Method to improve virus removal in protein purification

Abstract

To provide methods for increasing the filtration capacity of virus filters, by combined use of endotoxin removal and cation-exchange media in the prefiltration process.SOLUTION: The present invention provides a method of improving the filtration capacity of a virus filter during protein purification, comprising subjecting a composition comprising a protein to be purified to a cation exchange step and an endotoxin removal step, simultaneously or in either order, prior to passing through the virus filter.SELECTED DRAWING: None

Description

[Technical field]

[0001] The present invention is from the field of protein purification. In particular, the present invention relates to a method for increasing the filtration capacity of a virus filter by combining endotoxin removal in a pre-filtration step with a cation exchange medium. [Background technology]

[0002] Mammalian cell lines have become the first choice for the production of recombinant protein therapeutics due to their capacity for proper protein folding and post-translational modifications, such as glycosylation (Chu and Robinson Current Opinion in Biotechnology 12:180-187, 2001). However, these cell lines are also known to contain retrovirus-like particles (Lieber et al. Science 182:56-59, 1973; Lubiniecki et al. Dev Biol Stand 70:187-191, 1989) and pose a potential risk of adenovirus contamination (Garnick, Dev Biol Stand. Basel: Karger 93:21-29, 1998). While the biopharmaceutical industry, which produces recombinant protein drugs, has a good safety record, there have been cases of viral infections from blood and plasma-derived blood products in the past (Brown, Dev. Biol. Stand. 81, 1993; Thomas, Lancet 343:1583-1584, 1994). To mitigate the risk of viral contamination during recombinant protein production, downstream purification processes are designed to include process steps that remove endogenous and exogenous viruses. Adequate viral clearance is achieved by a combination of several process steps that provide either viral inactivation or viral removal from the process feedstream. Viral inactivation is achieved using techniques such as incubation at low pH, heat treatment, and detergents, while viral removal is typically performed using chromatography and filtration (Curtis et al., Biotechnology and Bioengineering 84(2):179-186, 2003).

[0003] Unlike chromatographic media, which remove viruses based on physicochemical properties such as effective charge, viral filtration removes viruses by size exclusion and is therefore considered a more powerful technique. So far, the use of viral filtration during downstream purification of mammalian cell culture-derived biotherapeutics has been limited to the removal of retroviruses (80-100 nm in diameter) due to the lack of high-throughput membranes with nominal pore sizes below 60 nm.

[0004] Recent advances in membrane technology have allowed the production of high-throughput membranes with nominal pore sizes of 20 nm. These virus filters are retentive to parvoviruses (18-26 nm in diameter) and allow the passage of proteins as large as 160 kD (approximately 8 nm), such as monoclonal antibodies (mAbs).

[0005] The high selectivity and high throughput associated with parvovirus filters are achieved by casting a thin retentive membrane on a microporous substrate. Although the thin retentive layer allows for very precise separation of proteins and viruses, it is also susceptible to fouling by impurities in the process feed stream, reducing filter capacity and flow rate. Fouling of virus filters is due to contaminants such as protein aggregates and denatured proteins. Bohonak and Zydney (Bohonak and Zydney, Journal of Membrane Science 254(1-2):71-79, 2005) showed that the loss of filter capacity can be due to cake formation or pore blocking. Other recent reports (Bolton et al., Biotechnol. Appl. Biochem. 43:55-63, 2006; Levy et al., Filtration in the Biopharamaceutical Industry. (Meltzer, TH and Jornitz, MW, eds.) pp. 619-646, Marcel Dekker, New York, 1998) attributed filter fouling to adsorption of impurities to the pore walls. Several publications (Bolton et al., Biotechnology and Applied Biochemistry 42:133-142, 2005; Hirasaki et al., Polymer Journal 26(11):1244-1256, 1994; Omar and Kempf, Transfusion 42(8):1005-1010, 2002) have also demonstrated that reduced filter capacity or pore blockage can reduce virus retention by several orders of magnitude, affecting the robustness of unit operations.

[0006] Thus, much recent research has focused on identifying pre-filters to remove contaminants from process feed streams, minimize viral filter fouling, and ensure high capacity, high throughput, and robust viral retention. Bolton et al. (2006) conducted an extensive study that included testing several membranes as pre-filters and demonstrated that it is possible to increase the capacity of normal flow parvovirus (NFP) membranes by almost an order of magnitude by using Viresolve™ depth filters as pre-filters. Brown et al. (2008, Use of Charged Membranes to Identify Soluble Protein Foulants in order to Facilitate Parvovirus Filtration. IBC's 20 th (Antibody Development and Production, San Diego, CA) evaluated a strong cation exchange membrane adsorber as a pre-filter for a parvovirus-retentive filter and showed that the capacity of the viral filter could be increased several-fold for 11 different mAb streams. The authors hypothesized that the cation exchange membrane adsorber would remove large molecular weight (approximately 600-1500 kD) protein aggregates from the feed stream by competitive adsorption, preventing the viral filter from clogging. U.S. Patent No. 7,118,675 (Siwak et al.) describes a process that utilizes charge-modified membranes to remove protein aggregates from protein solutions to prevent fouling of viral filters. Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention is based, at least in part, on the experimental findings that fouling of parvovirus filters may be due to impurities other than those described in the literature, and that a more comprehensive prefiltration solution is required to improve virus filtration capacity. Thus, the present invention provides a novel prefiltration solution that performs significantly better than the best prefiltration approach described in the literature (cation exchange membrane adsorber).

[0008] In one aspect, the invention relates to a method for improving the filtration capacity of a virus filter during protein purification, comprising passing a composition containing the protein to be purified through said virus filter after subjecting it to a cation exchange step and an endotoxin removal step, in either order.

[0009] In one embodiment, the virus filter has a pore size of about 15 to about 100 nm in diameter.

[0010] In another embodiment, the virus filter has a pore size of about 15 to about 30 nm in diameter.

[0011] In yet another embodiment, the pore size of the virus filter is about 20 nm.

[0012] In a further embodiment, the virus to be cleared is a parvovirus.

[0013] In a further embodiment, the parvovirus has a diameter of about 18 to about 26 nm.

[0014] In a different embodiment, the protein is an antibody or antibody fragment, such as an antibody or fragment thereof produced by recombinant DNA techniques.

[0015] In another embodiment, the antibody is a therapeutic antibody.

[0016] In yet another embodiment, the recombinant antibody or antibody fragment is produced in mammalian host cells, such as Chinese hamster ovary (CHO) cells.

[0017] In a further embodiment, the composition comprising the protein to be purified is first subjected to a cation exchange step, followed by an endotoxin removal step, followed by viral filtration.

[0018] In a further embodiment, the composition comprising the protein to be purified is first subjected to an endotoxin removal step, followed by a cation exchange step, followed by virus filtration.

[0019] In another embodiment, a composition containing the protein to be purified is subjected to a cation exchange step and an endotoxin removal step simultaneously, by keeping the two media together in a single module, followed by viral filtration.

[0020] In yet another embodiment, the endotoxin removal step is immediately followed by viral filtration.

[0021] In a further embodiment, the cation exchange step is immediately followed by viral filtration.

[0022] In a different embodiment, the viral filtration is carried out at a pH of about 4 to about 10.

[0023] In another embodiment, the protein concentration in the composition to be purified is about 1-40 g / L.

[0024] In yet another embodiment, the antibody to be purified is against one or more antigens selected from the group consisting of HER1 (EGFR), HER2, HER3, HER4, VEGF, CD20, CD22, CD11a, CD11b, CD11c, CD18, ICAM, VLA-4, VCAM, IL-17A and / or F, IgE, DR5, CD40, Apo2L / TRAIL, EGFL7, NRP1, mitogen-activated protein kinase (MAPK), and Factor D.

[0025] In further embodiments, the antibody is an anti-estrogen receptor antibody, an anti-progesterone receptor antibody, an anti-p53 antibody, an anti-cathepsin D antibody, an anti-Bcl-2 antibody, an anti-E-cadherin antibody, an anti-CA125 antibody, an anti-CA15-3 antibody, an anti-CA19-9 antibody, an anti-c-erbB-2 antibody, an anti-P-glycoprotein antibody, an anti-CEA antibody, an anti-retinoblastoma protein antibody, an anti-ras oncoprotein antibody, an anti-Lewis X antibody, an anti-Ki-67 antibody, an anti-PCNA antibody, an anti-CD3 antibody, an anti-CD4 antibody, an anti-CD5 antibody, an anti-CD7 antibody, an anti-CD8 antibody, an anti-CD9 / p24 antibody, an anti-CD10 antibody, an anti-CD11c antibody, an anti-CD13 antibody, an anti-CD14 antibody, an anti-CD15 antibody, an anti-CD19 antibody, an anti-CD2 3 antibody, anti-CD30 antibody, anti-CD31 antibody, anti-CD33 antibody, anti-CD34 antibody, anti-CD35 antibody, anti-CD38 antibody, anti-CD41 antibody, anti-LCA / CD45 antibody, anti-CD45RO antibody, anti-CD45RA antibody, anti-CD39 antibody, anti-CD100 antibody, anti-CD95 / Fas antibody, anti-CD99 antibody, anti-CD106 antibody, anti-ubiquitin antibody, anti-CD71 antibody, anti-c-myc antibody, anti-cytokeratin antibody, anti-vimentin antibody, anti-HPV protein antibody, anti-kappa light chain antibody, anti-lambda light chain antibody, anti-melanosome antibody, anti-prostate specific antigen antibody, anti-S-100 antibody, anti-tau antigen antibody, anti-fibrin antibody, anti-keratin antibody and anti-Tn-antigen antibody. [Brief description of the drawings]

[0026] [Figure 1] FIG. 1 is a schematic diagram of the experimental setup used for viral filtration testing. [Diagram 2] FIG. 1 shows the effect of sterilization and depth filters on the capacity of Viresolve Pro parvovirus-retentive filters. Experiments were performed at pH 5.5 and conductivity 8.5 mS / cm. mAb concentration was approximately 13 g / L. [Figure 3a] Figure 3 shows the effect of a cation exchange and endotoxin removal membrane adsorber as a pre-filter on the capacity of a Viresolve Pro parvovirus filter. Data in 3(a) and 3(b) were obtained for MAb1 at pH 5.0 and 6.5, respectively. [Figure 3b]Figure 3 shows the effect of a cation exchange and endotoxin removal membrane adsorber as a pre-filter on the capacity of a Viresolve Pro parvovirus filter. Data in 3(a) and 3(b) were obtained for MAb1 at pH 5.0 and 6.5, respectively. [Figure 4a] Figure 4 shows the effect of a novel prefiltration train containing both a cation exchange and endotoxin removal membrane adsorber on the capacity of a Viresolve Pro parvovirus-retentive filter for MAb1. Data in 4(a) and 4(b) were obtained at pH 5.0 and 6.5, respectively. [Figure 4b] Figure 4 shows the effect of a novel prefiltration train containing both a cation exchange and endotoxin removal membrane adsorber on the capacity of a Viresolve Pro parvovirus-retentive filter for MAb1. Data in 4(a) and 4(b) were obtained at pH 5.0 and 6.5, respectively. [Diagram 5] FIG. 1 shows the effect of a novel prefiltration train containing both a cation exchange and endotoxin removal membrane adsorber compared to a cation exchange prefiltration medium on the capacity of a parvovirus retentive filter for MAb2. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] I. Definition By "protein" is meant a sequence of amino acids whose chain length is sufficient to create a high level of tertiary and / or quaternary structure. Proteins are thus distinguished from "peptides," which are also amino acid-based molecules but lack such structure. Typically, proteins for use herein have a molecular weight of at least about 15-20 kD, preferably at least about 20 kD.

