Patent Application
20090062514
The present invention relates to a process of affinity purifying a protein of interest using an affinity matrix comprising recombinant plant viral particles or recombinant plant virus-like particles. The invention further relates to the affinity matrix and to the recombinant viral particles, whereby the recombinant viral particles expose one or more recombinant proteins on their surface. The invention also relates to a fusion protein as a building block for said recombinant viral particles, to a polynucleotide encoding the fusion protein and to a plant, plant tissue or plant cells comprising said polynucleotide. The invention further relates to a process of producing said affinity matrix and to a process of producing said recombinant viral particles. The invention also relates to the use of said fusion protein for affinity purifying a protein of interest.
It is known that in microbial protein expression systems that have been optimized with regard to product yield, up to 90% of total production costs are the costs related to purification of the protein of interest from the host, rather than expenses for the production itself. In order to make protein production more economical, strategies are needed that will allow rapid and inexpensive separation of the protein of interest or non-proteinaceous small molecule of interest, from other contaminating components.
It has been proposed (WO02/068927) to use plant viral particles having short peptides (e.g. a FLAG-tag) bound to the surface of plant viral particles for purification of a protein of interest by affinity purification. However, in the process of the prior art, the protein of interest to be purified has to be fused to a protein such as a single chain antibody or other purification tag that is capable of binding to the peptide bound to the surface of the viral particles. It is therefore necessary to cleave the protein to be purified after the affinity purification step, which represents an additional process step that one would like to avoid.
Antibodies and antibody derivatives constitute about 20% of biopharmaceutical products currently in development. The purification of antibodies accounts for 50-80% of the total production costs (for review: Roque et al., 2004, Biotechnol. Prog., 20:639-654). Protein A from Staphylococcus aureus is widely used as an affinity protein in processes of immunoglobulin purification (for review: Jungbauer & Hahn, 2004, Curr. Opin. Drug. Disc. & Dev., 7:248-256). Protein A reversibly interacts with the Fc domain of immunoglobulins (Lindmark et al., 1983, J. Immunol. Methods, 62:1-13; Gouda, et. al., 1998, Biochemistry, 37:129-136), predominantly via hydrophobic interactions (Dowd et al., 1998, Nat. Biotechnol, 16:190-195). The high stability and selectivity of protein A makes it a preferable generic ligand for immunoglobulin purification. The main source of protein A for the market has been recombinant protein A produced in E. coli (Duggleby & Jones, 1983, Nucleic Acids Res., 11:3065-3076; Engel et al., 1992, Protein Expr. Purif., 3:108-113). In prior art processes of purifying antibodies by affinity purification with protein A, protein A first has to be expressed and purified and then linked to a matrix such as sepharose that is then used for affinity purification of the antibodies. Thus, the production of the affinity matrix involves many steps and is laborious and expensive. Due to the costs for the affinity matrix, the affinity matrix is typically used for several purification runs, leading to a risk of contamination between consecutive samples purified on the same affinity matrix. A cheaper and readily producible affinity matrix for the purification of antibodies is therefore much needed. Such a cheap affinity matrix could be a single used matrix, avoiding the contamination risk.
It is therefore an object of the invention to provide a process of affinity purifying a protein of interest, wherein no cleavage step has to be performed on the protein of interest after the affinity purification step. It is a further object of the invention to provide a process of purifying immunoglobulins such as therapeutic antibodies or fusion proteins thereof using an economical and easily accessible affinity matrix. It is another object of the invention to provide a process of purifying therapeutic antibodies without the risk of contamination with human or other animal pathogens. It is a further object of the invention to provide an affinity matrix for said purifying processes.
These objects are achieved by a process of purifying a protein of interest, comprising the following steps:
The above objects are further achieved by recombinant viral particles or recombinant plant virus-like particles comprising a plurality of fusion protein molecules, said fusion protein comprising the following fusion protein domains:
The invention further provides recombinant viral particles or recombinant plant virus-like particles comprising fusion protein molecules, said fusion protein comprising the following fusion protein segments:
In one embodiment, the fusion protein comprises the following domains: a plant viral coat protein domain, a recombinant protein domain and at least one peptide linker linking said plant viral coat protein domain and said recombinant protein domain. In this embodiment, said recombinant protein may be present within the primary structure of the amino acid sequence of the coat protein, whereby the coat protein domain may be formed by two coat protein segments of the primary structure of the fusion protein. In this embodiment, said recombinant protein is linked to said coat protein by two peptide linkers, one peptide linker linking the N-terminal portion of said recombinant protein to the N-terminal segment of said coat protein, the second peptide linker linking the C-terminal portion of said recombinant protein to the C-terminal segment of said coat protein.
In another embodiment, the coat protein domain, the recombinant protein domain and one peptide linker are sequential segments in the primary structure of said fusion protein. In this embodiment, there are two possibilities for the sequence of the fusion protein segments (or domains) from the N-terminus to the C-terminus of the fusion protein: (i) said plant viral coat protein is located at the N-terminus of the fusion protein and is followed by said peptide linker followed by said recombinant protein that is located at the C-terminal end of the fusion protein; (ii) said recombinant protein is located at the N-terminus of the fusion protein and is followed by said peptide linker followed by said coat protein that is located at the C-terminal end of the fusion protein. Thus, in one embodiment of the invention, the fusion protein comprises one peptide linker.
In all these embodiments, the fusion protein may comprise further amino acid residues or sequence segments at the N-terminus, at the C-terminus or within said fusion protein. “Domain” and “segment” are used interchangeably herein.
