT AGA TTC TTC ATT AAC GGT ATT AGA GAT AAC ACT 94Ser Asp Glu Glu Ile Arg Phe Phe Ile Asn Gly Ile Arg Asp Asn Thr 20 25 30ATT TCT GAA GGT CAA ATT GCT GCT TTA GCT ATG ACT ATT TTC TTC CAT 142Ile Ser Glu Gly Gln Ile Ala Ala Leu Ala Met Thr Ile Phe Phe His 35 40 45GAT ATG ACT ATG CCA GAA AGA GTT TCT TTA ACT ATG GCT ATG AGA GAT 190Asp Met Thr Met Pro Glu Arg Val Ser Leu Thr Met Ala Met Arg Asp 50 55 60TCT GGT ACT GTT TTA GAT TGG AAA TCT TTA CAT TTA AAC GGT CCA ATT 238Ser Gly Thr Val Leu Asp Trp Lys Ser Leu His Leu Asn Gly Pro Ile 65 70 75GTC GAC AAA CAT TCT ACT GGT GGT GTT GGT GAT GTT ACT TCT TTA ATG 286Val Asp Lys His Ser Thr Gly Gly Val Gly Asp Val Thr Ser Leu Met 80 85 90 95TTA GGT CCA ATG GTT GCT GCT TGT GGT GGT TAC ATT CCA ATG ATT TCT 334Leu Gly Pro Met Val Ala Ala Cys Gly Gly Tyr Ile Pro Met Ile Ser 100 105 110GGT AGA GGT TTA GGT CAT ACT GGT GGT ACT TTA GAT AAA TTA GAA TCT 382Gly Arg Gly Leu Gly His Thr Gly Gly Thr Leu Asp Lys Leu Glu Ser 115 120 125ATT CCA GGT TTC GAT ATT TTC CCA GAT GAT AAC AGA TTC AGA GAA ATT 430Ile Pro Gly Phe Asp Ile Phe Pro Asp Asp Asn Arg Phe Arg Glu Ile 130 135 140ATT AAA GAT GTT GGT GTT GCT ATT ATT GGT CAA ACT TCT TCT TTA GCT 478Ile Lys Asp Val Gly Val Ala Ile Ile Gly Gln Thr Ser Ser Leu Ala 145 150 155CCA GCT GAT AAA AGA TTC TAC GCT ACT AGA GAT ATT ACT GCT ACT GTT 526Pro Ala Asp Lys Arg Phe Tyr Ala Thr Arg Asp Ile Thr Ala Thr Val160 165 170 175GAT TCT ATT CCA TTA ATT ACT GCT TCT ATT TTA GCT AAA AAA TTA GCT 574Asp Ser Ile Pro Leu Ile Thr Ala Ser Ile Leu Ala Lys Lys Leu Ala 180 185 190GAA GGT TTA GAT GCT TTA GTT ATG GAT GTT AAA GTT GGT TCT GGT GCT 622Glu Gly Leu Asp Ala Leu Val Met Asp Val Lys Val Gly Ser Gly Ala 195 200 205TTC ATG CCA ACT TAC GAA TTA TCT GAA GCC TTG GCT GAA GCG ATC GTT 670Phe Met Pro Thr Tyr Glu Leu Ser Glu Ala Leu Ala Glu Ala Ile Val 210 215 220GGT GTT GCT AAC GGT GCT GGT GTT AGA ACT ACT GCT TTA TTA ACT GAT 718Gly Val Ala Asn Gly Ala Gly Val Arg Thr Thr Ala Leu Leu Thr Asp 225 230 235ATG AAC CAA GTT TTA GCT TCT TCT GCT GGT AAC GCT GTT GAA GTT AGA 766Met Asn Gln Val Leu Ala Ser Ser Ala Gly Asn Ala Val Glu Val Arg240 245 250 255GAA GCT GTT CAA TTC TTA ACT GGT GAA TAC AGA AAC CCA AGA TTA TTC 814Glu Ala Val Gln Phe Leu Thr Gly Glu Tyr Arg Asn Pro Arg Leu Phe 260 265 270GAT GTT ACT ATG GCT TTA TGT GTT GAA ATG TTA ATT TCT GGT AAA TTA 862Asp Val Thr Met Ala Leu Cys Val Glu Met Leu Ile Ser Gly Lys Leu 275 280 285GCT AAA GAT GAT GCT GAA GCT AGA GCT AAA TTA CAA GCT GTT TTA GAT 910Ala Lys Asp Asp Ala Glu Ala Arg Ala Lys Leu Gln Ala Val Leu Asp 290 295 300AAC GGT AAA GCT GCT GAA GTT TTC GGT AGA ATG GTT GCT GCT CAA AAA 958Asn Gly Lys Ala Ala Glu Val Phe Gly Arg Met Val Ala Ala Gln Lys 305 310 315GGT CCA ACT GAT TTC GTT GAA AAC TAC GCT AAA TAC TTA CCA ACT GCT 1006Gly Pro Thr Asp Phe Val Glu Asn Tyr Ala Lys Tyr Leu Pro Thr Ala320 325 330 335ATG TTA ACT AAA GCT GTT TAC GCT GAT ACT GAA GGT TTC GTT TCT GAA 1054Met Leu Thr Lys Ala Val Tyr Ala Asp Thr Glu Gly Phe Val Ser Glu 340 345 350ATG GAT ACT AGA GCT TTA GGT ATG GCT GTT GTT GCT ATG GGT GGT GGT 1102Met Asp Thr Arg Ala Leu Gly Met Ala Val Val Ala Met Gly Gly Gly 355 360 365AGA AGA CAA GCC TCT GAT ACT ATT GAT TAC TCT GTT GGT TTC ACT GAT 1150Arg Arg Gln Ala Ser Asp Thr Ile Asp Tyr Ser Val Gly Phe Thr Asp 370 375 380ATG GCT AGA TTA GGT GAT CAA GTT GAT GGT CAA AGA CCA TTA GCT GTT 1198Met Ala Arg Leu Gly Asp Gln Val Asp Gly Gln Arg Pro Leu Ala Val 385 390 395ATT CAT GCT AAA GAT GAA AAC AAC TGG CAA GAA GCT GCT AAA GCT GTT 1246Ile His Ala Lys Asp Glu Asn Asn Trp Gln Glu Ala Ala Lys Ala Val400 405 410 415AAA GCT GCT ATT AAA TTA GCT GAT AAA GCT CCA GAA TCT ACT CCA ACT 1294Lys Ala Ala Ile Lys Leu Ala Asp Lys Ala Pro Glu Ser Thr Pro Thr 420 425 430GTT TAC AGA AGG ATA TC 1311Val Tyr Arg Arg Ile 435 436 amino acids amino acid linear protein not provided 67Leu Ala Gln Glu Ile Ile Arg Lys Lys Arg Asp Gly His Ala Leu Ser 1 5 10 15Asp Glu Glu Ile Arg Phe Phe Ile Asn Gly Ile Arg