Patent Application
20090098148
The present invention relates to methods and compositions that can be employed to introduce toxins and nucleic acids into the cytoplasm or nucleus of a eukaryotic cell, particularly a cell of a higher vertebrate. The invention allows for the efficient and specific delivery of the toxins and nucleic acids into cells that bind an antibody. The invention particularly concerns the use of a fusion protein of streptavidin and protein A sequences to form a non-covalent complex of a toxin or nucleic acid and an antibody.
The invention provides a method of treatment of human disease by the introduction of toxins or antisense nucleotides into human cells, e.g., tumor cells, in vivo or ex vivo. The invention also provides methods of conducting biological research and methods useful in the production of biological products by introducing exogenous duplex DNA molecules into cultured cells.
The selective introduction of compounds into the cytoplasm or nucleus of specific cells has been a valuable technique in biological and medical research and in medical practice. Cell specific targeting of cytotoxins has been accomplished by complexing toxins with cell binding proteins that can preferentially bind targeted cells. The cell-binding proteins of the complex can be either antibodies, particularly monoclonal antibodies, or protein ligands, e.g., /growth factors/ which recognize the corresponding surface antigens or receptor. Complexes of toxin and antibody have been termed immunotoxins.
Conventionally, the toxin and cell binding protein of the complex have been linked covalently through either chemical coupling or gene fusion. Conventional immunotoxins have been made by chemically linking a toxin component to an antibody, typically a monoclonal antibody, with a heterobifunctional cross-linking reagent that is non-specific. Accordingly, this method yields a heterogeneous product in which some toxin molecules block the antibody's ability to bind antigen by linking to the F(ab) portion of the antibody. Additionally, the coupling chemistry can partially destroy the toxins activity.
Many of the problems associated with chemical conjugation have been overcome through the generation of single-chain fusion toxins using recombinant DNA technology. However, this technology requires that a new recombinant toxin for each target cell. The biological activity of each new recombinant toxin is unpredictable.
Alternative techniques have been developed in which the cytotoxin is non-covalently linked to the antibody or ligand. One such technique exploits the specific interaction between Staphylococcal aureus protein A and immunoglobulins to generate antibody complexes with two specificities. According to this technique, protein A is complexed with antibodies of two different specificities: a toxin specific antibody and a cell surface specific antibody. Such complexes have been used to deliver ricin toxin into targeted cells (Laky, et al., 1986/1987, Immunology Letters 14:127-132). In a second immunotoxin targeting system, single chain antibodies are fused with streptavidin which has a strong and specific binding affinity for biotin. Using this construct, biotinylated toxin was delivered into a target cell (Dubel, et al., 1995, Journal of Immunological Methods, 178:201-209).
Recently, Sano et al. described a fusion protein consisting of streptavidin and one or two immunoglobulin G (IgG)-binding domains of protein A in Escherichia coli. (U.S. Pat. 5,328,985, issued Jul. 12, 1994, which is hereby incorporated by reference in its entirety). The streptavidin-protein A (ST-PA) fusion protein has functional biotin and IgG binding sites. Sano further described complexes of the streptavidin-protein A fusion protein, a monoclonal antibody to BSA, and biotinylated horseradish peroxidase.
Sano also described a method of labeling cells using the ST-PA fusion protein. Cells were incubated with an antibody to a cell surface antigen, Thy-1. The chimeric protein-biotinylated marker complex was subsequently added to the cell suspension. This technique was used to deliver biotinylated FITC to the surface of cells having Thy-1 antigen on their surface. However, Sano did not describe or suggest the use of the ST-PA fusion protein to deliver compounds into the cytoplasm or nucleus of specific cells.
Immunotoxins appear to enter the cell via receptor-mediated endocytosis (Pastan et al., 1986, Cell 47:1-44 and Pirker et al., 1987, Lymphokines 14:361-382). Binding of the antibody moiety of the immunotoxin complex to the surface receptor is followed by, first, clustering of the complex into coated pits and then by internalization of the complex into endosomes or receptosomes within the cell (Middlebrook et al., 1994, Microbiol. Rev., 48:199-221; Morris et al., 1985, Infect. Immun. 50:721-727; Fitzgerald et al., 1980, Cell 21:867-873). During the journey into the cell, the complex may be transported through different intracellular compartments that vary in pH and proteolytic enzyme activity before the toxins are translocated across an intracellular membrane and into the cell cytoplasm where they can cause cell death.
