MX2008002447A - Proteolysis resistant antibody preparations - Google Patents

Abstract

Antibody preparations with substantially homogeneous and unsialylated glycoforms, such as G0 and G2, are prepared by enzymatic treatment, expression under certain conditions, use of particular host cells, and contact with serum. These antibody preparations resist cleavage by proteases, such as papain, ficin, bromolein, pepsin, a matrix metalloproteinase, such as MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3) and macrophage elastase (MMP-12), and glycosylation modification enzymes. The antibody preparations with substantially homogeneous and unsialylated glycoforms and methods of testing for glycosylation in an antibody are usefulin connection with characterization of antibody properties and/or in diseases or conditions characterized by an increase in protease activity.

Description

PREPARATIONS OF RESISTANT ANTIBODIES TO PROTEOLISIS FIELD OF THE INVENTION The invention relates to the evaluation of the glucoform content of antibodies and in particular, to methods for preparing and using preparations of antibodies which are substantially homogeneous glucoforms, for example, non-sialidated glucoforms.

BACKGROUND OF THE INVENTION Antibodies are serum-soluble glycoproteins that play an important role in innate immunity. The carbohydrate structures of all antibodies naturally produced at conserved positions in the heavy chain constant regions vary with isotype (Figure 1). Each isotype has a different arrangement of N-linked oligosaccharide structures, which variably affect the protein assembly, secretion or functional activity (Wright, A., and Morrison, SL, Trends Biotech 15: 26-32 (1997 )). The structure of the bound N-linked oligosaccharides (Figure 2) varies considerably, depending on the degree of processing, and may include high mannose content, as well as biantennary complex oligosaccharides with or without fucose core residues and which divide GIcNAc (Wright , A., and Morrison, SL, supra).