[0028] Examples of proteins encompassed within the definition herein include mammalian proteins such as CD4, integrins and their subunits, e.g., beta 7, growth hormones, e.g., human growth hormone and bovine growth hormone; growth hormone releasing factor; parathyroid hormone; thyroid stimulating hormone; lipoproteins; alpha-1-antitrypsin; insulin A-chain; insulin B-chain; proinsulin; follicle stimulating hormone; calcitonin; luteinizing hormone; glucagon; clotting factors, e.g., factor VIIIC, factor IX, tissue factor, and von Willebrand factor; anticoagulants, e.g., protein C; atrial natriuretic factor; pulmonary surfactant; plasminogen activators, e.g., urokinase or tissue-type plasminogen activator (t-PA, e.g., Actibase®, TNKase®, Retebase®); bombadin; thrombin; tumor necrosis factors alpha and beta; enkephalinase; RANTES (regulated on activation normally T-Cell expressed and secreted (regulated upon activation and expressed and secreted by normal T cells); human macrophage inflammatory protein (MIP-1-α); serum albumin, e.g. human serum albumin; Müllerian inhibitory substance; mouse gonadotropin-related peptide; DNase; inhibin; activin; vascular endothelial growth factor (VEGF); receptors for IgE, hormones or growth factors; integrins; protein A or D; rheumatoid factor; neurotrophic factors, e.g. bone-derived neurotrophic factor (BDNF), neurotrophin-3, -4, -5 or -6 (NT-3, NT-4, NT-5 or NT-6), or nerve growth factors, such as NGF-β; platelet-derived growth factor (PDGF); fibroblast growth factors, such as aFGF and bFGF; epidermal growth factor (EGF); transforming growth factors (TGFs), such as TGF-α and TGF-β, such as TGF-β1, TGF-β2, TGF-β3, TGF-β4 or TGF-β5; insulin-like growth factors I and II (IGF-I and IGF-II); des(1-3)-IGF-I (brain IGF-I); insulin-like growth factor binding proteins; other CD proteins, such as CD3, CD8, CD19 and CD20;Erythropoietin (EPO); thrombopoietin (TPO); bone morphogenetic factors; immunotoxins; bone morphogenetic proteins (BMPs); interferons, such as interferon alpha, beta and gamma; colony stimulating factors (CSFs), such as M-CSF, GM-CSF and G-CSF; interleukins (ILs), such as IL-1 to IL-10; superoxide dismutase; T-cell receptors; surface membrane proteins; decay accelerating factors (DAFs); viral antigens, such as parts of the AIDS envelope; transport proteins; homing receptors; addressins; regulatory proteins; integrins, such as CD11a, CD11b, CD11c, CD18, ICAM, VLA-4 and VCAM; tumor associated antigens, such as the HER1 (EGFR), HER2, HER3 or HER4 receptors; apo2L / TRAIL, and fragments of any of the above polypeptides; as well as immunoadhesins and antibodies that bind any of the above proteins; and biologically active fragments or variants of any of the above proteins.

[0029] Specifically included within the definition of "protein" as defined herein are therapeutic antibodies and immunoadhesins, such as, but not limited to, antibodies against one or more of the following antigens: HER1 (EGFR), HER2, HER3, HER4, VEGF, CD20, CD22, CD11a, CD11b, CD11c, CD18, an ICAM, VLA-4, VCAM, IL-17A and / or F, IgE, DR5, CD40, Apo2L / TRAIL, EGFL7, NRP1, mitogen-activated protein kinase (MAPK) and Factor D, and fragments thereof.

[0030] Examples of other antibodies include, but are not limited to, those selected from the following: an anti-estrogen receptor antibody, an anti-progesterone receptor antibody, an anti-p53 antibody, an anti-cathepsin D antibody, an anti-Bcl-2 antibody, an anti-E-cadherin antibody, an anti-CA125 antibody, an anti-CA15-3 antibody, an anti-CA19-9 antibody, an anti-c-erbB-2 antibody, an anti-P-glycoprotein antibody, an anti-CEA antibody, an anti-retinoblastoma protein antibody, an anti-ras oncoprotein antibody, an anti-Lewis X antibody, an anti-Ki-67 antibody, an anti-PCNA antibody, an anti-CD3 antibody, an anti-CD4 antibody, an anti-CD5 antibody, an anti-CD7 antibody, an anti-CD8 antibody, an anti-CD9 / p24 antibody, an anti-CD10 antibody, an anti-CD11c antibody, an anti-CD13 antibody, an anti-CD14 antibody, an anti-CD15 antibody, an anti-CD16 antibody, an anti-CD17 antibody, an anti-CD18 antibody, an anti-CD19 antibody, an anti-CD20 antibody, an anti-CD21 antibody, an anti-CD22 antibody, an anti-CD23 antibody, an anti-CD24 antibody, an anti-CD25 antibody, an anti-CD26 antibody, an anti-CD27 antibody, an anti-CD28 antibody, an anti-CD29 antibody, an anti-CD30 antibody, an anti-CD40 antibody, an anti-CD50 antibody, an anti-CD29 antibody, an anti-CD31 antibody, an anti-CD32 antibody, an anti-CD33 antibody, an anti-CD34 antibody, an anti-CD35 antibody, an anti-CD36 antibody, an anti-CD37 antibody, an anti-CD38 antibody, an anti-CD39 antibody, an anti-CD39 antibody, an anti-CD31 antibody, an anti-CD32 antibody, an anti-CD33 antibody CD15 antibody, anti-CD19 antibody, anti-CD23 antibody, anti-CD30 antibody, anti-CD31 antibody, anti-CD33 antibody, anti-CD34 antibody, anti-CD35 antibody, anti-CD38 antibody, anti-CD41 antibody, anti-LCA / CD45 antibody, anti-CD45RO antibody, anti-CD45RA antibody, anti-CD39 antibody, anti-CD100 antibody, anti-CD95 / Fas antibody, anti-CD99 antibody, anti-CD106 antibody, anti-ubiquitin antibody, anti-CD71 antibody, anti-c-myc antibody, anti-cytokeratin antibody, anti-vimentin antibody, anti-HPV protein antibody, anti-kappa light chain antibody, anti-lambda light chain antibody, anti-melanosome antibody, anti-prostate specific antigen antibody, anti-S-100 antibody, anti-tau antigen antibody, anti-fibrin antibody, anti-keratin antibody and anti-Tn-antigen antibody.

[0031] An "isolated" protein, such as an antibody, is one that has been identified, separated, and / or recovered from a component of its natural environment. Contaminating components of its natural environment are substances that would interfere with diagnostic or therapeutic uses for the protein, such as an antibody, and may include, for example, enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. In a preferred embodiment, the protein, such as an antibody, is purified (1) to greater than 95% by weight, most preferably greater than 99% by weight, as determined by the Lowry method, (2) sufficiently to obtain at least 15 residues of N-terminal or internal amino acid sequence by use of a spinning cup sequenator, or (3) to homogeneity by SDS-PAGE under reducing or non-reducing conditions using Coomassie blue, preferably silver staining.

[0032] The protein is preferably essentially pure and desirably essentially homogeneous (i.e., free from contaminating proteins). By "essentially pure" protein is meant a composition that comprises at least about 90% by weight, preferably at least about 95% by weight, of the protein, based on the total weight of the composition.

[0033] "Essentially homogeneous" protein means a composition that contains at least about 99% by weight of protein, based on total weight of the composition.

[0034] The term "antibody" is used in the broadest sense and specifically includes monoclonal antibodies (e.g., full-length antibodies having an immunoglobulin Fc region), antibody compositions with polyepitopic specificity, bispecific antibodies, diabodies and single-chain molecules, as well as antibody fragments (e.g., Fab, F(ab')2 and Fv).

[0035] The basic four-chain antibody unit is a heterotetrameric glycoprotein composed of two identical light (L) chains and two identical heavy (H) chains. IgM antibodies consist of five basic heterotetrameric units with an additional polypeptide called the J chain and contain 10 antigen-binding sites, whereas IgA antibodies consist of two to five basic four-chain units, which can polymerize and combine with the J chain to form multivalent aggregates. For IgG, the four-chain unit is generally about 150,000 daltons. Each L chain is linked to a H chain by one covalent disulfide bond, whereas the two H chains are linked to each other by one or more disulfide bonds, depending on the H chain isotype. The H and L chains each also have regularly spaced intrachain disulfide bridges. Each H chain contains a variable domain (V) at the N-terminus. H ), followed by three constant domains in each of the α and γ chains (C H ) and four Cs for the μ and ε isotypes H Each L chain has a variable domain (V L ) followed by a constant domain at the other end. L is V HAligned with C L The first constant domain of the heavy chain (C H 1). Particular amino acid residues are believed to form an interface between the light and heavy chain variable domains. H and V L Together, these pairings form a single antigen-binding site. For the structure and properties of different classes of antibodies, see, e.g., Basic and Clinical Immunology, 8th Edition, Daniel P. Sties, Abba I. Terr and Tristram G. Parsolw (eds), Appleton & Lange, Norwalk, CT, 1994, page 71 and Chapter 6.

[0036] L chains from any vertebrate species can be assigned to one of two clearly distinct types, called kappa and lambda, based on the amino acid sequence of their constant domains. Depending on the amino acid sequence of the constant domain (CH) of their heavy chains, immunoglobulins can be assigned to different classes or isotypes. There are five classes of immunoglobulins, IgA, IgD, IgE, IgG, and IgM, with heavy chains called α, δ, ε, γ, and μ, respectively. The γ and μ classes are further divided into subclasses based on relatively minor differences in CH sequence and function; for example, humans express the following subclasses: IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2.

