Title:  Synthesis of polymer bio-active conjugates

United States Patent:  6,172,202

Assignee:  Pharmacia S.p.A. (Milan, IT)

Filed:  July 7, 1997

Foreign Application Priority Data:  Dec 04, 1992[GB] (9225448)

A process for the preparation of a conjugate between a poly (ethylene glycol) and a protein or glycoprotein, the process comprising specifically binding the domain to a specific binder, to shield the domain from the poly (ethylene glycol) in the following conjugating step, thereafter conjugating the poly (ethylene glycol) to the protein or glycoprotein, wherein conjugation of the poly (ethylene glycol) to the domain is avoided and thereafter releasing the specific binder from the domain without releasing the poly (ethylene glycol) from the protein or glycoprotein, wherein the protein or glycoprotein is other than a proteolytic enzyme selected from the group consisting of trypsin, urokinase, tissue plasminogen activator, plasmin, chymotrypsin, elastase and kallikrein.

Description of the Invention

The present invention relates to a process for the preparation of a conjugate between a polymer and a bioactive substance.

In the last decade a steadily increasing number of proteins have entered routine clinical use as therapeutic or diagnostic agents (see, for example, Waldmann, T. A., Science, 252 : 1657-1662, 1991 and Jaffe, H. S. and Sherwin, S. A., Drugs of Today, 25 : 311-320, 1989 and Foon, K. A., Cancer Research, 49 : 1621-1639, 1989). The efficacy of many of these agents, however, is limited for two main reasons.

First, the in vivo half-life is often very short (see, for example, Muhlradt, P. F. and Opitz, H. G., European Journal of Immunology, 12: 983-985, 1982 and Jacobs, C. A., Lyncl, D. H., Roux, E. R., Miller, R., Davis, B., Widmer, M. B., Wignall, J., VandenBos, T., Park, L. S. and Beeckmann, M. P., Blood, 77 : 2396-2403, 1991 and Blick, M., Sherwin, S.A., Rosenblum, M. and Gutterman, J., Cancer Research, 47 : 2986-2989, 1987).

Second, in case of heterologous proteins, another problem adds to the first. This is due to the proteins being recognized as foreign substances (antigens) by the immune system of the species treated, thereby leading to an immune response that abolishes, upon a second administration, the pharmacological activity of the heterologous protein (see, for example, Shawler, D. L., Bartholomew, R. M., Smith, L. M. and Dillman, R. O., Journal of Immunology, 125: 1530-1535, 1985 and Schroff, R. W., Foon, K. A., Beatty, S. M., Oldham, R. K. and Morgan, A. C., Cancer Research, 54 : 879-885, 1985 and Traub, U. C., De Jager, R. L., Primus, F. J., Losman, M. and Goldenberg, D. M., Cancer Research, 48 : 4002-4006, 1988).

For these reasons methods have been sought to overcome either one or both of the above mentioned problems, i.e. to prolong the in vivo half-life of proteins and to reduce their antigenicity in case of proteins heterologous with respect to the species to be treated. One particular approach that has been taken is to conjugate proteins to soluble synthetic polymers, in particular poly(ethylene glycol), poly(vinyl pyrrolidone), poly(vinyl alcohol), poly(amino acids), divinylether maleic anhydride, ethylene-maleic anhydride, N-(2-hydroxypropyl)methacrylamide and dextran (see, for example, Abuchowski, A., Van Es, T., Palczuk, N. C. and Davis, F. F. Journal of Biological Chemistry 11, 3578-3581, 1977; Yasuda, Y., Fujita, T., Takakura, Y., Hashida, M., and Sezaki, H. Chemistry and Pharmaceutical Bulletin 38, 2053-2056, 1990; Fagnani, R., Hagan, M. S. and Bartholomew, Cancer Research 50, 3638-3645 (1990); Suck, J. M. and Wild, B. S. U.S. Pat. No. 3,679,653, 1972; Flanagan, P. A., Duncan, R., Rihova, B., Subr, V. and Kopecek, J., Journal of Bioactive and Compatible Polymers 5, 151-166, 1990.