Typically, there is a heterogeneous processing of the oligosaccharide core structures attached to a particular glycosylation site so that even monoclonal antibodies exist as multiple glycoforms. Likewise, it has been shown that the main differences in glycosylation of antibodies occur between antibody producing cell lines, and even minor differences are seen for a certain cell line that grows under different culture conditions. Among the antibody isotypes (eg, IgE, IgD, IgA, IgM, and IgG), IgGs are the most abundant with the IgG1 subclasses presenting the most important degree and disposition of effector functions. IgG1-type antibodies are the most commonly used antibodies in cancer immunotherapy where the activity of ADCC and CDC is often considered important. Structurally, the region of hinge IgG and CH2 domains play a major role in the effector functions of antibodies. The N-linked oligosaccharides present in the Fe region (formed by the dimerization of the hinge, CH2 and CH3 domains) affect the effector functions. These covalently bound oligosaccharides are complex biantennary type structures and are highly heterogeneous (see Figure 2); NANA, 5-N-acetylneuraminic acid, (NeuAc) or NGNA, 5-N-glycolyl neuraminic acid (NeuGc) is typically "sialic acid". Other sialic acids have been found or can be chemically synthesized. An N-linked glycosylation site conserved in AsN297 extends into each CH2 domain. In the mature antibody, the two complex biantennary oligosaccharides attached to Asn297 are hidden between the CH2 domains, forming extensive contacts with the basic polypeptide structure. It has been found that its presence is essential for the antibody to mediate effector functions such as ADCC (Lifely MR, et al., Glycobiology 5: 813-822 (1995); Jefferis R., et al., Immunol Rev. 163: 59 -76 (1998); Wright, A. and Morrison, SL, supra). The presence and biological importance of individual saccharides in specific positions has also begun to be explored. For example, the degree of galactosylation of antibodies is carried out by age, gender and disease (Raju, T.S., et al., Glycobiology 2000. 10 (5): 477-86). In general, oligosaccharide structures are somewhat species specific and vary widely. In addition, the biological importance of oligosaccharide structures with or without fucose core residues and dividing GIcNAc has also been studied. Human IgG and many of the recombinantly produced IgGs contain minor amounts of sialylated (or non-sialylated or asialylated) oligosaccharides, however, the vast majority of IgGs contain non-sialylated oligosaccharide structures. The proteolytic breakdown of antibodies occurs naturally under physiological conditions and can also be an industrial processing step in the production of biological therapeutics based on the structure of the antibody. Therapeutic antibody fragments processed or generated with papain, Fabs, are having more widespread use. Although papain is an enzyme derived from plants, there are a number of protease cleavage sites identified in the lgG1 hinge and are summarized in Figure 3. Recombinant IgGs can be converted into IgG fragments, such as Fab and F (ab '). ) 2 and Fe, using various proteolytic enzymes (figures 1 and 3). Digestion fragments represent a major biotherapeutic class useful for managing and treating human diseases. Abciximab ((c7E3 Fab, marketed as REOPRO®), is an example of a therapeutic Fab.The Fab fragment of 47,615 daltons is purified from cell culture supernatant, papain digestion and column chromatography.Other examples include: DigiFab ( Digi ), a preparation of Fab fragments from sheep polyclonal antibodies, for the potential treatment of digoxin poisoning; CroFAb, a preparation of monovalent Fab fragments obtained from sheep immunized with snake venoms, such as an antivenom against bites from the four most common North American rattlesnakes (vipers) approved in the United States in October 2000; and EchiTAb, an antivenom based on Fab fragments of monospecific sheep polyclonal antibodies, for the treatment of bites of the carpet viper (Echins Ocellatus), a common snake in West Africa. Other developing Fabs include: ranibizumab (rhuFab V2; AMD-Fab; Lucentis), a high-affinity Fab variant of Genentech's bevacizumab, as a potential treatment for age-related macular degeneration; and 5G1.1, an intravenous humanized monoclonal antibody that prevents the breakdown of the human complement component C5 into its proinflammatory components, as a potential treatment for several chronic inflammatory diseases, including rheumatoid arthritis (RA), membranous and lupus nephritis, dermatomyositis, and paroxysmal nocturnal hemoglobinuha (PNH). Other compositions containing Fab with a potential therapeutic use include chemically modified Fabs, such as CDP-870 a humanized anti-TNFalpha-Fab fragment bound to polyethylene glycol (PEG). CDP-870 is derived from a mouse anti-human TNFalpha antibody that was selected for its high affinity binding and neutralizing potential. The Fab fragments of this antibody were constructed through recombinant DNA technology, humanized and synthesized by fermentation by feeding lots in E coli. The yield of this fermentation process reached between 300 and 1200 mg of culture / l of bacterial culture. To improve the plasma half-life, a PEG portion was added to the Fab fragments. For this purpose, a specific site conjugation method was developed in which an individual cysteine residue was introduced into the hinge region of the Fab fragment for the covalent addition of the hydrophilic polymer (PEG) portion for the purpose of increasing its circulating average life. Using a low cost E coli technology to produce the Fab fragments, the manufacturer (Celltech) was allowed to reduce the costs of manufacturing CDP-870 from 10 to 20 times compared to antibodies that are conventionally produced in cell culture. mammal. E. coli does not express glycosylated proteins. To date, the relationship between the presence and composition of glucans on the IgG susceptibility to proteolytic cleavage from human or other species has not been studied. Therefore, there is a need to understand the relationship between the proteolytic pattern and the glucan structure of therapeutically relevant antibody structures for the purpose of efficient antibody production and as a tool to identify the presence and / or composition of glucans from antibodies BRIEF DESCRIPTION OF THE INVENTION The present invention comprises a method for improving the ability of an antibody preparation to resist cleavage by a protease and methods for using said antibody preparations to treat pathological conditions associated with the presence of high levels of proteases, such as cancer. In one embodiment of the method of the invention, the preparation of antibodies is substantially free of sialylated glucoforms in the Fe region of the antibody. In another aspect of the invention, the protease is selected from the group consisting of papain, pepsin, a matrix metalloproteinase that includes MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3), macrophage elastase (MMP-12) , trypsin, chymotrypsin, and other proteases, including glycosylation modification enzymes, eg, sialidase-a, galactosidase, etc. In one embodiment, the antibody preparation is modified to be substantially homogeneous with respect to the GO glycogen. In another embodiment, the present invention comprises a method for increasing or reducing the ability of an antibody preparation to resist cleavage by a protease and methods for using said antibody preparations in diseases or conditions associated with the presence of increased or reduced levels of protease. . The present invention also comprises a method for improving the stability of an antibody with a protease, such as papain, pepsin, a matrix metalloproteinase including MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3), macrophage elastase ( MMP-12), trypsin, chymotrypsin, and other proteases, including glycosylation modification enzymes by treating an in vitro antibody preparation with a sialidase and optionally, further treatment of the antibody with a β-galactosidase and / or α-galactosidase to remove the galactose residues. The present invention further comprises a method for detecting or diagnosing a disease state in a cell or subject, which comprises determining the glycosylation status of antibodies in the cell or subject. The method for determining or diagnosing the disease state can be based on an analysis of the glycophor of the antibodies naturally or therapeutically administered in a subject together with a determination of the presence of Fab, F (ab ') 2, Fv, facb fragments o Faith in a biological sample of said subject. The present invention further provides any invention described herein.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 illustrates antibody IgG domains showing the relationship between the designated domains and major break fragments; Figure 2 is a schematic illustration of the variations in the structure of biantennary oligosaccharides found in human IgG; Figure 3 shows the amino acid sequence in the human IgG hinge region and break sites for various enzymes; Figures 4A-4G are MALDI-TOF-MS records with peak information on the identification of overlapping species (+1 is individually charged molecular ion, +2 is double charged molecular ion, +2 is triply charged molecular ion, and LC is free light chain) and show the formation of IgG fragments over time during digestion with papain for a glycosylated and deglycosylated preparation: Figure 4A: undigested, Figure 4B: half an hour, Figure 4C: 1 hour, Figure 4D: 2 hours, Figure 4E: 4 hours, Figure 4F: 8 hours and Figure 4G: after 24 hours; Figures 5A-5B show a comparison of the percentage of peak area of molecular ions +1 intact IgG (Figure 5A) and Fe fragments (Figure 5B) of glycosylated and deglycosylated IgG samples during papain digestion; Figure 6 shows traces of the MALDI-TOF-MS analysis of intact homogenous IgG glycophorus preparations described in Example 5 for the GO, G2 and G2S2 glyco- forms; Figures 7A-7D show traces of the MALDI-TOF-MS analysis of the glucans released from PGNase from the various preparations of homogeneous glycoform and from the control sample; Figure 8 shows traces of the MALDI-TOF-MS analysis of digested with papain from glycogen preparations of homogeneous IgG GO, G2 and G2S2 together with a control sample subjected to digestion with papain at a 50: 1 ratio at 37 ° C after 15 minutes with various marked peak identities; Figure 9 is a graphical representation of the integrated peak area of intact IgG from MALDI-TOF-MS analysis of digested with papain from glycogen preparations of homogeneous IgG G0, G2 and G2S2 and with respect to a control subjected to digestion with papain at a 50: 1 ratio at 37 ° C at various times; and Figure 10 is a graphical representation of the integrated peak area of the Fe domain from MALDI-TOF-MS analysis formed during papain digestion of glycogen preparations of homogeneous IgG, G2 and G2S2 along with a control sample at various times. .