Asp Asn Thr Ile 20 25 30Ser Glu Gly Gln Ile Ala Ala Leu Ala Met Thr Ile Phe Phe His Asp 35 40 45Met Thr Met Pro Glu Arg Val Ser Leu Thr Met Ala Met Arg Asp Ser 50 55 60Gly Thr Val Leu Asp Trp Lys Ser Leu His Leu Asn Gly Pro Ile Val 65 70 75 80Asp Lys His Ser Thr Gly Gly Val Gly Asp Val Thr Ser Leu Met Leu 85 90 95Gly Pro Met Val Ala Ala Cys Gly Gly Tyr Ile Pro Met Ile Ser Gly 100 105 110Arg Gly Leu Gly His Thr Gly Gly Thr Leu Asp Lys Leu Glu Ser Ile 115 120 125Pro Gly Phe Asp Ile Phe Pro Asp Asp Asn Arg Phe Arg Glu Ile Ile 130 135 140Lys Asp Val Gly Val Ala Ile Ile Gly Gln Thr Ser Ser Leu Ala Pro145 150 155 160Ala Asp Lys Arg Phe Tyr Ala Thr Arg Asp Ile Thr Ala Thr Val Asp 165 170 175Ser Ile Pro Leu Ile Thr Ala Ser Ile Leu Ala Lys Lys Leu Ala Glu 180 185 190Gly Leu Asp Ala Leu Val Met Asp Val Lys Val Gly Ser Gly Ala Phe 195 200 205Met Pro Thr Tyr Glu Leu Ser Glu Ala Leu Ala Glu Ala Ile Val Gly 210 215 220Val Ala Asn Gly Ala Gly Val Arg Thr Thr Ala Leu Leu Thr Asp Met225 230 235 240Asn Gln Val Leu Ala Ser Ser Ala Gly Asn Ala Val Glu Val Arg Glu 245 250 255Ala Val Gln Phe Leu Thr Gly Glu Tyr Arg Asn Pro Arg Leu Phe Asp 260 265 270Val Thr Met Ala Leu Cys Val Glu Met Leu Ile Ser Gly Lys Leu Ala 275 280 285Lys Asp Asp Ala Glu Ala Arg Ala Lys Leu Gln Ala Val Leu Asp Asn 290 295 300Gly Lys Ala Ala Glu Val Phe Gly Arg Met Val Ala Ala Gln Lys Gly305 310 315 320Pro Thr Asp Phe Val Glu Asn Tyr Ala Lys Tyr Leu Pro Thr Ala Met 325 330 335Leu Thr Lys Ala Val Tyr Ala Asp Thr Glu Gly Phe Val Ser Glu Met 340 345 350Asp Thr Arg Ala Leu Gly Met Ala Val Val Ala Met Gly Gly Gly Arg 355 360 365Arg Gln Ala Ser Asp Thr Ile Asp Tyr Ser Val Gly Phe Thr Asp Met 370 375 380Ala Arg Leu Gly Asp Gln Val Asp Gly Gln Arg Pro Leu Ala Val Ile385 390 395 400His Ala Lys Asp Glu Asn Asn Trp Gln Glu Ala Ala Lys Ala Val Lys 405 410 415Ala Ala Ile Lys Leu Ala Asp Lys Ala Pro Glu Ser Thr Pro Thr Val 420 425 430Tyr Arg Arg Ile 435 33 base pairs nucleic acid double linear DNA (genomic) NO NO not provided misc_feature 1..33 /function= “Fig.18 - region 82-92” CDS 1..33 68GTNGAYAARC AYWSNACNGG NGGNGTNGGN GAY 33 11 amino acids amino acid single linear peptide not provided Region 1..11 /note= “Fig.18 - Amino acid sequence encoded by region 82-92” 69Val Asp Lys His Ser Thr Gly Gly Val Gly Asp1 5 10 72 base pairs nucleic acid double linear DNA (genomic) NO NO not provided misc_feature 1..72 /function= “Fig.18 - region 110-133” CDS 1..72 70CCNATGATHW SNGGNMGNGG NYTNGGNCAY ACNGGNGGNA CNYTNGAYAA RYTNGARWSN 60ATHCCNGGNT TY 72 24 amino acids amino acid single linear peptide not provided Region 1..24 /note= “Fig.18 - amino acid sequence encoded by region 110-133” 71Pro Met Ile Ser Gly Arg Gly Leu Gly His Thr Gly Gly Thr Leu Asp1 5 10 15Lys Leu Glu Ser Ile Pro Gly Phe 20 78 base pairs nucleic acid double linear DNA (genomic) NO NO not provided misc_feature 1..78 /function= “Fig.18 - region 171-196” CDS 1..78 72MGNGAYATHA CNGCNACNGT NGAYWSNATH CCNYTNATHA CNGCNWSNAT HYTNGCNAAR 60AARYTNGCNG ARGGNYTN 78 26 amino acids amino acid single linear peptide not provided Region 1..26 /note= “Fig.18 - amino acid sequence encoded by region 171-196” 73Arg Asp Ile Thr Ala Thr Val Asp Ser Ile Pro Leu Ile Thr Ala Ser1 5 10 15Ile Leu Ala Lys Lys Leu Ala Glu Gly Leu 20 25 33 base pairs nucleic acid double linear DNA (genomic) NO YES not provided misc_feature 1..33 /function= “Fig.18 - anti-sense to region 82-92” 74RTCNCCNACN CCNCCNGTNS WRTGYTTRTC NAC 33 72 base pairs nucleic acid double linear DNA (genomic) NO YES not provided misc_feature 1..72 /function= “Fig.18 - anti-sense to region 110-133” 75RAANCCNGGD ATNSWYTCNA RYTTRTCNAR NGTNCCNCCN GTRTGNCCNA RNCCNCKNCC 60NSWDATCATN GG 72 78 base pairs nucleic acid double linear DNA (genomic) NO YES not provided misc_feature 1..78 /function= “Fig.18 - anti-sense to region 171-196” 76NARNCCYTCN GCNARYTTYT TNGCNARDAT NSWNGCNGTD ATNARNGGDA TNSWRTCNAC 60NGTNGCNGTD ATRTCNCK 78