Said plant viral coat protein may be derived from any plant virus listed below. In one embodiment, said plant viral coat protein is derived from a plant virus forming rod-shaped viral particles. “Being derived” means that the coat protein used in the fusion protein of the invention does not have to be identical to the natural coat protein of a plant virus. Instead, the coat protein used in the fusion protein may have additions, deletions, insertions or mutations relative to a natural coat protein of a plant virus. It is only necessary that the coat protein maintains its capability to form viral or virus-like particles under suitable conditions. In one embodiment, at most 20 amino acid residues of the natural plant viral coat protein are deleted and/or mutated. In another embodiment, at most 20 amino acid residues are inserted into the natural sequence of the plant viral coat protein of the plant virus from which the coat protein of the invention is derived.
Said coat protein may be derived from a plus-sense single-stranded RNA virus. Examples of plant viruses the coat protein of which may be used in the present invention include tobamoviruses such as tobacco mosaic virus (TMV), turnip vein clearing virus, potato virus X, potato virus Y and fragments or homologues thereof, provided said fragments or homologues are capable of forming viral particles or virus-like particles under suitable conditions. Preferably, the coat protein of the invention has a sequence identity of at least 40% to the coat protein of turnip vein clearing virus, to tobacco mosaic virus, potato virus X or potato virus Y. In another embodiment, said sequence identity is at least 50%; in a further embodiment, said sequence identity is at least 60%. In an important embodiment, said coat protein has a sequence identity to the coat protein of tobacco mosaic virus of at least 80%.
The recombinant protein of the invention is exposed on the surface of said recombinant viral particles. There are no restrictions with regard to said recombinant protein. Said recombinant protein may be any protein segment fused to a plant viral coat protein preferably via one or more peptide linkers. The type of said recombinant protein may be chosen depending on the application of the viral particles of the invention. The inventors have found for the first time that it is possible to create recombinant viral particles having a recombinant protein exposed on the surface of said viral particles without being restricted to small peptides of less than 40 or even less than 20 amino acids. Therefore, the invention shows its full potential with recombinant proteins having a size of at least 50 amino acid residues. However, in one embodiment, said recombinant protein has a size of at least 70 amino acid residues; in a further embodiment, said recombinant protein has a size of at least 90 amino acid residues; in a still further embodiment, said recombinant protein has a size of at least 110 amino acid residues.
The recombinant viral particles or virus-like particles of the invention are plant viral particles in that the coat protein domain or segment of said fusion protein is derived from a plant virus. The viral particles of the invention are recombinant in that they are assembled from a coat protein that is part of the fusion protein of the invention. The recombinant viral particles of the invention are also referred to herein as “said viral particles”.
Said recombinant protein may function as an affinity protein e.g. when a matrix of said viral particles is used for affinity purification of a protein of interest. Therefore, the terms “recombinant protein” and “affinity protein” are used interchangeably herein for a protein exposed on the surface of the viral particles of the invention. The recombinant protein of the invention is recombinant in that it is a segment or domain of the fusion protein of the invention.
For allowing affinity purification of a compound or protein of interest using the viral particles of the invention, said recombinant protein preferably has an affinity to the compound or protein of interest. Herein, a protein to be purified using the affinity matrix or the recombinant viral particles or virus-like particles, or the fusion protein of the invention is termed “protein of interest”. The protein of interest to be purified is a protein different from the fusion protein of the invention. In one embodiment, said recombinant protein has affinity to immunoglobulins or derivatives thereof such as therapeutic antibodies. The affinity to immunoglobulins or derivatives thereof may be to the constant region of the immunoglobulins. In this case, said recombinant protein may be staphylococcal protein A or a domain or derivative thereof having affinity to immunoglobulins. In another embodiment, said recombinant protein may be streptococcal protein G or a derivative thereof capable of binding immunoglobulins. In another embodiment, said recombinant protein may be streptavidin or a derivative thereof such as strepactin having affinity to the StrepTagII.
If the compound to be purified is a small molecule, the recombinant protein can be any protein having affinity to said small molecule. For example, said recombinant protein can be an antibody or a single-chain fragment of an antibody having affinity to said small molecule.
The peptide linker of the invention links said plant viral coat protein and said recombinant protein in the primary structure of said fusion protein. The peptide linker allows assembly of viral particles of said fusion protein despite of the presence of said recombinant protein that may have a size of at least 50 amino acid residues. Said peptide linker should be flexible. In one embodiment, said peptide linker has no secondary structure in order to be flexible. In another embodiment, said peptide linker forms a helix. Preferably, said peptide linker does not form a β-sheet. It belongs to the general knowledge of the skilled person to design peptides having a predetermined secondary structure or no secondary structure. For example, proline residues break helices and β-sheets. One may therefore include one or more proline residues into said peptide linker. Alternatively, said peptide linker may contain a large proportion of glycine residues, whereby highly flexible peptide linkers may be obtained.
Said peptide linker preferably has at least 10 amino acid residues. In another embodiment, said peptide linker has at least 15 amino acid residues; in a further embodiment, said peptide linker has at least 20 amino acid residues; in a further embodiment, said peptide linker has at least 30 amino acid residues. The bigger the recombinant protein to be bound to the surface of said viral particle, the longer should the peptide linker be made. For example, if said recombinant protein has more than 200 amino acid residues, the peptide linker preferably has at least 25 amino acid residues.
In one embodiment, the length of said peptide linker is between 10 and 70 amino acid residues. In another embodiment, the length of said peptide linker is between 13 and 50 amino acid residues. In a further embodiment, the length of said peptide linker is between 16 and 30 amino acid residues.
In one embodiment, said recombinant protein and said peptide linker together have at least 60 amino acid residues. In another embodiment, said recombinant protein and said peptide linker together have at least 80 amino acid residues. In a further embodiment, said recombinant protein and said peptide linker together have at least 100 amino acid residues. In a further embodiment, said recombinant protein and said peptide linker together have at least 130 amino acid residues.