[0037] The term "variable" refers to the fact that certain segments of the variable domains differ extensively in sequence among antibodies. The V domains mediate antigen binding and define the specificity of a particular antibody for its particular antigen. However, variability is not evenly distributed throughout the entire span of the variable domain. Instead, V regions consist of relatively invariant stretches called framework regions (FRs) of about 15-30 amino acid residues separated by short regions of very high variability called "hypervariable regions", each about 9-12 amino acid residues long, sometimes called "complementarity determining regions" (CDRs). Natural heavy and light chain variable domains each contain four FRs (mostly in a β-sheet configuration and connected by three hypervariable regions), which form loops that connect, and in some cases form part of, the β-sheet structure. The hypervariable regions in each chain are held together in close proximity by the FRs and, together with the hypervariable regions from the other chain, contribute to the formation of the antigen-binding site of antibodies (see Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991)). The constant domains are not directly involved in binding the antibody to an antigen, but exhibit various effector functions, such as involvement in antibody-dependent cellular cytotoxicity (ADCC).

[0038] The term "hypervariable region" (also known as "complementarity determining region" or CDR), as used herein, refers to the amino acid residues of an antibody (usually three or four short regions of very high sequence variability) within the V-region domains of immunoglobulins that form the antigen-binding site and are the primary determinants of antigen specificity. There are at least two methods for identifying CDR residues: (1) an approach based on cross-species sequence variability (i.e., Kabat et al., Sequences of Proteins of Immunological Interest (National Institute of Health, Bethesda, MS 1991); and (2) an approach based on crystallographic studies of antigen-antibody complexes (Chothia, C. et al., J. Mol. Biol. 196 : 901-917 (1987). However, so long as two residue specification techniques define areas of overlap but not identical regions, they can be combined to define a hybrid CDR.

[0039] The term "monoclonal antibody" as used herein refers to an antibody obtained from a population of substantially homogeneous antibodies, i.e., the individual antibodies comprising the population are identical except for possible natural mutations that may be present in minor amounts. Monoclonal antibodies are highly specific, being directed against a single antigenic site. Furthermore, unlike conventional (polyclonal) antibody preparations, which typically include a number of different antibodies directed against a number of different determinants (epitopes), each monoclonal antibody is directed against a single determinant on the antigen. In addition to their specificity, monoclonal antibodies are advantageous in that they are synthesized by a hybridoma culture uncontaminated by other immunoglobulins. The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies and is not intended to require production of the antibody by any particular method. For example, monoclonal antibodies to be used in accordance with the present invention may be prepared according to the methods described in Kohler et al., Nature, 2003, 144:1311-1325, 1997, 19 ... 256: 495 (1975), or may be made by recombinant DNA methods (see, e.g., U.S. Pat. No. 4,816,567). "Monoclonal antibodies" may be made by the hybridoma method first described by Clackson et al., Nature, 352 :624-628 (1991) and Marks et al., J. Mol. Biol., 222 Antibody fragments may also be isolated from phage antibody libraries using the techniques described in U.S. Pat. No. 6,313,392 (1991).

[0040] Monoclonal antibodies herein specifically include "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical to or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain(s) is identical to or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies (so long as they exhibit the desired biological activity) (U.S. Pat. No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 2002). 81 :6851-6855 (1984).

[0041] An "intact" antibody comprises an antigen-binding site, as well as a C domain and at least a long domain C H 1. C H 2 and C H It includes 3.

[0042] An "antibody fragment" comprises a portion of an intact antibody, preferably the antigen-binding and / or variable region of the intact antibody. Examples of antibody fragments include Fab, Fab', F(ab')2 and Fv fragments; diabodies; linear antibodies (U.S. Pat. No. 5,641,870, Example 2; Zapata et al., Protein Eng. 8(10) : 1057-1062

[1995] ); single-chain antibody molecules and multispecific antibodies formed from antibody fragments.

[0043] Papain digestion of antibodies produces two identical antigen-binding fragments called "Fab" fragments, as well as a residual "Fc" fragment (a name reflecting its ability to crystallize readily). The Fab fragments contain the variable region domains of the heavy chains (V H ) and the first constant domain of one heavy chain (C H Each Fab fragment is monovalent with respect to antigen binding, i.e., it has a single antigen-binding site. Pepsin treatment of an antibody produces a single large F(ab')2 fragment, which roughly corresponds to two disulfide-linked Fab fragments with different antigen-binding activities and are still capable of cross-linking antigen. The Fab' fragment contains one or more cysteines from the antibody hinge region that are not included in the C(ab')2 fragment. H It differs from a Fab fragment by having 2-3 additional residues at the carboxy terminus of one domain. Fab'-SH is the designation used herein for Fab' in which the cysteine ​​residue(s) of the constant domains bear a free thiol group. F(ab')2 antibody fragments were originally produced as pairs of Fab' fragments which have hinge cysteines between them. Other chemical couplings of antibody fragments are also known.

[0044] The Fc fragment contains the carboxy-terminal portions of both H chains held together by disulfides. The effector functions of an antibody are determined by sequences in the Fc region, the region that is also recognized by Fc receptors (FcR) found on certain types of cells.

[0045] "Fv" is the smallest antibody fragment that contains a complete antigen recognition and binding site. This fragment consists of a dimer of one heavy and one light chain variable region domain in tight, non-covalent association. Folding of these two domains results in six hypervariable loops (three loops each from the H and L chain) that contribute amino acid residues for antigen binding and confer antigen specificity to the antibody. However, even a single variable domain (or half of an Fv containing only three CDRs specific for an antigen) has the ability to recognize and bind antigen, albeit with a lower affinity than the complete binding site.

[0046] A "single-chain Fv" (also abbreviated as "sFv" or "scFv") is an antibody fragment comprising the VH and VL antibody domains linked into a single polypeptide chain. Preferably, the sFv polypeptide further comprises a V H and V L A polypeptide linker is included between the domains, which allows the sFv to form the desired structure for antigen binding. For a discussion of sFv, see Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds., Springer-Verlag, New York, pp. 269-315 (1994).

[0047] The term "diabody" refers to a V domain in which interchain pairing of V domains is achieved rather than intrachain pairing, thereby resulting in a bivalent fragment, i.e., a fragment with two antigen-binding sites. H and V L This refers to small antibody fragments prepared by constructing sFv fragments (see previous paragraph) with a short linker (approximately 5-10 residues) between the domains. Bispecific diabodies are small antibody fragments prepared by constructing sFv fragments (see previous paragraph) with a short linker (approximately 5-10 residues) between the domains. H and V L Diabodies are heterodimers of two "crossover" sFv fragments whose domains are present on different polypeptide chains. Diabodies are described, for example, in EP 404,097; WO 93 / 11161; Hollinger et al., Proc. Natl. Acad. Sci. USA 90 : 6444-6448 (1993).

[0048] An antibody that "binds" to a molecular target or antigen of interest is one that can bind to that antigen with sufficient affinity such that the antibody is useful for targeting cells expressing the antigen.

[0049] An antibody that "specifically binds to" or is "specific for" a particular polypeptide or epitope on a particular polypeptide is one that binds to the particular polypeptide or epitope on a particular polypeptide without substantially binding to any other polypeptides or polypeptide epitopes. In such embodiments, the extent of binding of the antibody to these other polypeptides or polypeptide epitopes is less than 10% compared to binding to the target polypeptide or epitope, as determined by fluorescence activated cell sorting (FACS) analysis or radioimmunoprecipitation (RIA).

[0050] "Humanized" forms of non-human (e.g., murine) antibodies are chimeric immunoglobulins, immunoglobulin chains or fragments thereof (e.g., Fv, Fab, Fab', F(ab')2 or other antigen-binding sequences of antibodies) of mostly human sequence that contain minimal sequence derived from non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (recipient antibody) in which residues from a hypervariable region (including CDRs) of the recipient are replaced by residues from a hypervariable region of a non-human species (donor antibody) such as mouse, rat or rabbit having the desired specificity, affinity, and capacity. In some cases, Fv framework region (FR) residues of the human immunoglobulin are replaced by corresponding non-human residues. Furthermore, "humanized antibodies," as the term is used herein, can also include residues that are not found in the recipient antibody or in the donor antibody. These modifications are made to further refine and optimize antibody performance. Humanized antibodies optimally also comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. For further details, see Jones et al., Nature, 321 :522-525 (1986); Reichmann et al., Nature, 332 :323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2 :593-596 (1992).

[0051] Antibody "effector functions" refer to the biological activities attributable to the Fc region of an antibody (either a native sequence Fc region or an amino acid sequence variant Fc region) and vary with antibody isotype. Examples of antibody effector functions include: C1q binding and complement dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; down-regulation of cell surface receptors (e.g., B cell receptor); and B cell activation.

[0052] "Antibody-dependent cell-mediated cytotoxicity" or ADCC refers to a form of cytotoxicity in which secreted Ig bound onto Fc receptors (FcR) present on certain cytotoxic cells (e.g., natural killer (NK) cells, neutrophils, and macrophages) allows these cytotoxic effector cells to specifically bind to antigen-bearing target cells and subsequently kill the target cells with cytotoxins. Antibodies "arm" the cytotoxic cells and are required for target cell killing by this mechanism. NK cells, the primary cells for mediating ADCC, express only FcγRIII, whereas monocytes express FcγRI, FcγRII, and FcγRIII. Fc expression on hematopoietic cells has been reported by Ravetch and Kinet, Annu. Rev. Immunol. 9 : 457-92 (1991), page 464, Table 3. To assess the ADCC activity of the molecule, an in vitro ACDD assay, e.g., as described in U.S. Pat. No. 5,500,362 or 5,821,337, can be performed. Useful effector cells for such assays include peripheral blood mononuclear cells (PBMC) and natural killer (NK) cells. Alternatively, or additionally, the ADCC activity of the molecule can be measured in vivo, e.g., as described in Clynes et al., PNAS USA 95 :652-656 (1998).