Claims

1. A method of modulating cellular proliferation in a mammal in need thereof which comprises administering to said mammal an amount of a pharmaceutical composition effective to modulate cellular proliferation, said composition comprising a pharmaceutically acceptable vehicle and a non-human thymidine phosphorylase polypeptide characterized by having a thymidine phosphorylase activity at least about 5% of native E. coli thymidine phosphorylase, wherein said activity is determined as the Vmax(μmoles min−1mg−1) of the enzyme at 25° C. in the presence of 1 mM thymidine in 0.2M KH2PO4 at pH 7.4.

2. A method in accordance with claim 1, wherein said polypeptide is characterized by having a thymidine phosphorylase activity at least about 50% of native E. coli thymidine phosphorylase.

3. A method in accordance with claim 1, wherein said polypeptide is characterized by having a thymidine phosphorylase activity at least about 90% of native E. coli thymidine phosphorylase.

4. A method in accordance with claim 1, wherein said polypeptide is native E. coli thymidine phosphorylase, or a functional portion thereof.

5. A method in accordance with claim 1, wherein the administration of said polypeptide to said mammal causes healing of a wound by increasing cellular proliferation at said wound.

6. A method in accordance with claim 5, wherein said polypeptide is characterized by having a thymidine phosphorylase activity at least about 50% of native E. coli thymidine phosphorylase.

7. A method in accordance with claim 5, wherein said polypeptide is characterized by having a thymidine phosphorylase activity at least about 90% of native E. coli thymidine phosphorylase.

8. A method in accordance with claim 5, where said polypeptide is applied topically to said wound.

9. A method in accordance with claim 1, wherein said polypeptide is administered as a conjugate with a targeting agent which causes said polypeptide to be accumulated at the site where modulation of cellular proliferation is required.

10. A method in accordance with claim 9, wherein said targeting agent is a monoclonal antibody.

11. A method in accordance with claim 9, wherein said conjugate is a fusion protein expressed by recombinant techniques in a suitable non-human host.

12. A method in accordance with claim 11, wherein said host is a yeast.