The viral particles or virus-like particles of the invention can be produced by expressing a polynucleotide encoding said fusion protein of the invention in a bacterial or plant host. Said plant host may be plant cells, plant tissue or entire plants. Apart from encoding said fusion protein, said polynucleotide will have regulatory elements required for the expression of said fusion protein in the chosen host. Upon expressing said polynucleotide, the viral particles of the invention generally assemble within host cells or may be assembled in vitro after isolating said fusion protein from the host cells under suitable conditions.
Said viral particles of the invention do preferably not require the presence of free viral coat protein for assembly. Therefore, in one embodiment, said fusion protein is expressed without co-expressing free viral coat protein. The fusion protein of the invention can, however, assemble to said viral particles in the presence of free coat protein. In one embodiment, said viral particles comprise at most 30 mol-% free viral coat protein, preferably at most 20 mol-%, most preferably at most 10 mol-% of free viral coat protein. The content of free viral coat protein in said viral particles may be determined by solubilizing said viral particles and performing mass spectroscopy such as MALDI or ESI mass spectroscopy for determining the molecular weights and the relative abundance of the proteinaceous components of said viral particles. Any viral RNA contained in said viral particles may either be removed before performing mass spectroscopy or the signal thereof may be neglected when determining the relative abundance of the proteinaceous components of said viral particles.
In a further embodiment, an SDS-PAGE is performed on the viral particles and stained by Coomassie or silver staining. The intensity of the band caused by free viral coat protein will be at most 20%, preferably at most 10% of the intensity of the band caused by said fusion protein, as determined by a commercial gel reader.
For the purpose of this invention, a viral particle or a virus-like particle is defined as an oligomeric particle comprising a plurality of viral coat protein molecules, a plurality of the fusion protein molecules of the invention, or of a mixture of viral coat protein molecules and said fusion protein molecules of the invention. Said particle typically has a size and shape as seen in electron microscopy similar as the size and shape of the viral particle of the wild-type virus from which said coat protein is derived. The sizes of the viral particles or virus-like particles as determined in electron microscopy as described in Analytical Biochem., 333 (2004) 230-235 is preferably at least 10 nm in the shortest dimension, more preferably at least 13 nm in the shortest dimension.
In one embodiment, the recombinant viral particles or virus-like particles are produced in plant cells or plants using plant viral vectors, whereby the coat protein open reading frame (ORF) of a natural plant virus is replaced by the ORF of the fusion protein of the invention. The use of plant viral vectors has the advantage that high amounts of the fusion protein of the invention is produced per host cell, since the plant viral coat protein is the most abundant protein expressed in host cells after infection with a plant virus or plant viral vector. Further, cell to cell movement or systemic movement of the viral vector may lead to spread of the viral vector and to a high number of plant cells expressing said fusion protein. Methods of expressing a protein such as the fusion protein of the invention using a viral vector are known in the art. In one embodiment, the viral vector is introduced into plant cells or cells of a plant as part of a binary vector using Agrobacterium-mediated transformation.
The invention also provides an affinity matrix for purifying a compound or protein of interest. Said affinity matrix comprises a plurality of the viral particles or virus-like particles of the invention. In one embodiment, said viral particles or virus-like particles in said affinity matrix are not covalently cross-linked. In another embodiment, the viral particles or virus-like particles in said affinity matrix may be cross-linked by a cross-linking agent. Cross-linking agents that can be used for cross-linking the viral particles of the invention are known in the art. Examples for such cross-linking agents are glutaraldehyde or bis-succinimides. A cross-linked affinity matrix has improved mechanical properties and a higher molecular weight. Further, covalent cross-linking allows to render said viral particles infection-deficient, which increases the safety of a product purified using said affinity matrix.
For purifying a protein of interest, the affinity matrix of the invention may be filled into a column for affinity chromatography. Affinity chromatography may then be carried out according to conventional methods. In another embodiment, said protein of interest may be purified using said affinity matrix by a batch method (cf. example 4). In any event, the affinity matrix of the invention is used in a solvent that does not dissolve said affinity matrix or said viral particles of the affinity matrix. A suitable solvent is an aqueous solvent, preferably the solvent is water. Due to the high molecular weight and said insolubility of the affinity matrix, the affinity matrix can easily be separated (e.g. by sedimentation) from the soluble contaminants in a solution from which a protein of interest is to be purified.
A protein of interest to be purified according to the invention is typically present in dissolved form in an aqueous solution or dispersion further containing soluble or insoluble contaminants. An example of such a solution is a cell lysate or cell supernatant. Insoluble matter is typically first separated by filtration or centrifugation for obtaining a clear solution. The clear solution may then be contacted with the affinity matrix of the invention, whereby the protein of interest binds to the affinity protein of said viral particles. Next, the affinity matrix having bound protein of interest is separated from the solution that originally contained the protein of interest. After washing the affinity matrix having bound protein of interest, the protein of interest may be detached from the affinity matrix under suitable conditions, whereby a solution containing purified protein of interest is obtained. A protocol for purifying immunoglobulins using an affinity matrix comprising viral particles having bound protein A is given in the examples.
The part of protein A gene encoding for the underlined region of the protein sequence was re-synthesized with a codon usage optimized for expression in N. tabacum and for the structure and stability of the mRNA. It was cloned into TMV-3′ provector (pICH21767).
The length of the cloned sequence is 133 aa (domain E: 56 aa; domain D: 61 aa).
The sequence fragment of mature streptavidin (aa 12-139) was cloned into the 3′-provector pICH21444 for fusion to CP; the mutated residues for increased affinity to StrepTagII are underlined (native sequence at this position is “ESAV”).