[0053] "Fc receptor" or "FcR" refers to a receptor that binds to the Fc region of an antibody. A preferred FcR is a native sequence human FcR. Further preferred FcRs are those that bind IgG antibodies (gamma receptors) and include receptors of the FcγRI, FcγRII and FcγRIII subclasses, including allelic variants, and alternatively spliced ​​forms of these receptors, with FcγRII receptors including FcγRIIA ("activating receptors") and FcγRIIB ("inhibiting receptors"), which have similar amino acid sequences that differ primarily in their cytoplasmic domains. Activating receptor FcγRIIA contains an immunoreceptor tyrosine-based activation motif (ITAM) in its cytoplasmic domain. Inhibiting receptor FcγRIIB contains an immunoreceptor tyrosine-based inhibition motif (ITIM) in its cytoplasmic domain (M. Daeron, Annu. Rev. Immunol. 15 :203-234 (1997). FcRs are disclosed in Ravetch and Kinet, Annu. Rev. Immunol. 9 : 457-92 (1991);Capel et al., Immunomethods 4 : 25-34 (1994); and de Haas et al., J. Lab. Clin. Med. 126 : 330-41 (1995). Other FcRs, including those to be identified in the future, are encompassed by the term "FcR" herein. The term also includes FcRn, the neonatal receptor involved in the delivery of maternal IgG to the fetus (Guyer et al., J. Immunol. 117 : 587 (1976) and Kim et al., J. Immunol. 24 : 249 (1994).

[0054] "Human effector cells" are leukocytes that express one or more FcRs and perform effector functions. Preferably, the cells express at least FcγRIII and perform ADCC effector functions. Examples of human leukocytes that mediate ADCC include peripheral blood mononuclear cells (PBMCs), natural killer (NK) cells, monocytes, cytotoxic T cells and neutrophils, with PBMCs and MNK cells being preferred. Effector cells can be isolated from native sources, such as blood.

[0055] "Complement-dependent cytotoxicity" or "CDC" refers to the lysis of target cells in the presence of complement. Activation of the classical complement pathway is initiated by the binding of the first component of the complement system (C1q) to antibodies (of the appropriate subclass) that are bound to their cognate antigen. To assess complement activation, see, for example, Gazzano-Santoro et al., J. Immunol. Methods 202 : 163 (1996) can be performed.

[0056] The terms "conjugate," "conjugated," and "conjugation" refer to any and all types of covalent or non-covalent attachment, including, but not limited to, direct genetic or chemical fusion, coupling via a linker or cross-linking agent, and non-covalent association, for example, using leucine zippers. An antibody conjugate has another entity, such as a cytotoxic compound, drug, composition, chemical compound, radioactive element, or detectable label, attached to the antibody or antibody fragment.

[0057] "Treatment" refers to both therapeutic treatment and prophylactic or preventative measures. Those in need of treatment include those already with the disorder as well as those in which the disorder is to be prevented.

[0058] "Mammal" for purposes of treatment refers to any animal classified as a mammal, such as humans, non-human higher primates, domestic and farm animals, and zoo, sport or pet animals, such as dogs, horses, rabbits, cows, pigs, hamsters, mice, cats, etc. Preferably, the mammal is a human.

[0059] A "disorder" is any condition that would benefit from treatment with the protein, including chronic and acute disorders or diseases, such as pathological conditions that predispose a mammal to the disorder.

[0060] A "therapeutically effective amount" is at least the minimum concentration required to bring about a measurable improvement or prevention of a particular disorder. Therapeutically effective amounts of known proteins are well known in the art, but effective amounts of future discovered proteins below can be determined by standard techniques well within the skill of the art, e.g., ordinary physician.

[0061] II. How to Carry Out the Invention A. Protein preparation According to the present invention, the proteins are produced by recombinant DNA techniques, i.e., by culturing cells that have been transformed or transfected with a vector containing a nucleic acid encoding the protein, as is well known in the art.

[0062] Preparation of proteins by recombinant means can be accomplished by transfecting or transforming suitable host cells with an expression or cloning vector and culturing in conventional nutrient media modified as appropriate to induce promoters, select transformants, or amplify genes encoding the desired sequences. Culture conditions, such as medium, temperature, pH, etc., can be selected by one of skill in the art without undue experimentation. In general, principles, protocols, and practical techniques for maximizing cell culture productivity can be found in Mammalian Cell Biotechnology: A Practical Approach, M. Butler, Ed. (IRL Press, 1991) and Sambrook et al., Molecular Cloning: A Laboratory Manual, New York: Cold Spring Harbor Press. Transfection methods are known to those of skill in the art and include, for example, CaPO4 and CaCl2 transfection, electroporation, microinjection, and the like. Suitable techniques are also described in Sambrook et al. (supra). Additional transfection techniques can be found in Shaw et al., Gene 23 : 315 (1983);WO 89 / 05859;Graham et al., Virology 52 : 456-457 (1978) and U.S. Pat. No. 4,399,216.

[0063] The nucleic acid encoding the desired protein can be inserted into a replicable vector for cloning or expression. Suitable vectors are publicly available and can take the form of a plasmid, cosmid, viral particle or phage. The appropriate nucleic acid sequence can be inserted into the vector by a variety of procedures. Generally, DNA is inserted into an appropriate restriction endonuclease site(s) using techniques known in the art. Vector components generally include, but are not limited to, a signal sequence, an origin of replication, one or more marker genes, and one or more of an enhancer element, a promoter and a transcription termination sequence. The construction of a suitable vector containing one or more of these components uses standard ligation techniques known to those skilled in the art.

[0064] The form of the protein can be recovered from the medium or from the host cell lysate. If membrane-bound, it can be released from the membrane using a suitable detergent or by enzymatic cleavage. The cells used for expression can also be disrupted by various physical or chemical means, such as freeze-thaw cycles, sonication, mechanical disruption, or cell lysing agents.

[0065] Purification of proteins may be accomplished by any suitable technique known in the art, such as fractionation on ion exchange columns, ethanol precipitation, reverse phase HPLC, chromatography on silica or cation exchange resins (e.g., DEAE), chromatofocusing, SDS-PAGE, ammonium sulfate precipitation, gel filtration using Protein A Sepharose columns (e.g., Sephadex® G-75) to remove contaminants such as IgG, and metal chelate columns to bind epitope-tagged forms.

[0066] B. Antibody preparation In certain embodiments of the invention, the preferred protein is an antibody. Techniques for the production of antibodies, including polyclonal, monoclonal, humanized, bispecific and heteroconjugate antibodies, are provided below.

[0067] (i) Polyclonal antibodies Polyclonal antibodies are generally raised in animals by multiple subcutaneous (sc) or intraperitoneal (ip) injections of the relevant antigen and an adjuvant. It may be useful to conjugate the relevant antigen with a protein that is immunogenic in the species to be immunized, such as keyhole limpet hemocyanin, serum albumin, bovine thyroglobulin, or soybean trypsin inhibitor. Examples of adjuvants that can be used include Freund's complete adjuvant and MPL-TDM adjuvant (monophosphoryl lipid A, synthetic trehalose dicorynomycolate). One skilled in the art can select the immunization protocol without undue experimentation.

[0068] One month later the animals are boosted with 1 / 5 to 1 / 10 the original amount of peptide or conjugate in Freund's complete adjuvant by subcutaneous injection at multiple sites. 7 to 14 days later the animals are bled and the serum is assayed for antibody titer. Animals are boosted until the titer plateaus. Preferably, animals are boosted with the conjugate of the same antigen but conjugated to a different protein and / or via a different cross-linking reagent. Conjugates can be made in recombinant cell culture as protein fusions. Furthermore, agglutinating agents, such as alum, are suitably used to enhance the immune response.

[0069] (ii) Monoclonal antibody Monoclonal antibodies are obtained from a population of substantially homogeneous antibodies, i.e., the individual antibodies comprising the population are identical except for possible naturally occurring mutations that may be present in minor amounts. Thus, the modifier "monoclonal" indicates the character of the antibody as not being a mixture of discrete antibodies.

[0070] For example, monoclonal antibodies are described in Kohler et al., Nature, 256 They may be made using the hybridoma method first described by Johns Hopkins et al., J. Clin. Immunol. 495 (1975), or may be made by recombinant DNA methods (U.S. Patent No. 4,816,567).

[0071] In the hybridoma method, a mouse or other suitable host animal, such as a hamster, is immunized as described herein above to elicit lymphocytes that produce or can produce antibodies that specifically bind to the protein used for immunization. Alternatively, lymphocytes can be immunized in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusing agent, such as polyethylene glycol, to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp.59-103 (Academic Press, 1986)).

[0072] The immunizing agent typically contains the protein to be formulated. Generally, peripheral blood lymphocytes ("PBL") are used if cells of human origin are desired, or spleen cells or lymph node cells are used if non-human mammalian origin is desired. The lymphocytes are then fused with an immortalized cell line using a suitable fusing agent, such as polyethylene glycol, to form hybridoma cells (Goding, Monoclonal antibodies: Principles and Practice, Academic Press (1986), pp. 59-103). The immortalized cell line is usually a transformed mammalian cell, particularly a myeloma cell of rodent, bovine and human origin. Usually, rat or mouse myeloma cell lines are used. The hybridoma cells thus prepared are seeded and grown in a suitable medium, preferably containing one or more substances that inhibit the growth or survival of the unfused parental myeloma cells. For example, if the parent myeloma cells lack the enzyme hypoxanthine-guanine phosphoribosyltransferase (HGPRT or HPRT), the culture medium for the hybridomas typically contains hypoxanthine, aminopterin, and thymidine (HAT medium), which prevent the growth of the HGPRT-deficient cells.

[0073] Preferred myeloma cells are those that fuse efficiently, support stable high-level production of antibody by the selected antibody-producing cells, and are sensitive to a medium such as HAT medium. Among these, preferred myeloma cell lines are those derived from murine myeloma lines, such as MOPC-21 and MPC-11 mouse tumors available from the Salk Institute Cell Distribution Center, San Diego, California USA, and SP-2 cells available from the American Type Culture Collection, Rockville, Maryland USA. Human myeloma and mouse-human heteromyeloma cell lines have also been described for the production of human monoclonal antibodies (Kozbor, J. Immunol., 2002). 133 :3001 (1984); Brodeur et al., Monoclonal Antibody Production Techniques and Applications, pp. 51-63 (Marcel Dekker, Inc., New York, 1987)).

[0074] Culture medium in which hybridoma cells are growing is assayed for production of monoclonal antibodies directed against the antigen. Preferably, the binding specificity of the monoclonal antibodies produced by the hybridoma cells is determined by immunoprecipitation or by an in vitro binding assay, such as radioimmunoassay (RIA) or enzyme-linked immunosorbent assay (ELISA).

[0075] The binding affinity of a monoclonal antibody can be determined, for example, by the method described in Munson et al., Anal. Biochem., 107 This can be confirmed by Scatchard analysis of :220 (1980).

[0076] After hybridoma cells that produce antibodies with the desired specificity, affinity and / or activity are identified, the clones can be subcloned by limiting dilution procedures and grown by standard methods (Goding, supra). Suitable media for this purpose include, for example, D-MEM or RPMI-1640 medium. In addition, hybridoma cells can be grown in vivo as ascites tumors in animals.