More than 40 proteins have now been modified, mainly using polyethylene glycol), but it has been shown that a variety of other polymers can be substituted to provide reduced immunogenicity and protein stabilisation. In particular, the approach has been used to modify enzymes including arginase, asparaginase, adenosine deaminase galactosidase, lipase, pro-urokinase, streptokinase, superoxide dismutase, trypsin and uricase (see, for example, Nucci, M. L., Shorr, R. and Abuchowski, A. Advanced Drug Delivery Reviews 6, 133-151, 1991; Veronese, F. M., Caliceti, P., Pastorino A., Schisvon, 0., and Sartore. Journal of Controlled Release 10, 145-154, 1989); cytokines and growth factors such as interleukin 2 and human granulocyte colony-stimulating factor (see, for example, Katre, N. V. Journal of Biological Chemistry 144, 209-213, 1990; Tanaka, H., Satake-Ishikawa, R., Ishikawa, M., Natsuki, S. and Asano, K. Cancer Research 51, 3710-3714, 1991) and antibodies (see, for example, Kitamura, K., Takahashi, T., Yamaguchi, T., Noguchi, A., Takashina, K., Tsurumi, H., Inagake, M., Toyokuni, T. and Hakomori, S. Cancer Research 51, 4310-4315, 1991).

A number of methods have been described for linkage of polymers to bio-active protein molecules (see, for example, U.S. Pat. No. 4,179,337, U.S. Pat. No 4,732,863, Jackson C.-J., Charlton, J. L., Kuzminski, K., Lang, G. M. and Sehon A. H. Analytical Biochemistry 165, 114-127, 1987; Veronese, F., M., Largajolli, R., Boccu, E., Benassi, C. A. and Schiavon, O. Applied Biochemistry and Biotechnology 11, 141-152, 1985 ; WO93/15189). However, the methods currently available for conjugation have two major drawbacks. Usually, other than the Veronese approach in WO93/15189, first, the derivatisation procedures reported are inherently random thereby leading to the introduction of polymeric moieties into domains of the molecule that mediate the therapeutically or diagnostically desirable activity(ies). Consequently, the molecule may acquire a prolonged half-life in vivo and, in case of heterologous proteins, reduced immunogenicity, but at the expense of a significant or complete loss of the desired biological activity(ies) (see, for example, Kitamura, K., Takahashi,T.,Yamaguchi,T.,Noguchi,A.,Noguchi, A., Takashima, K.-i., Tsurumi, H., Inagake, M., Toyokuni, T. and Hakanori, S.-i., Cancer Research, 51: 4310-4315, 1991 and Maiti, P. K., Lang, G. M. and Sehon, A. H., International Journal of Cancer, Supplement 3: 17-22, 1988).

Loss of biological activity following polymer conjugation has been observed in the case of both antibodies and enzymes, particularly when access of the modified protein to a macromolecular substrate or receptor is essential to produce biological activity. However, it has been found that inactivation of enzymatic activity is not necessarily a result of polymer conjugation if the domain(s) mediating activity either do not contain functional groups suitable for polymer derivatisation, and/or the binding of polymer molecule(s) does not sterically hinder access of low molecular weight enzyme substrates.

In fact, enzymes like adenosine deaminase and L-asparaginase have been successfully conjugated with polyethylene glycol (see, for example, Hershfield, M. S., Buckley, R. H., Greenberg, M. L., Melton, A. L., Schiff, R., Hatem, C., Kurtzberg, J., Markert, M. L., Kobayashi, R. H., Kobayashi, A. L. and Abuchowski, A., The New England Journal of Medicine, 316 : 589-596, 1987 and Teske, E., Rutteman, G. R., van Heerde, P. and Misdorp, W., European Journal of Cancer, 26: 891-895, 1990). In one case (adznosine deaminase) the product thereby obtained has received approval for clinical use in humans. These examples, however, are the exception rather than the rule.

A second problem associated with synthesis of polymer-protein conjugates has been heterogeneity of the product formed. Polymers are by nature heterogeneous, displaying within any sample a range of molecular weights, ie. they are polydisperse, and in addition any preparation also displays a heterogeneity in the number of functional groups available for attachment to the protein to be modified. Thus, during the conjugation reaction there is opportunity to form a multitude of products. This problem has been previously been exacerbated by the need to control carefully the degree of protein modification to a minimum to ensure retention of substantial biological activity of the protein, whilst concurrently introducing a sufficient number of polymer molecules into the conjugate to facilitate the needed reduction in immunogenicity, and protein stabilisation.