Viral coat protein is the main building block of viral particles and virus-like particles (VLP). Viral particles and VLP are structured multimolecular biopolymers. By fusing a recombinant protein with a viral coat protein, it is possible to obtain viral particles with foreign epitopes (said recombinant protein) on their surface. The translational fusion of a recombinant protein with a viral coat protein (Hamamoto et al., 1993, BioTechnology, 11, 930-932; Gopinath et al., 2000, Virology, 267, 159-173; Porta et al., 1994, Virology, 202:949-955; Porta et al., 2003, Virology, 310: 50-63; JP6169789; U.S. Pat. No. 5,977,438; WO02068927) has, however, been restricted in the prior art, as the recombinant protein that could be fused to a plant viral coat protein has been limited to 20-25 amino acid residues (Turpen et al., 1995 Biotechnology, 13: 53-57; Sugiyama et al., 1995, FEBS Lett., 359: 247-250; US 2002/0107387). Speculations in the prior art such as in US 2002/0107387 that larger recombinant proteins could be fused to viral particles could not be confirmed experimentally and thus turned out to be wrong. The experiment shown in
Therefore, in the prior art, significant amounts (up to 95%) of free viral coat protein have been co-expressed with the fusion protein for co-assembly with the fusion protein, whereby viral particles consisting of up to 95% of free vial coat protein are obtained. Obviously, this limits any application of viral particles displaying a peptide on the surface.
In general, the limited size of peptides that could be fused to the CPs of plant viruses such as TMV, cowpea mosaic virus (CPMV), alfalfa mosaic virus, etc., while retaining the ability to assemble into functional virions has restricted, in the prior art, these systems to the expression of short immunogen epitopes and peptide hormones. Another publication (Santa-Cruz et al., 1996, Proc Natl Acad Sci USA, 93:6286-6290) describes formation of Potato Virus X (PVX) virions containing on its surface green fluorescent protein (GFP) expressed as N-terminal fusion with potato virus X coat protein. However, as in the cases with TMV vectors, the expression of significant amount of wild type coat protein was necessary for assembly of viral particles. The latter was achieved either by expression of viral coat protein from independent vector stably integrated into plant chromosomal DNA or by using a cleavable linker peptide in GFP—CP fusion, thus providing a source for CP for virion formation. It was not possible to obtain viral particles displaying GFP on their surface in the absence of free viral coat protein. In a similar study, a single chain antibody was displayed on potato virus X viral particles containing free viral coat protein (Smolenska et al., FEBS Lett. 441 (1998) 379-382). Another work (Bendahmane et al., 2002, Proc. Natl. Acad. Sci. USA, 99:3645-3650) showed that fusion of GFP with coat protein of tobacco mosaic virus in the absence of a peptide linker does not produce recombinant viral particles.
There are many applications for which the recombinant viral particles of the invention are useful such as affinity chromatography. For example, purification of antibodies and antibody derivatives that constitute 20% of biopharmaceutical products currently in development, accounts for 50-80% of total manufacturing costs (for review: Roque et al., 2004, Biotechnol. Prog., 20:639-654). Protein A from Staphylococcus aureus is broadly used as affinity protein in the process of immunoglobulin purification (for review: Jungbauer & Hahn, 2004, Curr. Opin. Drug. Disc. & Dev., 7:248-256). Protein A reversibly interacts with the Fc domain of immunoglobulins (Lindmark et al., 1983, J. Immunol. Methods, 62:1-13; Gouda, et. al., 1998, Biochemistry, 37:129-136), predominantly via hydrophobic interactions (Dowd et al., 1998, Nat. Biotechnol., 16:190-195). The high stability and selectivity of protein A makes it a useful generic affinity protein for immunoglobulin purification. The main source of protein A for the market has been recombinant protein A produced in E. coli (Duggleby & Jones, 1983, Nucleic Acids Res., 11:3065-3076; Engel et al., 1992, Protein Expr. Purif., 3:108-113). Display of protein A as affinity protein on the surface of an affinity matrix, such as matrix comprising viral particles or virus-like particles opens the opportunity for a cheap source of an affinity matrix having bound protein A or having bound another immunoadsorbent to be used in downstream processing of recombinant monoclonal antibodies. Another protein that can be used in this invention is streptococcal protein G (Guss et al., 1986, EMBO J., 5: 1567-1575), that also has strong affinity to Fc domain of IgG (Sauer-Eriksson et al., 1995, Structure, 3:275-278) and also weak affinity to the Fab fragment (Derrick & Wigley, 1992, Nature, 359: 752-754).