[0077] The immunizing agent typically contains an epitope protein to which an antibody binds. Generally, peripheral blood lymphocytes ("PBLs") are used if cells of human origin are desired, or spleen cells or lymph node cells are used if non-human mammalian sources are desired. The lymphocytes are then fused with an immortalized cell line using a suitable fusing agent, such as polyethylene glycol, to form a hybridoma cell (Goding, Monoclonal antibodies: Principles and Practice, Academic Press (1986), pp. 59-103).

[0078] Immortalized cell lines are usually transformed mammalian cells, especially myeloma cells of rodent, bovine and human origin. Usually, rat or mouse myeloma cell lines are used. Hybridoma cells can be cultured in a suitable medium, preferably containing one or more substances that inhibit the growth or survival of unfused immortalized cells. For example, if the parent cells lack the enzyme hypoxanthine-guanine-phosphoribosyltransferase (HGPRT or HPRT), the medium for the hybridoma typically contains hypoxanthine, aminopterin and thymidine (HAT medium), which prevents the growth of HGPRT-deficient cells.

[0079] Preferred immortalized cell lines are those that fuse efficiently, support stable high-level expression of antibody by the selected antibody-producing cells, and are sensitive to a medium such as HAT medium. Further preferred immortalized cell lines are murine myeloma lines, which are available, for example, from the Salk Institute Cell Distribution Center, San Diego, California USA and the American Type Culture Collection, Rockville, Maryland. Human myeloma and mouse-human heteromyeloma cell lines have also been described for the production of human monoclonal antibodies (Kozbor, J. Immunol., 2002). 133 :3001 (1984); Brodeur et al., Monoclonal Antibody Production Techniques and Applications, Marcel Dekker, Inc., New York, (1987) pp. 51-63).

[0080] The medium in which the hybridoma cells are cultured can then be assayed for the presence of monoclonal antibodies directed against the protein to be formulated. Preferably, the binding specificity of the monoclonal antibodies produced by the hybridoma cells is determined by immunoprecipitation or by in vitro binding assays, such as radioimmunoassay (RIA) or enzyme-linked immunosorbent assay (ELISA). Such techniques and assays are known in the art. The binding affinity of the monoclonal antibodies can be determined, for example, by the method described in Munson et al., Anal. Biochem., 2003, 111:1311-1325, 1999. 107 This can be confirmed by Scatchard analysis of :220 (1980).

[0081] After the desired hybridoma cells are identified, the clones can be subcloned by limiting dilution procedures and grown by standard methods (Goding, supra). Suitable media for this purpose include, for example, Dulbecco's modified Eagle's medium and RPMI-1640 medium. Alternatively, the hybridoma cells can be grown in vivo as ascites tumors in animals.

[0082] The monoclonal antibodies secreted by the subclones are suitably separated from the culture medium, ascites fluid, or serum by conventional immunoglobulin purification procedures, such as, for example, protein A-Sepharose, hydroxylapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

[0083] DNA encoding a monoclonal antibody is readily isolated and sequenced using conventional techniques (e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of a murine antibody). Hybridoma cells serve as a preferred source of such DNA. Once isolated, the DNA is placed into an expression vector, which is then transfected into host cells that do not otherwise produce immunoglobulin proteins, such as E. coli cells, monkey COS cells, Chinese hamster ovary (CHO) cells, or myeloma cells, to synthesize the monoclonal antibody in the recombinant host cells. Review articles on recombinant expression of antibody-encoding DNA in bacteria include Skerra et al., Curr. Opinion in Immunol., 5 :256-262 (1993) and Pluckthun, Immunol. Revs. 130 :151-188 (1992).

[0084] In a further embodiment, the antibody is selected from the group consisting of those described in McCafferty et al., Nature, 348 Antibody phage libraries can be isolated from antibodies generated using the techniques described in Clackson et al., Nature, 200:552-554 (1990).352 :624-628 (1991) and Marks et al., J. Mol. Biol., 222 :581-597 (1991) describe the isolation of murine and human antibodies, respectively, using phage libraries. Subsequent publications have demonstrated the production of high affinity (nM range) human antibodies by chain shuffling (Marks et al., Bio / Technology, 10 :779-783 (1992)), as well as combinatorial infection and in vivo recombination as a strategy for constructing very large phage libraries (Waterhouse et al., Nuc. Acids. Res., 21 :2265-2266 (1993)). These techniques are thus viable alternatives to traditional monoclonal antibody hybridoma techniques for isolation of monoclonal antibodies.

[0085] The DNA may also be modified by substituting, for example, coding sequences for human heavy and light chain constant domains in place of the homologous murine sequences (U.S. Pat. No. 4,816,567; Morrison, et al., Proc. Natl. Acad. Sci. USA, 2002). 81 :6851 (1984)), or by covalently linking all or part of the coding sequence for a non-immunoglobulin polypeptide to the immunoglobulin coding sequence.

[0086] Typically, such non-immunoglobulin polypeptides are substituted for the constant domains of an antibody, or they are substituted for the variable domains of one antigen-binding site of an antibody to create a chimeric bivalent antibody containing one binding site with specificity for an antigen and another antigen-binding site with specificity for a different antigen.

[0087] Chimeric or hybrid antibodies can also be prepared in vitro using known methods in synthetic protein chemistry, including cross-linking agents.For example, immunotoxins can be constructed by using disulfide exchange reaction or by forming thioether bond.Examples of suitable reagents for this purpose include iminothiolate and methyl-4-mercaptobutyrimidate.

[0088] (iii) Humanized and Human Antibodies The antibodies subjected to the present formulation method may further include humanized or human antibodies. Humanized forms of non-human (e.g. murine) antibodies are chimeric immunoglobulins, immunoglobulin chains or fragments thereof (e.g. Fv, Fab, Fab', F(ab')2 or other antigen-binding subsequences of antibodies) that contain minimal sequence derived from non-human immunoglobulin. Humanized antibodies include human immunoglobulins (recipient antibody) in which residues from a complementarity determining region (CDR) of the recipient are replaced by residues from a CDR of a non-human species (donor antibody) such as mouse, rat or rabbit having the desired specificity, affinity and capacity. In some cases, Fv framework residues of the human immunoglobulin are replaced by corresponding non-human residues. Humanized antibodies may also include residues that are found neither in the recipient antibody nor in the imported CDR or framework sequences. Generally, a humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the CDR regions correspond to those of a non-human immunoglobulin and all or substantially all of the FR regions correspond to those of a human immunoglobulin consensus sequence. The humanized antibody optimally also will comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin (Jones et al., Nature, 2003). 321 :522-525 (1986); Reichmann et al., Nature, 332 :323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2:593-596 (1992).

[0089] Methods for humanizing non-human antibodies are well known in the art. Generally, a humanized antibody has one or more amino acid residues introduced into it from a source that is non-human. These non-human amino acid residues are often referred to as "import" residues, which typically come from an "import" variable domain. Humanization is essentially the process described by Winter and co-workers, Jones et al., Nature 2003, 144:1311-1321, 1999. 321 :522-525 (1986);Riechmann et al., Nature 332 :323-327 (1988);Verhoeyen et al., Science 239 :1534-1536 (1988), or by substituting rodent CDR(s) or CDR sequences for the corresponding sequences of a human antibody. Such "humanized" antibodies are thus chimeric antibodies (U.S. Pat. No. 4,816,567) in which substantially less than an intact human variable domain has been substituted by the corresponding sequences from a non-human species. In practice, humanized antibodies are typically human antibodies in which some CDR residues and possibly some FR residues are substituted by residues from analogous sites in rodent antibodies.

[0090] The choice of human variable domains to be used in the generation of a humanized antibody, both light and heavy chains, is very important to reduce antigenicity. According to the so-called "best-fit" method, the sequence of the variable domain of a rodent antibody is screened against the entire library of known human variable domain sequences. The human sequence that is closest to the rodent sequence is then accepted as the human framework (FR) for the humanized antibody (Sims et al., J. Immunol., 2003). 151 :2296 (1993);Chothia et al., J. Mol. Biol., 196:901 (1987)). Another method uses a particular framework derived from the consensus sequence of all human antibodies of a particular subgroup of light or heavy chains. The same framework can be used for several different humanized antibodies (Carter et al., Proc. Natl. Acad. Sci. USA, 89 :4285 (1992);Presta et al., J. Immnol., 151 :2623 (1993).

[0091] Moreover, it is important that antibodies be humanized while retaining high affinity for the antigen and other favorable biological properties. To achieve this goal, according to a preferred method, humanized antibodies are prepared by analysis of the parental sequences and various conceptual humanized products with three-dimensional models of the parental and humanized sequences. Three-dimensional immunoglobulin models are commonly available and are familiar to those skilled in the art. Computer programs are available which illustrate and display possible three-dimensional conformational structures of selected candidate immunoglobulin sequences. Inspection of these displays allows analysis of the possible role of the residues in the functioning of the candidate immunoglobulin sequence, i.e., analysis of residues that influence the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and combined from the recipient and import sequences so that the desired antibody characteristic, such as increased affinity for the target antigen(s), is achieved. In general, the CDR residues are directly and most substantially involved in influencing antigen binding.

[0092] Alternatively, it is now possible to produce transgenic animals (e.g., mice) that are capable, upon immunization, of producing a full repertoire of human antibodies in the absence of endogenous immunoglobulin production. For example, expression of the antibody heavy chain joining region (J) in chimeric and germ-line mutant mice has been demonstrated. HIt has been described that homozygous deletion of the .) gene results in complete suppression of endogenous antibody production. Transfer of the human germline immunoglobulin gene array in such germline mutant mice results in the production of human antibodies upon antigen challenge. See, e.g., Jakobovits et al., Proc. Natl. Acad. Sci. USA, 90 :2551 (1993);Jakobovits et al., Nature, 362 :255-258 (1993);Bruggermann et al., Year in Immuno., 7 :33 (1993). Human antibodies can also be obtained from phage display libraries (Hoogenboom et al., J. Mol. Biol., 227 :381 (1991);Marks et al., J. Mol. Biol., 222 :581-597 (1991).