DETAILED DESCRIPTION OF THE INVENTION Abbreviations AA, anthranilic acid; «1, 3GT, f -1, 3-galactosyltransferase; β1, 4GT, β-1, 4-galactosyltransferase; a2,3ST, oc-2,3-sialyltransferase; ADCC, antibody-dependent cellular cytotoxicity; CDC, complement-dependent cytotoxicity; CMP-Sia, cytidine monophosphate N-acetylneuraminic acid; FBS, fetal bovine serum; IgG, immunoglobulin G; MALDI-TOF-MS, time-of-flight mass spectrometry by ionization matrix-assisted laser desorption; NANA, isomer of N-acetylneuraminic acid of sialic acid; NGNA, isomer of N-glycolylneuraminic acid of sialic acid; PNGase F, peptide N-glucosidase F; HPLC, reverse phase high performance liquid chromatography; SA, sinapic acid; Sia, sialic acid; SDHB, dihydroxybenzoic acid containing sodium chloride; UDP-Gal, uridine diphosphate galactose; UDP-GIcNAc, uridine diphosphate N-acetylglucosamine.

Definitions The terms "Fe," "Fe-containing protein" or "Fe-containing molecule" as used herein, refer to a monomeric, dimeric or heterodimeric protein having at least one CH2 and CH3 domain of immunoglobulin. The CH2 and CH3 domains can form at least a portion of the dimeric region of the protein / molecule (eg, antibody). The term "antibody" is intended to encompass antibodies, fragments of digestion, portions and specified variants thereof, including without limitation, antibody mimetics or comprising portions of antibodies that mimic the structure and / or function of an antibody or fragment or portion specified thereof, including without limitation, single chain antibodies, single domain antibodies, minibodies, and fragments thereof. Functional fragments include antigen-binding fragments that bind to the target antigen of interest. For example, antibody fragments capable of binding to a target antigen or portions thereof, including but not limited to Fab fragments (e.g., by papain digestion), Fab '(e.g., by digestion with pepsin and reduction). partial) and F (ab ') 2 (e.g., by digestion with pepsin), facb (e.g., by digestion with plasmin), pFc' (e.g., by digestion with pepsin or plasmin), Fd (e.g. digestion with pepsin, partial reduction and reaggregation), Fv or scFv (for example, by molecular biology techniques) are encompassed by the term antibody (see for example, Colligan, Immunology, supra).

The term "monoclonal antibody" as used herein, is a specific form of a Fe-containing fusion protein comprising at least one ligand binding domain which retains substantial homology to at least one of an antibody variable domain of heavy or light chain of at least one species of animal antibody.

Enzymatic enzyme digestion Because of their binding to high affinity targets, Fabs also provide an ideal target portion, for example, for conjugation of toxins or for embedding in more complex structures, such as liposomes. As an improvement for large circulating lipid vesicles that carry encapsulated drug, it is expected that the immunoliposomes or targets allow a more precise delivery of active to diseased or pathogenic tissue and at the same time saving normal cells thus reducing the side effects. The use of IgG and Fab as target portions for therapeutic liposomes is described in US4957735 and Maruyama et al. (1995) Biochim Biophys Acta 1234: 74-80. Papain is a sulfhydryl protease that has been used to digest IgG antibodies in either Fab or F (ab ') 2 fragments, depending on whether L-cysteine is present or absent during the reaction, respectively. Prolonged treatment, or excessive amounts of papain, usually results in an overdigestion of the Fe domain, although Fab domains often remain resistant to overdigestion with papain. This is because the Fe domain contains additional (secondary) papain break sites (Figure 1). The histidine residue is the C-terminal position of abciximab when digestion with papain is carried out in presence of cysteine.

Human IgG1: A-E-P-K-S-C-D-K-T-H-T-C-P-P-C-P-A-P-E-L-L-G-G human IgG2: C-P-P-L-K-E-C-P-P-C-P-A-P-P -_- V-A-G human IgG3: C-D-T-P-P-P-C-P-R-P-C-P-A-P-E-L-L-G human IgG4: S-K-Y-G-P-P-C-P-S-C-P-A-P Although papain is an industrially useful enzyme, it is of plant origin initially isolated from the green fruit and leaves of Carica papaya . { Caricaceae spp.) An industrially useful mammalian enzyme is pepsin.

Pepsin is self-activated and active at low pH since it is a normal component of gastric fluid secreted into the lumen of the stomach after eating.

Low levels of the precursor of the pepsinogen enzyme can be found in serum, but because the activation and activity are dependent on the acid, is not physiologically relevant to circulating antibodies. The pepsin breaks human IgG1 between Ieucine234-leucine235 in the hinge lower. This breaking site is downstream of the hinge core (-C-P-P-C-) containing two cysteine residues that link the two chains heavy through disulfide bonds creating a molecule F (ab ') 2 which is divalent for antigen binding.