The present invention utilizes various properties of plant viruses for the purposes of purifying and visualizing proteins of interest produced in different hosts (which for purposes of this invention is meant to include any biological protein production host or any non-biological protein production method). The general principle of the invention is shown in
A list of viruses to which this invention can be applied is presented below. Taxa names in quotes (and not in italic script) indicate that this taxon does not have an ICTV international approved name. Species (vernacular) names are given in regular script. Viruses with no formal assignment to genus or family are indicated):
DNA Viruses: Circular dsDNA Viruses: Family: Caulimoviridae, Genus: Badnavirus, type species: commelina yellow mottle virus, Genus: Caulimovirus, Type species: cauliflower mosaic virus, Genus “SbCMV-like viruses”, Type species: Soybean chloroticmottle virus, Genus “CsVMV-like viruses”, Type species: Cassaya vein mosaicvirus, Genus “RTBV-like viruses”, Type species: Rice tungro bacilliformvirus, Genus: “Petunia vein clearing-like viruses”, Type species: Petunia vein clearing virus; Circular ssDNA Viruses: Family: Geminiviridae, Genus: Mastrevirus (Subgroup I Geminivirus), Type species: maize streak virus. Genus: Curtovirus (Subgroup II Geminivirus), Type species: beet curly top virus, Genus: Begomovirus (Subgroup III Geminivirus). Type species: bean golden mosaic virus;
RNA Viruses:
ssRNA Viruses: Family: Bromoviridae, Genus: Alfamovirus, Type species: alfalfa mosaic virus, Genus: Ilarvirus, Type species: tobacco streak virus, Genus: Bromovirus, Type species: brome mosaic virus, Genus: Cucumovirus, Type species: cucumber mosaic virus;
Family: Closteroviridae, Genus: Closterovirus, Type species: beet yellows virus, Genus: Crinivirus, Type species: Lettuce infectious yellows virus, Family: Comoviridae, Genus: Comovirus, Type species: cowpea mosaic virus, Genus: Fabavirus, Type species: broad bean wilt virus 1, Genus: Nepovirus, Type species: tobacco ringspot virus;
Family: Potyviridae, Genus: Potyvirus, Type species: potato virus Y, Genus: Rymovirus, Type species: tyegrass mosaic virus, Genus: Bymovirus, Type species: barley yellow mosaic virus;
Family: Sequiviridae, Genus: Sequivirus, Type species: parsnip yellow fleck virus, Genus: Waikavirus, Type species: rice tungro spherical virus; Family: Tombusviridae, Genus: Carmovirus, Type species: carnation mottle virus, Genus: Dianthovirus, Type species: carnation ringspot virus, Genus: Machlomovirus, Type species: maize chlorotic mottle virus, Genus: Necrovirus, Type species: tobacco necrosis virus, Genus: Tombusvirus, Type species: tomato bushy stunt virus, Unassigned Genera of ssRNA viruses, Genus: Capillovirus, Type species: apple stem grooving virus;
Genus: Carlavirus, Type species: carnation latent virus; Genus: Enamovirus, Type species: pea enation mosaic virus,
Genus: Furovirus, Type species: soil-borne wheat mosaic virus, Genus: Hordeivirus, Type species: barley stripe mosaic virus, Genus: Idaeovirus, Type species: raspberry bushy dwarf virus;
Genus: Luteovirus, Type species: barley yellow dwarf virus; Genus: Marafivirus, Type species: maize rayado fino virus; Genus: Potexvirus, Type species: potato virus X; Genus: Sobemovirus, Type species: Southern bean mosaic virus, Genus: Tenuivirus, Type species: rice stripe virus,
Genus: Tobamovirus, Type species: tobacco mosaic virus,
Genus: Tobravirus, Type species: tobacco rattle virus,
Genus: Trichovirus, Type species: apple chlorotic leaf spot virus; Genus: Tymovirus, Type species: turnip yellow mosaic virus; Genus: Umbravirus, Type species: carrot mottle virus;
Negative ssRNA Viruses: Order: Mononegavirales, Family: Rhabdoviridae, Genus: Cytorhabdovirus, Type Species: lettuce necrotic yellows virus, Genus: Nucleorhabdovirus, Type species: potato yellow dwarf virus;
Negative ssRNA Viruses: Family: Bunyaviridae, Genus: Tospovirus, Type species tomato spotted wilt virus;
dsRNA Viruses: Family: Partitiviridae, Genus: Alphacryptovirus, Type species: white clover cryptic virus 1, Genus: Betacryptovirus, Type species: white clover cryptic virus 2, Family: Reoviridae, Genus: Fijivirus, Type species: Fiji disease virus, Genus: Phytoreovirus, Type species: wound tumor virus, Genus: Oryzavirus, Type species: rice ragged stunt virus;
Unassigned Viruses:
Genome: ssRNA, Species Garlic viruses A,B,C,D, Species grapevine fleck virus, Species maize white line mosaic virus, Species olive latent virus 2, Species: ourmia melon virus, Species Pelargonium zonate spot virus.
Sizes and shapes of selected viruses are as follows.
Rod shaped viruses—TMV: the virions have ≈300 nm in length and ≈18 nm in diameter; PVX (filamentous; usually flexuous; with a clear modal length): 515 nm long and 13 nm in diameter; Brome Mosaic Virus: 26 nm in diameter.
Symmetry/shape—icosahedral
Alfalfa mosaic virus (Nucleocapsids bacilliform, or quasi-isometric elongated): 35 nm long (Tb) or 30 nm long; Ta that occurs either in bacilliform (Ta-b) or ellipsoidal (Ta-t) shape) with no clear modal length: 56 nm long (B); 43 nm long (M); 18 nm in diameter.
Preferred viruses are plant viruses having a single-stranded plus-sense RNA genome. Other preferred viruses are plant viruses having rod-shaped viral particles.
The viruses (tobacco mosaic virus and potato virus X) used in the examples were predominantly chosen because of the ready availability of well-established expression systems for said viruses (Donson et al., 1991, Proc Natl Acad Sci USA, 88:7204-7208; Shivprasad et al., 1999, Virology, 255:312-323; Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857; Marillonnet et al., 2005, Nat Biotechnol, 23:718-723; Chapman, Kavanagh & Baulcombe, 1992, Plant J., 2:549-557; Baulcombe, Chapman & Santa Cruz, 1995, Plant J., 7:1045-1053; Angell & Baulcombe, 1997, EMBO J., 16:3675-3684) including the very recently developed system for expression of hetero-oligomeric proteins (EP Application No. 05 001 819.1; WO 2006/079546). Other plant viruses including DNA viruses also can be used for practicing this invention (for reviews please refer to: Mullineaux et al., 1992, Genetic Engineering in Plant Viruses, CRC Press Inc., pp 187-215; Timmermans et al., 1994, Ann. Rev. Plant Physiol. Plant Mol. Biol., 45:79-112; Porta & Lomonossoff, 2002, Biotechnol. Genet. Engineering Rev., 19:245-291).