According to the present invention, there is provided a process for the preparation of a conjugate between a polymer and a first substance having a biological activity mediated by a domain thereof, which process comprises:

(a) contacting the first substance with a second substance which specifically binds to the said domain of the first substance;

(b) conjugating a polymer to the first substance having the second substance bound thereto; and

(c) freeing the second substance from the first substance having the polymer conjugated thereto, and wherein, when the first substance is a proteolytic enzyme chosen from trypsin, urokinase, tissue plasminogen activator, plasmin, chymotrypsin, elastase and kallikrein then the polymer is other than polyethylen glycol.

A significant improvement can thus be achieved. The procedure of the invention preserves the advantages deriving from the conjugation of polymers to therapeutically or diagnostically useful molecules, i.e. prolonged half-life in vivo and reduced immunogenicity in case of heterologous proteins. Further, a desired biological activity is not lost and a homogeneous product can be attained.

The invention relies upon the use of a second substance that specifically recognizes a domain that mediates the desired biological activity of a first substance which is to be derivatized. The second substance can be viewed as a specific binder substance. The first substance is allowed to interact with the specific binder before effecting polymer conjugation. This ensures that the domain of the first substance that mediates the desired biological activity of that substance is shielded and consequently unavailable for derivatization by the polymer. Following its elution from the specific binder, the conjugate between the polymer and the first substance can be recovered with fully preserved biological activity.

The first substance may be any molecule having a desired activity. The substance may be a physiologically active substance. The first substance is typically an organic macromolecular entity such as a protein or a glycoprotein. The second substance binds to a site on the first substance which mediates the activity of the first substance. The second substance therefore protects the active site of the first substance.

The invention may be applied broadly. The first substance may be an antibody or antibody fragment, cytokine, antigen, enzyme, ligand or receptor. The second substance may be, respectively, an antigen or antidiotypic antibody, receptor or anticytokine antibody, antibody, enzyme substrate, receptor or ligand. In each case the second substance specifically binds to the first substance to shield the domain of the first substance which is responsible for the activity of the first substance. Examples of pairs of first and second substances are as follows

1. When the first substance is a monoclonal antibody (mAb), including anti-idiotypic and anti-anti-idiotypic antibodies, or mAb fragment such as Fab'and F(ab)₂ fragments, the second substance may be the antigen to which the mAb or mAb fragment binds or an antibody including idiotypic, anti-idiotypic, anti-anti-idiotypic antibodies. The mAb or mAb fragment which constitutes the first substance may be specific for any appropriate antigen, for example human tumor necrosis factor .alpha. (hu TNF.alpha.). The mAb fragment may be non-human, for example a murine mAb or mAb fragment.

2. The first substance may be a protein and the second substance may be a ligand which specifically binds to an active site of the protein. The first substance may be a fibrinolytic enzyme such as pro-urokinase (pro-UK), urokinase (UK) or tissue plasminogen activator (tPa) or a fibrinolytic agent such as streptokinase or a non-protease enzyme, such as .beta.-glucuronidase, purine nucleoside phosphorylase,bilirubin oxidase or superoxide dismutase. The second substance may be a mAb which is directed against the specific active site of the first substance. Alternatively the second substance may be the ligand with which the first substance normally interacts. For example:

First substance: streptokinase or hirudin, second substance plasminogen or thrombin respectively; and

first substance: a growth factor such as FGF, EGF, PDGF, HGH or HGF, a growth factor receptor, a lymphokine or cytokine such as an inteteron, interleukin, stimulating factor, TNF.alpha. or TIF.beta. or a lymphokine or cytokine receptor; second substance the corresponding growth factor receptor, the corresponding growth factor, the corresponding lymphokine or cytokine receptor or the corresponding lymphokine or cytokine respectively.

The second substance may be a low molecular weight ligand. This applies generally to all therapeutic proteins having enzymic activity and to proteolytic enzymes. The therapeutic protein may be a fibrinolytic enzyme such as pro-UK, UK or tPA and the second substance may be benzamidine or a derivative thereof. A heparin-binding protein, for example a growth factor such as FGF or HGF, may be the first substance and heparin or a heparin-like molecule or a heparin derivative typically having a low molecular weight and a negative charge may constitute the second substance.

In other instances, the first substance may be a protein that interacts with a peptide ligand. This may be applied to any protein having a therapeutic application. The first substance may be a DNA binding protein and the second substance may be DNA or an oligonucleotide.