A second area which has been developed concerns methods for introducing nucleic acids into cells. The most widely used methods employ calcium phosphate or DEAE-dextran to promote uptake of nucleic acids. These methods appear to involve the steps of DNA attachment to the cell surface, entry into the cytoplasm by endocytosis, and subsequent transfer into the nucleus. Maniatis, Laboratory Cloning Manual, volume 2, 16.30. Depending upon the cell type, up to 20% of a population of cultured cells can take up DNA using calcium phosphate or DEAF-dextran.
Electroporation is an alternative transfection method in which an electric field is applied to open pores in the cell plasma membrane. DNA appears to enter the cell through these pores.
Liposomes have also been used to introduce nucleic acids into cells. According to this technique, artificial lipid-bilayer vesicles containing cationic and neutral lipids mediate the transfer of DNA or RNA into cells. The mechanism of liposome-mediated transfection, is not well understood, but it appears that negatively charged phosphate groups on DNA bind to the positively charged surface of the liposome, and that the residual positive charge binds to negatively charged sialic acid residues on the cell surface.
Sano did not use the complex to introduce nucleic acid into the cell. Sano et al. described DNA-antibody complexes with the ST-PA fusion protein by incorporating a single biotin molecule at one end of a linearized pUC 19 plasmid. In contrast to the methods of transfecting nucleic acids into the cell which are described above.
The invention relates to a method of delivering toxins or nucleic acids into specific cell types and to the complexes for the practice of the method. According to the invention, an antibody that recognizes a cell surface antigen is non-covalently bound to the antibody binding site of a ST-PA fusion protein; a biotinylated toxin or nucleic acid is bound to the biotin-binding site. In an alternative embodiment, the toxin or nucleic acid can be bound to a third biotinylated molecule, an adapter, which is bound to the biotin binding-site.
In one embodiment of the present invention, a nucleic acid is delivered into a specific cell type. The nucleic acid can be a biotinylated single stranded nucleic acid bound to the biotin binding site of the ST-PA fusion protein. In an alternative embodiment, the nucleic acid can be a duplex nucleic acid that forms a complementary triplex with a biotinylated single stranded nucleic acid, which is in turn bound to the biotin binding site.
The method of the invention relates to the steps of forming a complex between a streptavidin-protein A fusion protein; an antibody, that is specific for a cell surface protein, which undergoes endocytosis after binding with the antibody; and some targeted material e.g. a biotinylated multidrug resistance (mdr) gene product, prodrug, toxin or nucleic acid; isolating the complex from toxin that is not bound to the biotin binding site; and exposing the target cell, to the complex so that the targeted material enters the cell.
As used herein, toxin, refers to holotoxins, modified toxins, catalytic subunits of toxins, or any enzymes not normally present in a cell that under defined conditions cause the cell's death. A biotinylated toxin refers to a toxin that is either directly biotinylated or one that is bound to a biotinylated adapter.
The term “antibody” refers to any molecule which contains one or more functional antigen binding domains and an Fc domain that specifically binds protein A.
The present invention provides a method of specific delivery of targeted material, e.g., toxins, prodrugs, mdr gene products, or nucleic acids into cells by a complex of the targeted material, an antibody specific for a cell surface antigen on the cell, which antigen is endocytosed when the cell is exposed to an effective concentration of the antibody, and a ST-PA fusion protein.
The streptavidin-protein A fusion protein (ST-PA) binds, non-covalently the cell specific antibody in the antibody binding site and the biotinylated targeted material in the biotin binding site. The manner and method by which the targeted material is biotinylated is not critical; the invention includes the use of any and all such methods and reagents.
In one embodiment, the invention provides a method for the specific destruction of cells, e.g., the destruction of tumor cells in a host or ex vivo in short term or long term culture. A toxin can be selected from the group consisting of: thymidine kinase, endonuclease, RNAse, alpha toxin, ricin, abrin, Pseudomonas exotoxin A, diphtheria toxin, saporin, momordin, gelonin, pokeweed antiviral protein, alpha-sarcin and cholera toxin. The toxin is biotinylated and the complex of PA-ST/Ab/biotinylated toxin is formed. An effective amount of the complex can then be administered to the host or to the ex vivo culture system.