We have surprisingly found that (a) flexible peptide linker(s) (either unable to form secondary structure or capable of forming a helical secondary structure) allows overcoming size restrictions in generating translational fusion of recombinant protein with plant viral coat protein. Said linker peptide presumably removes or significantly reduces a negative effect of fusion partners on each other's functionality. The linker peptides used in this invention may be flexible peptide linkers such as (GGGGS)n or helix-forming peptide linkers such as (EAAAK)n, wherein n may be 2-5. The peptide linkers are segments of said fusion protein. The use of this type of peptide linkers in fusion proteins of different proteins and their effect on the function of fusion proteins is described (Arai et al., 2001, Protein Eng., 14:529-532; Arai et al., 2004, Proteins, 57:829-838). In example 1, we describe the design of constructs containing linker peptides with n=3 (see also
The recombinant plant viral particles or plant virus-like particles can be produced by expressing said fusion protein of the invention. In one embodiment, said fusion protein is expressed in plant cells or plants using plant viral vectors. In such plant viral vectors, the coat protein of the virus from which the viral vector is derived may be replaced by a polynucleotide encoding said fusion protein of the invention.
Plant viral vectors are efficient tools for transient high yield expression of recombinant proteins such as the fusion protein of the invention in plants (for review see: Porta & Lomonossoff, 1996, Mol. Biotechnol., 5, 209-221; Yusibov et al., 1999, Curr. Top. Microbiol. Immunol., 240, 81-94; Gleba et al., 2004, Curr Opin Plant Biol. 7:182-188; Gleba et al., 2005, Vaccine, 23:2042-2048). Viral vector-based expression systems offer a significantly higher yield of transgene product (such as the fusion protein of the invention) compared to plant nuclear transgenes. For example, the level of recombinant protein can reach 5-50% of the total cellular plant protein content, when expressed from a viral vector (Kumagai et al., 2000, Gene, 245, 169-174; Shivprasad et al., 1999, Virology, 255, 312-323; Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857; Marillonnet et al., 2005, Nat Biotechnol, 23:718-723). There are several published patents which describe viral vectors suitable for systemic expression of the fusion protein of the invention in plants (U.S. Pat. No. 5,316,931; U.S. Pat. No. 5,589,367; U.S. Pat. No. 5,866,785). In general, these vectors can express a foreign gene as a translational fusion with a viral protein (U.S. Pat. No. 5,491,076; U.S. Pat. No. 5,977,438), from an additional subgenomic promoter (U.S. Pat. No. 5,466,788; U.S. Pat. No. 5,670,353; U.S. Pat. No. 5,866,785), or from polycistronic viral RNA using IRES elements for independent protein translation (WO0229068). Other systems (WO2005049839) rely on agrobacteria for systemic delivery of viral replicon and do have significantly higher capacity for the size of a foreign gene compared to systemic viral vectors.
In example 2 we describe the production and analysis of viral particles displaying different recombinant proteins of different sizes on its surface. The electrophoretic analysis of different viral CP-recombinant protein fusions expressed with the help of a viral vector is shown in
Another embodiment of this invention demonstrates the functionality of a recombinant protein displayed on the surface of a viral particle in biotechnology applications. We have chosen fusion proteins comprising the domains E and D of protein A (133 amino acid residues, see
The surface of viral particles displaying protein A may serve as a high-affinity ligand suitable for purification of immunoglobulins. The relatively high molecular weight of virus particles allows their used as an affinity matrix and to develop simple procedures for purifying immunoglobulins that may be bound to the recombinant viral particles of the invention (or other protein of interest).
In addition, the recombinant viral particles can be further polymerized by cross-linking, yielding even higher molecular weight structures that are suitable for serving as an affinity matrix e.g. in protein purification procedures (Smith, Petrenko & Matthew, 1998, J. Immunol. Methods, 215:151-161). Another method of cross-linking viral particles can be by forming disulfide bridges between modified (cystein-added) coat proteins of different viral particles (Wang et al. 2002, Chem. Biol, 9: 813-819). This method also allows to inactivate viral particles, preventing viral vectors from replication. Additionally, various cross-linking agents can be used for inactivating viral particles, such as but not limited to formaldehyde (Barteling & Cassim, 2004, Dev. Biol., 119:449-455), ethyleneimine, N-acetylethyleneimine (Burrage et al., 2000, Vaccine, 18:2454-2461), UV irradiation (Freitas et al., 2003, J Virol Methods., 108:205-11) and other approaches.
Viruses, whether naturally occurring wild-type or mutant viruses or genetically engineered viral vectors are self-replicating and as such are very inexpensive. Plant viral particles are also much larger than the great majority of proteins or small molecules for which purification procedures are required. The great difference in molecular weight or in physico-chemical properties can be effectively exploited to separate a protein or non-proteinaceous compound of interest from a mixture such as a tissue homogenate by binding the protein of interest to a virus particle according to the invention and then separating the resultant complex from the rest of the mixture. The association between the viral particle and the molecule of interest can later be dissolved in a number of ways known to those skilled in the art. In one embodiment of our invention, we demonstrate isolation of IgG from a plant extract using recombinant viral particles displaying IgG-binding domains on its surface. Viral particles displaying IgG binding domains of protein A were produced and isolated as described in Example 2. After evaluation of their binding capacity (Example 3,
This invention also allows generating and utilizing recombinant viral particles having on their surface more than one (two or more) types of recombinant proteins, thus creating complex structures on the surface of plant viral particles. This may be achieved e.g. by using a recently developed plant virus-based expression system permitting to express more than one fusion of CP and a recombinant protein of interest with high yield (EP Application No. 05 001 819.1; WO 2006/079546). For example, two recombinant fusion proteins in roughly equimolar amounts can be expressed and assembled in viral particles using the invention. Alternatively, if different fusion proteins are required in a molar ratio other than equimolar, one of said fusion proteins could be expressed from a standard (e.g. driven by 35S promoter) expression cassette either transiently, or from a vector stably incorporated into plant chromosomal DNA, while the other fusion protein could be expressed from a viral vector. In yet another approach, viral particles can be reconstructed in vitro by mixing different recombinant viral particles in required proportions, deconstructing them by changing pH and/or ionic strength of the solution, and then reassembling them de novo, thus producing a different type of viral particles with different recombinant proteins on their surface. A schematic representation of a plant viral particle displaying more than one recombinant protein is shown on the left, bottom, of
Preferably, the viral particles of the invention are produced (expressed) in plants, as plants are practically free of human and animal pathogens, thus reducing the danger of infection by using viral particles isolated from viral or bacterial source. The cost of producing viral particles in plants, plant tissue or plant cells will be significantly lower compared to viral particles produced by an animal or bacterial source. In principle, however, the method may be practiced using a wide variety of host expression systems including plants (including cell and tissue cultures thereof), animals including non-human animal organisms, and animal and human cell cultures, fungi, bacteria and yeast.