[0093] Human antibodies can also be produced by various techniques known in the art, such as phage display libraries (Hoogenboom and Winter, J. Mol. Biol. 227 : 381 (1991);Marks et al., J. Mol. Biol. 222 : 581 (1991)). The techniques of Cole et al. and Boerner et al. are also available for the preparation of human monoclonal antibodies (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77 (1985) and Boerner et al., J. Immunol. 147(1): 86-95 (1991)). Similarly, human antibodies can be made by introducing human immunoglobulin loci into transgenic animals, such as mice, in which the endogenous immunoglobulin genes have been partially or completely inactivated. Upon antigen challenge, human antibody production is observed, which closely resembles that observed in humans in all respects, including gene rearrangement, assembly, and antibody repertoire. This approach is described, for example, in U.S. Pat. Nos. 5,545,807; 5,545,806, 5,569,825, 5,625,126, 5,633,425, 5,661,016, as well as in the following scientific publications: Marks et al., Bio / Technology 10 : 779-783 (1992);Lonberg et al., Nature 368 : 856-859 (1994);Morrison, Nature 368 : 812-13 (1994);Fishwild et al., Nature Biotechnology 14 : 845-51 (1996);Neuberger, Nature Biotechnology 14 : 826 (1996) and Lonberg and Huszar, Intern. Rev. Immunol. 13 : 65-93 (1995).

[0094] (iv) Antibody-Dependent Enzyme-Mediated Prodrug Therapy (ADEPT) The antibodies of the invention may also be used in ADEPT by conjugating them to prodrug-activating enzymes that convert prodrugs (e.g., peptidyl chemotherapeutic drugs, see WO 81 / 01145) into active anti-cancer drugs (see, e.g., WO 88 / 07378 and U.S. Pat. No. 4,975,278).

[0095] The enzyme component of an immunoconjugate useful for ADEPT includes any enzyme capable of acting on a prodrug in such a way as to convert it into its more active, cytotoxic form.

[0096] Enzymes that are useful in the methods of the invention include glycosidases, glucose oxidase, human lysozyme, human glucuronidase, alkaline phosphatase (useful for converting phosphate-containing prodrugs to free drug); arylsulfatases (useful for converting sulfate-containing prodrugs to free drug); cytosine deaminases (useful for converting the non-toxic 5-fluorocytosine to the anticancer drug 5-fluorouracil); proteases, such as Serratia protease, thermolysin, subtilisin, carboxypeptidases (e.g., carboxypeptidase G2 and carboxypeptidase A), and cathepsins (e.g., cathepsins B and L) (peptide-containing Examples of enzymes that can be used to convert prodrugs containing D-amino acid substituents include, but are not limited to, carbohydrate cleaving enzymes, such as β-galactosidase and neuraminidase, useful for converting glycosylated prodrugs to free drugs; β-lactamases, useful for converting β-lactam derivatized drugs to free drugs; and penicillin amidases, such as penicillin V amidase or penicillin G amidase, useful for converting drugs derivatized at their amine nitrogen with phenoxyacetyl or phenylacetyl groups, respectively, to free drugs. Alternatively, antibodies with enzymatic activity, also known in the art as "abzymes," can be used to convert the prodrugs of the invention to free active drugs (see, e.g., Massey, Nature 1999, 144:131-132). 328 : 457-458 (1987). Antibody-abzyme conjugates can be prepared as described herein for delivery of the abzyme to a tumor cell population.

[0097] The enzymes of the invention can be covalently linked to anti-IL-17 or anti-LIF antibodies by techniques well known in the art, such as the use of heterobifunctional cross-linkers as described above. Alternatively, fusion proteins comprising at least the antigen-binding region of an antibody of the invention linked to at least a functionally active portion of an enzyme of the invention can be constructed using recombinant DNA techniques well known in the art (e.g., Neuberger et al., Nature 1999, 143:131-132). 312 : 604-608 (1984).

[0098] (iv) Bispecific and multispecific antibodies A bispecific antibody (BsAb) is an antibody that has binding specificities for at least two different epitopes. Such antibodies can be derived from full length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies).

[0099] Methods for making bispecific antibodies are known in the art. Traditional production of full-length bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two chains have different specificities (Millstein et al., Nature, 305 :537-539 (1983)). Because of the random assortment of immunoglobulin heavy and light chains, these hybridomas (quadromas) produce a potential mixture of 10 different antibody molecules, of which only one has the correct bispecific structure. Purification of the correct molecule, which is usually done by affinity chromatography steps, is quite laborious, and the product yields are low. Similar techniques are described in WO 93 / 08829, as well as in Traunecker et al., EMBO J., 10 :3655-3659 (1991).

[0100] Antibody variable domains with the desired binding specificities (antibody-antigen combining sites) can be fused to immunoglobulin constant domain sequences. The fusions preferably have an immunoglobulin heavy chain constant domain, including at least part of the hinge, CH2, and CH3 regions. It is preferred to have the first heavy chain constant region (CH1) containing the site necessary for light chain binding present in at least one of the fusions. DNAs encoding the immunoglobulin heavy chain fusions and, optionally, the immunoglobulin light chain, are inserted into separate expression vectors and co-transfected into a suitable host organism. For further details on the generation of bispecific antibodies, see, for example, Suresh et al., Methods in Enzymology 121 : 210 (1986).

[0101] According to a different approach, antibody variable domains (antibody-antigen binding sites) with the desired binding specificity can be fused to immunoglobulin constant domain sequences. The fusions preferably have an immunoglobulin heavy chain constant domain, including at least a portion of the hinge, CH2 and CH3 regions. It is preferred to have the first heavy chain constant region (CH1) containing the site necessary for light chain binding present in at least one of the fusions. DNA encoding the immunoglobulin heavy chain fusions and, optionally, the immunoglobulin light chain are inserted into separate expression vectors and co-transfected into a suitable host organism. This provides great flexibility in adjusting the mutual ratio of the three polypeptide fragments in embodiments where unequal ratios of the three polypeptide chains used in the construction provide optimal yields. However, it is possible to insert the coding sequences for two or all three polypeptide chains into one expression vector when expression of at least two polypeptide chains in equal ratios results in high yields or when the ratios are not of particular importance.

[0102] According to another approach described in WO96 / 27011, the interface between a pair of antibody molecules can be engineered to maximize the percentage of heterodimers recovered from recombinant cell culture. A preferred interface comprises at least a portion of the CH3 region of the antibody constant domain. In this method, one or more small amino acid side chains from the interface of the first antibody molecule are replaced with larger side chains (e.g., tyrosine or tryptophan). By replacing the large amino acid side chains with smaller ones (e.g., alanine or threonine), a compensatory "cavity" of identical or similar size to the large side chain(s) is created on the interface of the second antibody molecule. This provides a mechanism for increasing the yield of heterodimers over other unwanted end-products such as homodimers.

[0103] In a preferred embodiment of this approach, the bispecific antibody is composed of a hybrid immunoglobulin heavy chain with a first binding specificity in one arm, and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. It has been found that this asymmetric structure facilitates separation of the desired bispecific compound from undesired immunoglobulin chain combinations, since the presence of an immunoglobulin light chain in only one half of the bispecific molecule provides an easy method of separation. This approach is disclosed in WO 94 / 04690 (published March 3, 1994). For further details on the generation of bispecific antibodies, see, for example, Suresh et al., Methods in Enzymology, 121 :210 (1986).

[0104] Bispecific antibodies include cross-linked or "heteroconjugate" antibodies. For example, one of the antibodies in the heteroconjugate can be conjugated with avidin and the other with biotin. Such antibodies have been proposed, for example, to target immune system cells to unwanted cells (U.S. Pat. No. 4,676,980) and for the treatment of HIV infection (WO91 / 00360, WO92 / 200373). Heteroconjugate antibodies can be made using any convenient cross-linking method. Suitable cross-linking agents are well known in the art and are disclosed in U.S. Pat. No. 4,676,980, along with a number of cross-linking techniques.

[0105] Techniques for generating bispecific antibodies from antibody fragments have also been described in the literature. For the production of bivalent antibody fragments that are not necessarily bispecific, the following techniques can also be used. For example, Fab' fragments recovered from E. coli can be chemically coupled in vitro to form bivalent antibodies (Shalaby et al., J. Exp. Med., 2003). 175 :217-225 (1992).

[0106] Bispecific antibodies can be prepared as full length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies). Techniques for generating bispecific antibodies from antibody fragments are described in the literature. For example, bispecific antibodies can be prepared using chemical linkage. Brennan et al., Science 229 : 81 (1985) describes a procedure in which intact antibodies are proteolytically cleaved to generate F(ab')2 fragments. These fragments are reduced in the presence of the dithiol complexing agent sodium arsenite to stabilize vicinal dithiols and prevent intermolecular disulfide formation. The Fab' fragments generated are then converted to thionitrobenzoate (TNB) derivatives. One of the Fab'-TNB derivatives is then reconverted to a Fab'-TNB derivative to form the bispecific antibody. The bispecific antibodies produced can be used as agents for the selective immobilization of enzymes.

[0107] Fab' fragments can be directly recovered from E. coli and chemically coupled to form bispecific antibodies. Shalaby et al., J. Exp. Med. 175 : 217-225 (1992) describes the production of a fully humanized bispecific antibody F(ab')2 molecule. Each Fab' fragment was secreted separately from E. coli and subjected to in vitro directed chemical coupling to form the bispecific antibody. The bispecific antibody thus formed was capable of binding to cells overexpressing the ErbB2 receptor and to normal human T cells, and of inducing the lytic activity of human cytotoxic lymphocytes against human breast tumor targets.

[0108] Various techniques for making and isolating bivalent antibody fragments directly from recombinant cell culture have also been described. For example, bivalent heterodimers have been produced using leucine zippers (Kostelny et al., J. Immunol., 2003). 148 (5):1547-1553 (1992)). The leucine zipper peptides from the Fos and Jun proteins were linked to the Fab' portions of two different antibodies by gene fusion. The antibody homodimers were reduced at the hinge region to form monomers and then reoxidized to form the antibody heterodimers. Hollinger et al., Proc. Natl. Acad. Sci. USA, 90 The "diabody" technology described by G. H. et al., J. Immunol. 1999: 6444-6448 (1993), provides an alternative mechanism for making bispecific / bivalent antibody fragments. The fragments consist of a light-chain variable domain (V ) bound to the light-chain variable domain (V ) by a linker that is too short to allow pairing between the two domains on the same chain. L ) and a heavy chain variable domain (V H ) is included. Therefore, the V H and V L The domain is the complementary V L and V HAnother strategy for making bispecific / bivalent antibody fragments by the use of single-chain Fv (sFv) dimers has also been reported (Gruber et al., J. Immunol., 2003, 111:1311-1323, 2003). 152 :5368 (1994).

[0109] Antibodies with more than two valencies are contemplated. For example, trispecific antibodies can be prepared (Tutt et al., J. Immunol. 147 : 60 (1991).