In an alternative embodiment the invention provides a method of using an antibody to a cell to introduce into the cell a cytotoxic prodrug, i.e., a non-toxic compound that is converted by an enzyme, normally present in the cell, into a cytotoxic compound. The embodiment contemplates the use of a complex of the ST-PA chimeric protein with a tumor selective monoclonal antibody and prodrug. Compounds suitable as prodrugs include by way of example glutamyl derivative of a benzoic acid mustard alkylating agent, phosphate derivatives of etoposide or mitomycin C, and phenoxyacetamide derivative of doxorubicin.
A further embodiment of the invention contemplates a complex for the delivery of single stranded nucleic acid into cells. As used herein, the term “single stranded nucleic acid” refers to both naturally and non-naturally occurring nucleic acids. The nucleic acid to be delivered can be biotinylated using any of the methods currently available, such as, random incorporation, extension reactions using DNA polymerase, and PCR with biotinylated primers. Such nucleic acids can be used to specifically destroy mRNAs to which they are complementary.
A further embodiment of the invention concerns the introduction into a cell of a duplex DNA that can integrate into the cell's genome or replicate episomally and that can be transcribed. The direct incorporation of biotin into a duplex DNA can interfere with both replication and transcription of the DNA. To overcome these problems, the duplex DNA to be delivered into the cell is modified to include a sequence that can form a region of triplex nucleic acid with a single stranded nucleic acid.
Triplex DNA formations have been described and typically consist of T-A-T and C-G-C nucleotide triads. The third strand of the triplex occupies the major groove of an A-form DNA helix and forms Hoogsteen base pairs with the homopolymeric duplex DNA. Alternatively, a triplex can be formed with duplex DNA. As used herein, the term “triplex” refers to a structure formed by Hoogsteen base pairing between a duplex DNA and a single stranded nucleic acid.
In this embodiment, the complex contains a duplex DNA Hoogsteen paired to a single stranded nucleic acid, which is biotinylated and complexed with the ST-PA/mAb complex. The duplex DNA can be a linear duplex DNA, which is suitable for recombination into the genome of a cell or, alternatively, the duplex can be a circular or supercoiled circular DNA, which can episomally replicate.
This embodiment of invention can be used under any circumstances it is desired to introduce cloned DNA into a cell, e.g., to express a product or to alter the phenotype of the cell, to investigate the function of any cloned gene.
A further embodiment of the invention comprises a complex of an antibody to a cell surface protein found on an antigen presenting cell, the ST-PA fusion protein, and a biotinylated protein of a pathological bacteria or virus, such a complex can be used to localize the antigen to antigen presenting cells and thereby enhance the immune response of CD4 positive T cells relative to that of other lymphocytes.
The complexes of the present invention can be formed by simply admixing ST-PA, a monoclonal antibody, and the biotinylated material in the appropriate ratios. The components can be mixed in any order.
The ST-PA fusion protein forms tetramers which bind up to four biotinylated molecules and four IgG molecules, which are each bivalent. Without limitation as to theory, the octovalent binding of complexes of the invention to the cell surface is believed to cause the complexes to have superior binding and internalization properties compared to other immunotoxins and immunopharmaceutical complexes.
Biotin-blocked streptavidin is capable of specifically interacting with cell surfaces through an Arg-Tyr-Asp sequence present in the protein (the “RYD site”). This site is distinct from the biotin-binding cleft of the protein and bears high homology to the RGD-containing cell binding domain of fibronectin which mediates fibronectin-cell surface interactions (Alon et al, 1993, Europ. J. Cell Biol. 60: 11). Studies have suggested that streptavidin acts as a close mimetic of fibronectin (Alon et al, 1993, Europ. J. Cell Biol. 60: 1-11).
The conserved RYD and RGD domains of fibronectin and streptavidin function as universal recognition sequences for interactions with many membrane-bound receptors. In one embodiment of the invention, the streptavidin component of the complex can be modified to alter the RYD site. As used herein, the term “modified RYD sequence” includes any alteration to the RYD site or flanking region which eliminates the non-biotin binding site-related interaction of streptavidin with cell surface proteins. One such modification is the replacement of aspartic acid by glutamic acid.