The present method of purifying proteins of interest can be practiced in many different ways depending on several factors such as the nature of the protein of interest to be purified relative to the host and the manner in which the protein is produced in the host and the nature of the affinity between the virus particle and the protein. In embodiments where the protein of interest or small molecular compound to be purified is produced endogenously (naturally) by a host, the host may be cultured and lysed. The lysate or a refined solution thereof containing the protein of interest may be contacted with the affinity matrix comprising the recombinant viral particles of the invention. Purification of proteins that are not produced endogenously by a host typically requires a genetic manipulation in order to supply the host with the machinery i.e., at least one transgene that encodes the protein of interest to be purified. In these embodiments, the transgene(s) may be introduced into a host as part of the viral expression/replication vector, or via a separate transformation event. The affinity of the recombinant protein displayed on the surface of the affinity matrix for the protein of interest produced by the host may be direct or indirect in the sense that the transgene may encode the protein of interest in the form of a fusion with a binding peptide that is recognized and bound by the affinity protein on the viral particles. Thus, the protein of interest may itself be a fusion protein.
In one embodiment, an exogenous (e.g., heterologous) protein of interest is expressed in a plant host (e.g., plant cells, tissue, homogenate or whole plant). This embodiment entails providing said viral particles displaying a recombinant protein as an affinity protein, wherein said recombinant protein has an affinity to the protein of interest to be purified.
Alternatively, said viral particle displays a recombinant protein that has an affinity to a small molecular compound to be purified. Therapeutic agents and herbicides are examples of such small molecular compounds. In general, any non-peptidic organic molecule produced by a host such as a plant, animal, bacterial or yeast cell, and that is recognizable by (e.g. has a binding affinity for) a recombinant protein displayed on the surface of plant viral particles may be isolated or detected in accordance with the present invention. The conditions employed for dissociating the plant viral particle from the protein (or small molecule) depends on the specific type of interactions and can be created by varying physico-chemical parameters e.g., pH; temperature; ions, chelating agents concentration, etc. Selecting appropriate conditions will be within the level of skill in the art of protein purification. Ultrafiltration is one such way of separating protein from the affinity matrix of said viral particles.
This invention is suitable for the purification of transgenic and endogenous proteins of interest alike as well as non-proteinaceous molecules occurring naturally or as a consequence of transgene expression in wide variety of hosts including but not limited to members of the plant, animal and bacterial kingdoms. Examples of such proteins can be, but not limited to pharmaceutically and industrially important proteins, e.g. immune response proteins, enzymes including DNA modifying enzymes, starch-, cell wall modifying enzymes, proteases, lipases etc.
In the case of proteins or small molecular compounds that are exogenous to the host, transgenes encoding the protein of interest (by itself or in the form of a fusion with a peptide that binds the recombinant protein on the virus particle) or the expression of which result in the production of the small molecule, are introduced into a non-human host in accordance with standard techniques. In general, these techniques may include stable or transient transformation or viral delivery (e.g., infection of the cell by the viral expression vector). Methods of creating transgenic organisms with stably integrated foreign genes are well described in the literature. For example, DNA can be transformed into plant cells via Agrobacterium-mediated delivery. See, U.S. Pat. Nos. 5,591,616; 4,940,838; and 5,464,763. Other methods include particle or microprojectile bombardment (U.S. Pat. No. 5,100,792; European Patent (EP) 444,882 B1; EP 434,616 B1), microinjection (WO 09209696; WO 09400583 A1; EP 175,966 B1) and electroporation (EP 564,595 B1; EP 290,395 B1; WO 08706614 A1). Procedures of transgene delivery into animal, bacterial and yeast cells are well established. A popular method of transgene delivery into animal cells is retrovirus-mediated (Robbins & Givizzani, 1998; Reynolds et al., 1999). Other methods with synthetic (non-viral) carriers are also suitable (for review see: Bown et al., 2001). Transformation methods for yeast and bacterial cells are well described in many manuals e.g., Yeast Protocol Handbook (2000) and Sambrook et al., (1989).
The present invention is well amenable to industrial application and scaling-up because it can accommodate techniques such as tissue homogenization, centrifugation and ultrafiltration. It can be applied to production of proteins and small molecules in any prokaryotic or eukaryotic system. Thus, the invention represents a universal, inexpensive and scale-up method of purification of any protein of interest from any kind of prokaryotic or eukaryotic system.
The recombinant viral particles of the invention can be of interest for applications in many different areas—not only in biotechnology, but also in nanotechnology and molecular electronics applications. Plant viruses are very convenient for such purposes, as they are easy to produce and isolate and provide for high yield (up to 10 g of viral particles per kilogram of fresh tobacco leaves). Also, the viral particles (virions) can be purified industrially using simple ‘low-tech’ protocols (Creager et al, 1999, Plant Cell, 11:301-308)
The following examples illustrate the invention further without limiting the scope of the invention.