[0110] Exemplary bispecific antibodies may bind to two different epitopes on a given molecule. Alternatively, to focus cellular defense mechanisms on cells expressing a particular protein, an anti-protein arm may be combined with an arm that binds to a trigger molecule on white blood cells, such as a T cell receptor molecule (e.g., CD2, CD3, CD28, or B7), or an Fc receptor (FcγR) for IgG, such as FcγRI (CD64), FcγRII (CD32), and FcγRIII (CD16). Bispecific antibodies may also be used to localize cytotoxic agents to cells expressing a particular protein. Such antibodies possess a protein-binding arm and an arm that binds to a cytotoxic agent or a radionuclide chelator, such as EOTUBE, DPTA, DOTA, or TETA. Another such bispecific antibody binds to the protein of interest and also binds to tissue factor (TF).

[0111] (v) Heterozygous antibody Heteroconjugate antibodies are also within the scope of the present invention. Heteroconjugate antibodies are composed of two covalently linked antibodies. Such antibodies have been proposed, for example, to target immune system cells to unwanted cells (US Pat. No. 4,676,980) and for the treatment of HIV infection (WO91 / 00360, WO92 / 200373 and EP 03089). It is contemplated that antibodies can be prepared in vitro using known methods in synthetic protein chemistry, such as those involving cross-linking agents. For example, immunotoxins can be constructed by using disulfide exchange reactions or by forming thioether bonds. Examples of suitable reagents for this purpose include iminothiolate and methyl-4-mercaptobutyrimidate, and those disclosed, for example, in US Pat. No. 4,676,980.

[0112] C. Purification of proteins including antibodies When the target polypeptide is expressed in a recombinant cell other than a cell of human origin, the target polypeptide is completely free of proteins or polypeptides of human origin. However, in order to obtain a preparation that is substantially homogeneous with respect to the target polypeptide, it is necessary to purify the target polypeptide from recombinant cell proteins or polypeptides. As a first step, the culture medium or lysate is typically centrifuged to remove particulate cell debris. The membrane and soluble protein fractions are then separated. The target polypeptide can then be purified from the soluble protein fraction as well as from the membrane fraction of the culture lysate, depending on whether the target polypeptide is membrane-bound or not. The following procedures are examples of suitable purification procedures: fractionation on immunoaffinity or ion exchange columns; ethanol precipitation; reversed-phase HPLC; chromatography on silica or on cation exchange resins, such as DEAE; chromatofocusing; SDS-PAGE, ammonium sulfate precipitation; gel filtration, such as with Sephadex® G-75; and protein A Sepharose columns to remove contaminants such as IgG.

[0113] Most companies currently producing monoclonal antibodies (MAbs) use a three-column platform approach that includes Protein A affinity chromatography for product capture, followed by anion exchange chromatography in a flow-through mode to extract negatively charged contaminants such as host cell proteins (HCPs), endotoxins, host DNA and leached Protein A, and then cation exchange or hydrophobic interaction chromatography (HIC) in a retention mode to remove positively charged contaminant species, including residual HCPs and product aggregates.

[0114] Viruses that may be present in a protein solution are larger than the protein itself, and it is therefore presumed that viruses can be removed from the protein according to size by filtration.

[0115] Virus filtration can remove larger viruses, such as retroviruses (diameter 80-100 nm), using high-throughput membranes, typically with a nominal pore size of about 60 nm. High-throughput membranes with a nominal pore size of 20 nm are also commercially available, allowing filtration to remove smaller viruses, such as parvoviruses (diameter 18-26 nm), while allowing the passage of proteins as large as 160 kD (about 8 nm), such as monoclonal antibodies. The present invention is primarily intended to solve the problems typically associated with filtration of such small viruses, using virus removal filters with smaller pore sizes.

[0116] Typically, a viral filtration step may be performed at any one of several points in a given downstream process. For example, in a typical monoclonal antibody purification process, viral filtration may occur after a low pH viral inactivation step, or after an intermediate column chromatography step, or after the final column chromatography step.

[0117] According to the present invention, a virus filtration unit operation can be performed at any stage of the downstream process. Virus filtration during downstream processing of monoclonal antibodies is typically performed after an affinity chromatography step (capture step) and an ion exchange purification step (final purification step).

[0118] The experimental set-up used in the experiments disclosed herein is illustrated in Figure 1. However, it is emphasized that the invention is not so limited. Other configurations known in the art are suitable and may be used in the methods of the invention.

[0119] In parallel flow virus filtration, a protein solution is typically forced into the environment at a constant flow rate onto the retentive side. A pressure differential developed across the virus removal filter causes the protein solution to permeate through the filter while the viruses are retained on the retentive side.

[0120] In the case of so-called "vertical flow" or "dead end" virus filtration, the same virus filter as that used in parallel virus filtration can be used, but the peripheral equipment and operating procedures are much simpler and cheaper than in the case of parallel flow virus filtration. Thus, in principle, "vertical flow" filtration involves placing a polymer-containing solution in a pressure vessel prior to filtration, and then pressurizing the solution through a virus removal filter with the aid of a pressure source, suitably nitrogen (gas) or air. Alternatively, a pump can be used on the retention side to filter the liquid through the virus removal filter at a predetermined flow rate.

[0121] The fineness of a filter is commonly expressed either as a pore size or as the approximate molecular weight (relative molecular weight) at which molecules are stopped by the filter (the so-called cut-off).

[0122] Virus filters are known in the art and are supplied by Millipore (Massachusetts, USA) and Asahi Chemical Industry Co., Ltd. (Japan), among others. Suitable parvovirus-retaining filters include Viresolve® Pro (Millipore Corp., Billerica, MA). Viresolve® Pro membranes have an asymmetric double membrane structure and are made of polyethersulfone (PES). The membrane structure is designed to retain viruses with a size greater than 20 nm, while allowing proteins with a molecular weight less than 180 kDa to pass through the membrane. Other filters suitable for removing small viruses, such as parvoviruses, from protein solutions include Novasip™ DV20 and DV50 virus removal filter capsules (Pall Corp., East Hills, NY), Virosart® CPV, Planova 20N (Asahi Kasei) and BioEX (Asahi Kasei). Novasip DV20 grade capsule filters utilize an Ultipor VF grade DV20 grade pleated membrane cartridge to remove parvoviruses as well as other viruses as small as 20 nm from up to 5-10 liters of protein solution. Novasip DV50 grade capsule filters incorporate an Ultipor VF DV50 grade Ultipleat® membrane cartridge to remove viruses larger than 40-50 nm. Novasip Ultipor VF capsule filters are supplied non-sterile and can also be gamma irradiated. Virosart® CPV utilizes a dual layer polyethersulfone asymmetric membrane and retains greater than 4 log of parvoviruses and 6 log of retroviruses.

[0123] Pre-filtration of the feed solution can have a dramatic effect on filter performance. Pre-filtration is typically aimed at removing impurities and contaminants, such as protein aggregates, DNA and other trace materials, that can lead to fouling of the virus filter.

[0124] According to the present invention, a significant enhancement in the efficacy of the virus filter can be achieved by a pre-filtration step that includes the use of both cation exchange and endotoxin removal media. In this context, the term "medium" or "media" is used to encompass any means for carrying out the cation exchange and endotoxin removal steps, respectively. Thus, the term "cation exchange media" specifically includes, but is not limited to, cation exchange resins, matrices, absorbers, and the like. The term "endotoxin removal media" includes, but is not limited to, any positively charged membrane surface, such as chromatographic endotoxin removal media, endotoxin affinity removal media, and the like.

[0125] Cation exchange media suitable for use in the prefiltration step of the present invention include, but are not limited to, Mustang® S, Sartobind® S, Viresolve® Shield, SPFF, SPXL, Capto® S, Poros® 50HS, Fractogel® S, Hypercel® D, and the like, which are commercially available.

[0126] Endotoxin removal media suitable for use in the prefiltration stage of the present invention include, but are not limited to, Mustang® E, Mustang® Q, Sartobind® Q, Chromasorb®, Possidyne®, Capto® Q, QSFF, Poros® Q, Fractogel® Q, and the like, which are commercially available.

[0127] The pre-filtration step can be carried out, for example, by taking an in-process chromatography pool and processing the pool on a filtration train that includes endotoxin removal and cation exchange media and a parvovirus filter. The endotoxin removal and cation exchange media act as a pre-filtration step, and the capacity of the parvovirus filter is independent of the order of the two steps in the filtration train. The filtration train can be operated in series as a single step, or it can be operated as different unit operations. For example, in one embodiment, the chromatography pool is first processed on the endotoxin removal media, the collected pool is then processed on the cation exchange media, and the subsequent pool is filtered on the parvovirus filter. As noted above, the order of application of the cation exchange media and the endotoxin removal media in the process sequence does not affect the parvovirus filtration capacity. The process can be operated over a wide pH range, for example, pH 4-10, with the optimal filter capacity depending on the target impurity profile and product attributes. Similarly, the protein concentration can vary over a wide range, for example, 1-40 g / L, but does not limit the mass throughput of the parvovirus filter.

[0128] The present invention will be further understood by reference to the following examples, which, however, are not intended to limit the scope of the invention. All citations throughout this disclosure are incorporated herein by reference in their entirety. EXAMPLES

[0129] material and method 1. Protein solution Since virus filtration during downstream processing of monoclonal antibodies is performed after affinity chromatography (capture step) and ion exchange steps (final purification step), all filtration experiments were performed using ion exchange (cation or anion exchange) chromatography pools in commercially suitable manufacturing processes. The mAb concentrations and pool conductivities of the cation exchange and anion exchange pools were 10 mg / ml and 10 mS / cm and 8 mg / ml and 4 ms / cm, respectively. Filtration experiments were performed with fresh feedstock (within 24 h of production) or with feedstock frozen at -70 °C after production and thawed at 4-8 °C before use. No significant differences were observed in the results obtained with fresh or frozen-thawed feedstock. Protein concentration was determined using a UV-vis spectrophotometer (NanoDrop ND-1000, NanoDrop Technologies, Wilmington, DE) and absorbance was measured at 280 nm.