Plasmid, pTSAPA-2, described in U.S. Pat. No. 5,328,985, issued Jul. 12, 1994, carries the chimeric gene of streptavidin and protein A (region E and D). Expression and purification of the gene fusion of ST-PA was carried out according to the methods that follow.
Bacterial Strain lysogen BL21 (DE3) (pLysS) was transformed with the pTSAPA-2 streptavidin-protein A fusion expression vector. The transformed strain was grown at 37° C. in LB media supplemented with 50 μg/ml ampicillin, 34 μg/ml chloramphenicol and 0.2% glucose. When the absorbance at 600 nm of the culture was between 0.8 and 1.0 OD, 100 mM isopropyl β-D-thiogalactopoyranoside (IPTG) dissolved in water was added to a final concentration of 0.4 mM to induce the T7 RNA polymerase gene placed under the lac UV5 promoter. After the induction, the cells were incubated at 37° C. with shaking for 2 hours.
Purification of streptavidin-protein A fusion chimeric protein was carried out at 4° C. or on ice unless otherwise indicated. The culture (100 ml) of BL21 (DE3) (pLysS)(pTSAPA-2) incubated for 2 hours after the induction was centrifuged at 2,900×g for 15 min. The cell pellet was suspended in 10 ml of 2 mM EDTA, 30 mM Tris-Cl (pH 8.0), 0.1% Triton X-100, 0.5 mM PMSF to lyse the cells and the lysate was stored at −70° C. until used. To the thawed cell lysate, PMSF, leupeptin, and pepstatin A were added to final concentrations of 0.5 mM, 1 μM, and 1 μM, respectively. The lysate was then treated with 10 μg/ml of deoxyribonuclease I and 10 μg/ml ribonuclease A in the presence of 12 mM MgSO4 at room temperature for 20 minutes. The mixture was centrifuged at 39,000×g for 15 minutes and the pellet was dissolved in 100 ml of 7 M guanidine hydrochloride overnight at 4° C. with stirring. After the pellet was dissolved the protein was then dialyzed against 150 mM NaCl, 50 mM Tris-C1 (pH 7.5), 0.05% Tween 20, 0.1 mM PMSF, 1 μM leupeptin, 1 μM pepstatin A, 0.02% NaN3. To achieve slow removal of the guanidine hydrochloride, the dialysis bag containing the protein solution was left overnight in the dialysis solution (˜1,000 ml) without stirring, followed by 3 changes of the dialysis solution and dialysis with stirring at 4° C. The dialysate was centrifuged at 39,000×g for 15 minutes, and the supernatant was applied to an IgG Sepharose 6 Fast Flow column (1.2×1.1 cm) previously washed with 5-10 bed volumes of TST Buffer. The column was then equilibrated with 2-3 bed volumes of each: 1) 0.5 M Acetic acid, pH 3.4 (pH adjusted with NH4CH3COOH(NH4Ac); 2) 150 mM NaCl, 50 mM Tris-Cl (pH 7.5), 0.05% Tween 20 (TST Buffer); 3) 0.5 M Acetic Acid, pH 3.4; and 4) TST. The sample was applied to the column and the unbound protein was removed by washing the column with: 1) 10 bed volumes of TST, and 2) 2 bed volumes of 5 mM NH4Ac, pH 5.0. Elution was performed with 0.5 M Acetic Acid, pH 3.4. The eluate was collected in 1-2 ml fractions, and the fractions having the greatest OD at 280 were dialyzed against 1 M NaCl, 50 mM sodium carbonate (pH 11.0). The dialysate was clarified by centrifugation at 39,000×g for 15 minutes, and applied to a 2-iminobiotin agarose column (1.2×1.2 c.m) previously equilibrated with 1 M NaCl, 50 mM sodium carbonate (pH 11.0). After the unbound proteins were removed with the same solution, the bound proteins were eluted with 6 M urea, 50 mM ammonium acetate (pH 4.0). The eluted proteins were dialyzed against Tris-buffered saline [TBS; 150 mM NaCl, 20 mM Tris-Cl (pH 7.5)] containing 0.02% NaN3, and the dialysate was stored at 4° C. after filtration through a 0.22 μm filter (Millex-GV, Millipore).