The vectors used in the following examples are generally described in two recent publications (Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857; Marillonnet et al., 2005, Nat. Biotechnol., 23:718-723). The cDNA for Potato Virus X (PVX) was generated from PVX isolate PV-0014 received from DSMZ collection (http://www.dsmz.de) by RT-PCR and used for creating PVX provectors. The descriptions of PVX-based expression system are provided in numerous publications (Chapman, Kavanagh & Baulcombe, 1992, Plant J., 2:549-557; Baulcombe, Chapman & Santa Cruz, 1995, Plant J., 7:1045-1053; Angell & Baulcombe, 1997, EMBO J., 16:3675-3684).
The 3′-part of the TVCV Coat protein was amplified by PCR using primers cptv1 and cpfus4 or cpfus5 thus introducing a (GGGGS)3-linker or a (EAAAK)3-linker to the C-terminus of CP. The PCR products were cut with NcoI and BsaI and ligated into 5′-provector pICH20697 (
The CP with linkers was amplified by PCR with primers pv5 cptv and pv5p5r2 using pICH20701 or pICH20723 as template. PCR products were cloned as NheI-SacI fragments into PVX 5′-provector giving constructs pICH23407 and pICH23411 (
Protein A from Staphylococcus aureus contains five immunoglobulin (IgG) binding domains (
Streptactin is a mutant form of streptavidin with increased affinity towards StrepTag II (Voss S. & Skerra A. 1997. Prot Engin 10, 975-982). The 5′- and 3′-part of streptactin were amplified separately by PCR using primers streppr1 and streppr2 or streppr3 and streppr4 on genomic DNA from Streptomyces avidinii as template. PCR products encoding for protein fragment shown in
A mutant form of streptactin (V55T, T76R, L109T, V125R) that is supposed to be monomeric (Wu S C & Wong S L., 2005, J Biol Chem 280:23225-31) was engineered by site directed mutagensis with oligonucleotides streppr5-streppr12 leading to construct pICH23478 (
This tag was introduced into 3′-provector pICH21595 cut with XbaI, BsaI by adapter ligation with oligos streptag5 and streptag6. The resulting construct was named pICH23463 (
A number of other genes (GFP, DsRed, antigens, cytokines) was cloned in a similar way into 3′-provectors (Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857) yielding in the constructs pICH7410 (GFP), pICH10580 (DsRED) (
All constructs were electroporated into Agrobacterium tumefaciens GV3101. Agroinfiltrations of N. benthamiana plants were done essentially as described in Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857. Three agrobacterial strains containing 5′ provector encoding CP, 3′ provector encoding the recombinant protein and a source of a site-specific recombinase (pICH14011,
Analysis of CP Fusions Expressed in N. benthamiana Leaves
All recombinant protein fusions were extracted from infiltrated N. benthamiana leaves 6-11 days after infiltration and analysed by electrophoretic separation in polyacrylamide gels as previously described (Marillonnet et al., 2004, Proc Natl Acad Sci USA, 101:6852-6857; Marillonnet et al., 2005, Nat. Biotechnol., 23:718-723). The results of electrophoretic analysis of different CP-recombinant protein fusions are shown in
Specifically, the use of pICH20697 5′-provector (
Infiltrated leaves were homogenized in 0.1 M K-phosphate buffer, pH 7.0 (2-3 ml buffer/g FW) using a leaf juice press or mortar and pestle. Insoluble matter was removed by filtration through miracloth. Leaf juice was treated once with one volume of chloroform and viral particles were precipitated with polyethylenglycol (PEG-6000) using standard procedures (Turpen & Reinl, 1998, Methods in Biotechnol, 3:89-101, eds. C. Cunningham & A. J. R. Porter, Humana Press, Totowa, N.J.). Particles were resuspended in K-phosphate buffer and further purified by sucrose density centrifugation. The samples of viral particles containing CP fused with protein A fragment were analyzed by SDS-electrophoresis and by electronic microscopy. The samples for electron microscopy were prepared as described by Negrouk and colleagues (2004, Analytical Biochem., 333: 230-235). The results of analysis are shown in
Aliquots (0.5 mg) of purified viral particles having bound recombinant protein A were mixed with different amounts of human IgG (Sigma 14506), incubated on ice for 1 hour and precipitated by centrifugation (10 min, 12.000 g). Pellets and supernatants were analysed by SDS-PAGE (
Antibodies were expressed in planta using ICONs viral expression system for production of hetero-oligomeric proteins in plants (EP Application No. 05 001 819.1; WO 2006/079546). Leaf material containing monoclonal antibodies of the IgG class was ground in liquid nitrogen and extracted with 3 ml of PBS (Sambrook, Fritsch & Maniatis, 1989, Molecular Cloning: a Laboratory Manual, CSH, NY) per gram of fresh leaf weight (FW). Insoluble material was removed by two rounds of centrifugation (10 min, 16000 g). One hundred milligram of viral particles displaying protein A was added per one ml of plant extract and samples were incubated on ice for at least 1 hour. Antibodies bound to viral particles were precipitated by centrifugation (15 min, 12000 g) and resuspended in 0.25 volumes 0.1 M glycine pH 2.5. In order to remove viral particles, samples were adjusted to 1% NaCl, 4% PEG, incubated 30 min on ice and centrifuged 15 min at 10.000 g. Antibody-containing supernatants were transferred to fresh tubes, neutralized with 1/10 volume 1 M Tris/HCl pH 9.0 and adjusted to 14% PEG by adding an appropriate volume of 25% PEG-solution in PBS buffer. Samples were kept on ice for at least 1 hour and antibodies were precipitated by centrifugation (15 min, 16000 g). Summary of purification protocol is shown in Table 1. Antibodies were dissolved in a convenient volume of PBS and analyzed by gel-electrophoresis. An electrophoretic analysis of proteins from the purification procedure is shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 05 020 311.6 | Sep 2005 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2006/009029 | 9/15/2006 | WO | 00 | 6/26/2008 |