[0130] 2. Membrane Filtration experiments were performed using Viresolve® Pro (Millipore Corp., Billerica, MA) parvovirus-retentive filters. Viresolve® Pro membranes have an asymmetric bilayer structure and are made of polyethersulfone (PES). The membrane structure is designed to retain viruses with a size greater than 20 nm, while proteins with a molecular weight less than 180 kDa are permeable through the membrane. Pre-filters for Viresolve® Pro evaluated in this study included Viresolve® Optiscale 40 depth filters (Millipore Corp., Billerica, MA), Fluorodyne EX Mini 0.2 μm sterile filters (Pall Corp., East Hills, NY) and membrane absorbers from the Mustang® family (Pall Corp., East Hills, NY). The membrane absorbers were obtained from the vendor in fully enclosed Acrodisc® units. Table 1 summarizes the important characteristics (functional group, bed volume, pore size, etc.) of all the prefilters used in this study. [Table 1]

[0131] 3. Experimental Setup Filtration experiments were performed using a custom-built apparatus as shown in Figure 1. The load material, i.e., the in-process mAb pool, was placed in a loading reservoir and filtered through a filtration train consisting of different combinations of pre-filters and a commercial parvovirus filter. In all filtration experiments, a constant filtration flow rate (P max ) method was used. Pressure transducers were placed upstream of each filter and interfaced with a Millidaq or Netdaq system to record differential pressure data as a function of time or mass throughput. Filtrate from the parvovirus filters was collected in a reservoir, which was held on a load cell and mass throughput was recorded as a function of time.

[0132] Results and Discussion Downstream purification of mAbs expressed in mammalian cell culture typically utilizes centrifugation and depth filtration as the first step to remove cells and cell debris, followed by affinity chromatography for mAb capture and removal of host cell proteins (HCPs), followed by cation exchange chromatography, viral filtration and anion exchange chromatography to further remove impurities such as aggregates, viruses, leached Protein A and HCPs. Most experiments in this study were performed with the cation exchange pool, with cation exchange chromatography being the second chromatographic step.

[0133] Figure 2 shows the therapeutic mAb feed flow at 200 L / m using different pre-filters. 2 Experimental data on differential pressure through Viresolve Pro at a constant flow rate per hour are shown. The X-axis represents the mass of mAb loaded per square meter of virus filter. The Y-axis represents the differential pressure through the virus filter as a function of mass throughput. The data clearly show that depth filters provide an order of magnitude increase in virus filtration capacity compared to sterile filters. A similar observation was made when Bolton et al. (Bolton et al. Appl. Biochem. 43:55-63, 2006) evaluated the effect of Viresolve Prefilter™ (depth filter media) as a prefilter for NFP parvovirus-retentive filters (Millipore Corp.) using a polyclonal IgG solution. The authors attributed the increased capacity to preferential adsorption of foulants (denatured proteins) due to hydrophobic interactions.

[0134] Depth filters have traditionally been used successfully for the clarification of cell culture fluids, but have a number of limitations that must be considered, particularly when used downstream of the capture step, e.g., as a pre-filter to a parvovirus-retentive filter. (a) Depth filters are not substrate stable, preventing sanitation of the process line after installation, creating open processing and the potential for microbial contaminant growth. (b) The composition of depth filters contains diatomaceous earth as a key component, which is typically food grade and presents quality issues. (c) Diatomaceous earth is generally sourced naturally (lacking a well-defined chemical process) and therefore may exhibit lot-to-lot variations. (d) Depth filters also tend to leach out metals, β-glycans and other impurities, and this scavenging needs to be demonstrated and confirmed with downstream operations.

[0135] These limitations place a significant burden on process development because unit operations downstream of the depth filter must be designed to provide adequate leachate clearance. However, even if the leachate clearance requirements are met, there is reason to fear that a particular lot of depth filter may have a significantly higher leachate than the process is capable of clearing, because the major components are naturally sourced, i.e., they lack a well-defined chemical synthesis process.

[0136] Therefore, there has been great interest in developing prefilters that do not exhibit these limitations. As mentioned above, Brown et al. th (Antibody Development and Production, San Diego, CA, 2008) recently showed that Mustang S, a strong negatively charged ion exchanger, can increase the capacity of a parvovirus-retentive filter several-fold when used as a prefilter. Experiments were therefore performed to evaluate the effect of different prefiltration media on Viresolve® Pro. Experimental data at pH 5.0 and 6.5 are shown in Figures 3(a) and (b). The data show that the cation exchange media showed a slight advantage over the endotoxin removal sorbent at pH 5.0, but that advantage disappeared at pH 6.5. The overall capacity of both media was higher than that with the sterilizing filter (Figure 2); however, it was significantly below the capacity required to successfully run the unit operation at manufacturing scale.

[0137] Based on the hypothesis that both cation exchange and endotoxin removal media can remove two different foulants and both can filter foulants, an experiment was designed using a novel prefiltration train containing both ion exchange and endotoxin removal media. The experimental results are shown in Figures 4(a) and (b) (pH 5.0 and pH 6.5, respectively). The data clearly show that the combination of the two media is significantly better than each of the media by itself. For example, at pH 5.0, the combination of cation exchange and endotoxin removal media provides an order of magnitude improvement in capacity at a pressure differential of 20 psa. A similar trend was observed at pH 6.5, but the overall capacity was lower than that obtained at pH 5.0. This is likely due to the fact that impurities are more robustly removed at lower pH.

[0138] The results of the experiment with MAb2 are shown in Figure 5. Consistent with the data in Figure 4, the novel prefiltration train containing both endotoxin removal and cation exchange media significantly increased capacity, suggesting that the endotoxin removal and cation exchange media act synergistically to remove two different types of foulants.

[0139] conclusion The majority of previous studies have focused on using depth filters or cation exchange membrane adsorbers as pre-filters to increase the capacity of parvovirus-retentive filters. Depth filters provide a robust mechanism for increasing virus filtration capacity, but limitations associated with them, such as filtrate, limit their application to certain stages in downstream purification trains. Cation exchange membrane adsorbers can increase parvovirus filter capacity for some monoclonal antibody (mAb) feed streams, but they are not universally applicable, as seen by the data in this study, suggesting that there may be a number of foulants that need to be addressed to further improve the performance of parvovirus removal filters.

[0140] The present invention, as demonstrated by the above experimental results, emphasizes two aspects: (1) endotoxin removal media itself can effectively increase the capacity of a parvovirus filter when used for prefiltration, and (2) the combination of endotoxin removal and cation exchange media in a prefiltration train can increase the parvovirus filtration capacity several fold, lowering the raw material cost and facilitating the successful operation of virus filtration at a manufacturing scale.

Claims

1. A method for improving the filtration capacity of a virus filter during protein purification, comprising subjecting a composition containing the protein to be purified to a cation exchange step and an endotoxin removal step, either simultaneously or in either order, and then passing the composition through the virus filter.

2. 2. The method of claim 1, wherein the virus filter has a pore size of about 15 to about 100 nm in diameter.

3. The method of claim 2, wherein the virus filter has a pore size of about 15 to about 30 nm in diameter.

4. The method of claim 3, wherein the pore size of the virus filter is about 20 nm.

5. 5. The method according to claim 3 or 4, wherein the virus to be removed is a parvovirus.

6. The method of claim 5, wherein the parvovirus has a diameter of about 18 to about 26 nm.

7. The method of claim 1, wherein the protein is an antibody or an antibody fragment.

8. The method of claim 7, wherein the protein is a recombinant antibody or antibody fragment.

9. The method of claim 8, wherein the recombinant antibody or antibody fragment is produced in a mammalian host cell.

10. The method of claim 9, wherein the mammalian host cell is a Chinese Hamster Ovary (CHO) cell.

11. 2. The method of claim 1, wherein the composition containing the protein to be purified is first subjected to a cation exchange step, then an endotoxin removal step, followed by virus filtration.

12. 2. The method of claim 1, wherein the composition containing the protein to be purified is first subjected to an endotoxin removal step, then a cation exchange step, followed by virus filtration.

13. 2. The method of claim 1, wherein the composition containing the protein to be purified is subjected simultaneously to an endotoxin removal step and a cation exchange step, followed by virus filtration.

14. The method of claim 11, wherein the endotoxin removal step is immediately followed by viral filtration.

15. The method of claim 12, wherein the cation exchange step is immediately followed by viral filtration.

16. 14. The method of claim 13, wherein the simultaneous endotoxin removal and cation exchange step is immediately followed by viral filtration.

17. 2. The method of claim 1, wherein the viral filtration is carried out at a pH of about 4 to about 10.

18. The method of claim 1, wherein the protein concentration in the composition is about 1 to 40 g / L.

19. 9. The method of claim 8, wherein the antibody is directed against one or more antigens selected from the group consisting of HER1 (EGFR), HER2, HER3, HER4, VEGF, CD20, CD22, CD11a, CD11b, CD11c, CD18, ICAM, VLA-4, VCAM, IL-17A and / or F, IgE, DR5, CD40, Apo2L / TRAIL, EGFL7, NRP1, mitogen-activated protein kinase (MAPK) and Factor D.

20. The antibody is an anti-estrogen receptor antibody, an anti-progesterone receptor antibody, an anti-p53 antibody, an anti-cathepsin D antibody, an anti-Bcl-2 antibody, an anti-E-cadherin antibody, an anti-CA125 antibody, an anti-CA15-3 antibody, an anti-CA19-9 antibody, an anti-c-erbB-2 antibody, an anti-P-glycoprotein antibody, an anti-CEA antibody, an anti-retinoblastoma protein antibody, an anti-ras oncoprotein antibody, an anti-Lewis X antibody, an anti-Ki-67 antibody, an anti-PCNA antibody, an anti-CD3 antibody, an anti-CD4 antibody, an anti-CD5 antibody, an anti-CD7 antibody, an anti-CD8 antibody, an anti-CD9 / p24 antibody, an anti-CD10 antibody, an anti-CD11c antibody, an anti-CD13 antibody, an anti-CD14 antibody, an anti-CD15 antibody, an anti-CD19 antibody, an anti-CD23 antibody, an anti-CD30 ...

9. The method of claim 8, wherein the antibody is selected from the group consisting of an antibody, an anti-CD31 antibody, an anti-CD33 antibody, an anti-CD34 antibody, an anti-CD35 antibody, an anti-CD38 antibody, an anti-CD41 antibody, an anti-LCA / CD45 antibody, an anti-CD45RO antibody, an anti-CD45RA antibody, an anti-CD39 antibody, an anti-CD100 antibody, an anti-CD95 / Fas antibody, an anti-CD99 antibody, an anti-CD106 antibody, an anti-ubiquitin antibody, an anti-CD71 antibody, an anti-c-myc antibody, an anti-cytokeratin antibody, an anti-vimentin antibody, an anti-HPV protein antibody, an anti-kappa light chain antibody, an anti-lambda light chain antibody, an anti-melanosome antibody, an anti-prostate specific antigen antibody, an anti-S-100 antibody, an anti-tau antigen antibody, an anti-fibrin antibody, an anti-keratin antibody and an anti-Tn-antigen antibody.

Images