In some instances it may be desirable to preserve the biological activities mediated by two separate domains of the first substance. In such cases both biological activities can be preserved through the use of two specific binders each one recognizing one of the two biologically active domains. Thus, for example, monoclonal antibodies against tumor-associated antigens destroy the relevant target cells through binding on the one hand, to the antigen with their variable regions and, on the other hand, to killer cells (K cells) through their constant regions. The K cells become at this point the actual effector of the lytic process that leads to the destruction of the tumorigenic target cell. This process is called antibody-dependent cellular cytotoxicity.

In this case it would be desirable to preserve binding of the monoclonal antibody to both cell types (tumor cell and K cell). This can be achieved, according to the present invention, by performing the conjugation step after having shielded, on the one hand, the antigen-binding regions with the tumor-associated antigen and, on the other hand, the binding site for K cells with the K cell-receptor for the constant region of the monoclonal antibody.

The first substance's active domain is therefore protected by the second substance. The first substance can then be conjugated to a polymer. Generally the polymer is an inert, synthetic, polymeric carrier. The polymer is usually water-soluble.

The polymer may be poly(ethylene glycol) (PEG), poly(vinyl pyrrolidone), poly(vinyl alcohol), poly(amino acids), divinylether maleic anhydride, ethylene-maleic anhydride, N-(2-hydroxypropyl)-methacrylamide or dextran. The polymer may be derivatised or activated itself prior to use. PEGylation may be achieved using, for example, monomethoxypolyethylene glycol-succinimidyl succinate (mPEG). Coupling may be effected using techniques analogous to those already known in the art, for example according to U.S. Pat. No. 4,179,337, U.S. Pat. No. 4,732,863, WO 91/01758, WO 91/15242, U.S. Pat. No. 4,935,465 and U.S. Pat. No. 4,415,665.

Generally there is a molar excess of polymer with respect to the first substance. In accordance with the present invention the molar excess may be from 1 to 500 times, or higher, without there being any loss of function activity. Conjugation must be carried out under such conditions that the second substance does not disassociate from the first substance during the binding procedure. When the first substance is an antibody and the second substance is an antiidiotypic antibody, therefore, the pH of the reaction medium is about 7, typically from about 6.3 to about 7.8, in particular about 7.2. Preferably conjugation is effected under conditions such that polymer is conjugated to the first substance at substitution levels for the reduction of immunogenic activity or elongation of the half-life of the biologically active substance. The person skilled in the art will appreciate that many other polymers can be used for this purpose. In a preferred embodiment, the specific binder that affords protection during the polymer conjugation step is first covalently linked to a solid phase such as column packing materials, for instance sephadex or agarose beads, or a surface, e.g. reaction vessel. This allows the polymer-derivatized first molecule to be separated from the specific binder by elution. The fluid phase containing the derivatized first substance is separated from the solid phase to which the specific binder remains covalently-linked. Such separation can be achieved by many other means. Thus, the specific binder may be derivatised itself with a third molecule (e.g. biotin) that can itself be recognized by a second specific binder (e.g. streptavidin). The second specific binder may be linked to a solid phase thereby allowing the separation of the polymer-derivatized first molecule from the first specific binder-third molecule complex through passage over a second specific binder-solid phase column which will retain, upon subsequent elution, the first specific binder-third molecule complex, but not the polymer-derivatized first molecule. The first substance to which the polymer is conjugated may be released from the second substance in any appropriate fashion. Deprotection may be achieved by providing conditions in which the second substance dissociates from the active domain of the first substance. A complex between an antibody to which a polymer is conjugated and an antiidiotypic antibody can be dissociated by adjusting the pH to an acid or alkaline pH.

Claim 1 of 8 Claims

What is claimed is:

1. A process for the preparation of a conjugate between a poly (ethylene glycol) and a monoclonal antibody having an antigen-binding domain, the process comprising

specifically binding the domain to an antiidiotypic antibody, to shield the domain from the poly (ethylene glycol) in the following conjugating step;

thereafter, conjugating the poly (ethylene glycol) to the monoclonal antibody, wherein conjugation of the poly (ethylene glycol) to the domain is avoided; and

thereafter, releasing the antiidiotypic antibody from the domain without releasing the poly (ethylene glycol) from the monoclonal antibody.

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.