The lower hinge region / CH2, P-A-P-E-L-L-G-G-P-S-V-F is within the domain where there are break sites for MMP-3 and MMP-12 (P-A-P * E-L-L-G for each) as well as pepsin and MMP-7 (P-A-P-E-L * L-G for each). In addition, a group of relevant physiological enzymes; Neutrophil elastase (HNE), stromelysin (MMP-3) and macrophage elastase (MMP-12) break IgG into different positions to generate subtly different F (ab ') 2, Fab and Fe fragments (Figure 3). Unexpectedly, it was discovered that the level of glycosylation of Fe alters the susceptibility to enzymatic degradation of said antibodies, resulting in the modulation of various aspects of the production processes and biological actions of said antibodies. More specifically, in the course of these experiments, it was discovered that glycosylated Abs of Abs is more resistant to digestion with papain than deglycosylated, agglucosylated or non-glycosylated Abs. Substantially deglycosylated, aglycosylated or non-glycosylated will mean that most of the actual and / or potential glycosylation sites are not occupied (with glucan), ie they are not glycosylated. The present invention further comprises a method for controlling the properties of an Fe-containing molecule by altering the glycosylation of the CH2 domains of Fe and the use of altered Fe-containing molecules. The presence or absence of glucan in the molecule containing Fe affects affinity for one or more of the Fc? RI, Fc? RIIA, and Fc? RIIIA receptors, ADCC activity, activation of macrophages or monocytes, and serum half-life (Lifely et al., Jeffreis, and Wright et al. Morrison, supra). Therefore, because proteolytic degradation is a measure of glycosylation and glycosylation is a requirement for the secondary functions of an IgG class antibody, the susceptibility to proteolysis becomes a marker for the aforementioned functions of said IgG antibody. IgG class. For example, sialic acid has a net negative charge at physiological pH and, therefore, it could be expected that the presence of sialic acid in the carbohydrate bound to Fe will alter the three-dimensional structure and therefore the conformation of the CH2 domain and affect this way the accessibility of Fe by proteolytic enzymes. Accordingly, the sialic acid content of the oligosaccharide linked to the CH2 domain is a determinant of proteolytic susceptibility and the proteolytic breaking ratio is a measure of the sialic acid content of the IgG or other Fe-containing protein.

Enrichment of protein-containing proteins containing Fe One method to prepare sublots of a particular protein containing Fe that differ in glucan content and structure is to take a protein preparation containing Fe with heterogenous Fe oligosaccharides, including glycosylated and aglycosylated molecules, and passing over a column containing an immobilized lectin having differential affinity, for example, for sialylated and asialylated oligosaccharides. The non-union through flow (T, through) or the unbound column fraction can be separated from the bound fraction (B, bound), the latter collected at the time of passing the elution pH regulator through the column. It may also be possible to separately collect a weakly bound fraction or the delayed column fraction (R, delayed), for example, by collecting Fe-containing protein that elutes during continuous washing of the column with the original sample pH regulator. Depending on the lectin used, the binding fraction is expected to have a higher content of saccharide, for example, sialic acid, and therefore, oligosaccharide content, than the non-binding fraction. Examples of lectins that can enrich the proteins containing sialylated or asialylated Fe are the lectin from Maackia amurensis (MAA), which specifically binds oligosaccharides with terminal sialic acid, and the wheat germ agglutinin lectin (WGA), which specifically binds oligosaccharides with terminal sialic acid or terminal N-acetyl glucosamine (GIcNAc). Another example is lectin Ricin I (RCA), which binds oligosaccharides with terminal galactose. In the last example, the non-binding through-flow fraction can be enriched for molecules containing sialylated Fe. Other lectins known in the art include those provided by Vector labs and EY labs.