Formation of the ST-PAS/mAb/biotinylated β-Galactosidase complex:
A mixture of ˜2 μg of antibody, ˜28 μg of fusion protein, and ˜2 units of biotinylated β-Galactosidase was incubated at room temperature for at least 10 minutes. After this incubation, the complex is ready to use.
β-Galactosidase staining of cells that adhere to the plate was performed according to Sanes, et al., 1986, EMBO J. 5: 3133-3142.
The protocol of Molecular Probes, Inc. was used to detect lacZ β-Galactosidase gene expression in cells that grow in suspension.
The above protocols were used to determine if the complex that contains the antibody coupled to the fusion protein and the biotinylated β-Galactosidase enzyme was successfully transduced into the cell of choice. If the transduction was complete, the cells were blue after overnight incubation.
Biotin-blocked streptavidin binds specifically (Kd=3×108M) to cell surfaces, presumably via an RYD containing sequence that is distinct from the biotin-binding cleft of the protein.
Alternation of the RYD domain (sequence) to other amino acid residues is expected to eliminate the non-biotin related specific surface binding of a large variety of cells.
One way to modify the RYD sequence would be to change the RYD sequence to RYE. The change of RYD sequence to RYE can be achieved by introducing a point mutation using sequential PCR steps. This can be achieved by designing two primers, DES (SEQ ID NO: 3) and DER (SEQ ID NO: 4) (see
3) Delivery of Biotin-β-Galactosidase into A431 Cells.
In order to study the ability of ST-PA to deliver compounds into a cell, the delivery of this complex was compared with that of a complex in which core streptavidin was covalently linked to the TGFα receptor.
pTSA-TGFα, an expression vector for streptavidin-TGF-α (ST-TGF) was constructed by replacing the gene of protein A in pTSAPA-2 with a mature human TGF-α gene (amino acids 1-50).
To demonstrate the capability of ST-PA and ST-TGFα fusion proteins to deliver biotinylated protein into specific cells types, biotinylated β-galactosidase was complexed with the streptavidin component of the fusion protein. Delivery of β-galactosidase into A431 cells was quantitated using known staining techniques and FACS analysis.
ST-PA/biotin β-galactosidase was complexed with anti-EGFR mAb and the resulting complex was then incubated with A431 human epidermoid cells over-expressing epidermal growth factor receptor (EGFR). Alternatively, ST-TGF was mixed with biotin-β-galactosidase and the resulting complex was administered to A431 cells.
As shown in
The highly efficient delivery by ST-PA fusion protein of biotin-β-galactosidase into A431 cells may be due to the four binding sites for biotin and IgG contained on each ST-PA tetramer (
4) Delivery of Biotin β-Galactosidase into Cells Using mAbs of Different Specificity.
The experimental procedure applied in studying the delivery of biotin β-galactosidase into A431 cells was also used to study the delivery of the antibody/ST-PA/biotinylated β-galactosidase complex into other cell types. The study utilized antibodies that recognize several different cell surface molecules. The cell lines and cell surface molecules used in this experiment are summarized in
5) pAT-β-Galactosidase Expression Vector.
Although biotin can be incorporated into DNA (e.g. expression plasmids), random incorporation of biotin into DNA may result in a loss of transcriptional activity. In order to retain the transcriptional activity of the DNA to be introduced into the cell, a new gene transfer system was developed which utilizes a pAT-expression vector having a poly(dA)/poly(dT) tract downstream of the expression cassette (
The protocol for triplex DNA formation and DNA transfer of the β-galactosidase gene into cells is as follows. The β-galactosidase gene is cloned into the expression cassette of the pAT-expression vector (
The present application is a continuation of application Ser. No. 10/289,921, filed on Nov. 6, 2002, which is a divisional of application Ser. No. 08/566,421, filed Nov. 30, 1995, now U.S. Pat. No. 6,497,881. The prior applications are hereby incorporated herein by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 08566421 | Nov 1995 | US |
| Child | 10289921 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10289921 | Nov 2002 | US |
| Child | 12335010 | US |