Enzymatic Modification of Fe-Containing Proteins An alternative method for preparing sublots of a Fe-containing protein that differ in glucan content is to treat a portion of a Fe-containing protein preparation with a sucrase, such as a fucosidase or sialidase enzyme, eliminating in this way the specific sugar residues, for example, fucose or sialic acids. The resulting afucosylated or asylated material can be compared to the original, partially fucosylated or sialylated material for differences in biological activity. The addition of saccharides to the Fe region can also be obtained using in vitro glycosylation methods. The glucosyltransferases naturally work to synthesize oligosaccharides. They produce specific products with excellent stereochemical and regiochemical geometry. The transfer of glucosyl residues results in the elongation or synthesis of an oligo- or polysaccharide. A number of types of glycosyltransferase have been described, including sialyltransferases, fucosyltransferases, galactosyltransferases, mannosyltransferases, N-acetylgalactosaminyltransferases, N-acetylglucosaminyltransferases, and the like. The glycosyltransferases that are useful in the present invention include, for example, a-sialyltransferases, ot-glucosyltransferases, α-galactosyltransferases, o-fucosyl-transferases, a-mannosyltransferases, α-xylosyltransferases, or < -N-acetylhexosaminyltransferases, ß-sialyltransferases, ß-glucosyltransferases, ß-galactosyltransferases, ß-fucosyltransferases, ß-mannosyltransferases, ß-xylosyltransferases, and ß-N-acetylhexosaminyltransferases, such as those of Neisseria meningitidis, or other bacterial sources, and those of rat, mouse, rabbit, cow, pig, human and insect and viral sources. Preferably, the glycosyltransferase is a truncation variant of the glycosyltransferase enzyme in which the membrane binding domain has been deleted.

Exemplary galactosyltransferases include (1, 3) galactosyltransferase (EC No. 2.4.1.151, see, for example, Dabkowski et al., Transplant Proc. 25: 2921 (1993) and Yamamoto et al., Nature 345: 229-233 ( 1990)) and (1, 4) galactosyltransferase (EC No. 2.4.1.38). Other glycosyltransferases, such as a sialyltransferase, can be used. An α (2,3) siallyltransferase, often referred to as the sialyltransferase, can be used in the production of sialyl lactose or higher order structures. This enzyme transfers sialic acid (NeuAc) of CMP-sialic acid to a Gal residue with the formation of a linkage between the two saccharides. The junction (linkage) between the saccharides is between the 2- position of NeuAc and the 3- position of Gal. An exemplary f (2,3) sialyltransferase referred to as a (2,3) sialyltransferase (EC 2.4.99.6) transfers sialic acid to the non-reducing terminus Gal of a disaccharide or Galß1? 3Gl glucoside. See, Van den Eijnden et al., J. Biol.

Chem., 256: 3159 (1981), Weinstein et al, J. Biol. Chem., 257: 13845 (1982) and Wen et al., J. Biol. Chem., 267: 21011 (1992). Another exemplary α-2,3-sialyltransferase (EC 2.4.99.4) transfers sialic acid to the non-reducing terminal Gal of the disaccharide or glucoside. See, Rearick et al., J. Biol. Chem., 254: 4444 (1979) and Gillespie et al., J. Biol. Chem., 267: 21004 (1992). Other exemplary enzymes include Gal-ß-1, 4-GlcNAc OI-2,6 sialyltransferase (See, Kurosawa et al., Eur. J. Biochem, 219: 375-381 (1994)). Other glycosyltransferases particularly useful for preparing oligosaccharides of the invention are mannosyltransferases including (1, 2) mannosyltransferase, OI (1, 3) mannosyltransferase, β (1, 4) mannosyltransferase, Dol-P-Man synthase, OCh1, and Pmt1 . Even other glycosyltransferases include N-acetylgalactosaminyltransferases including (1, 3) N-acetylgalactosaminyltransferase, β (1,4) N-acetylgalactosaminyltransferases (Nagata et al., J. Biol. Chem. 267: 12082-12089 (1992) and Smith et al., J. Biol Chem. 269: 15162 (1994)) and N-acetylgalactosaminyltransferase polypeptide (Homa et al., J. Biol Chem. 268: 12609 (1993)). Suitable N-acetylglucosaminyltransferases include GnTI (2.4.1.101, Hull et al., BBRC 176: 608 (1991)), GnTII, and GnTIII (Hiara et al., J. Biolchem, 113: 692 (1993)), GnTV (Shoreiban et al., J. Biol. Chem. 268: 15381 (1993)). For those modalities in which the method will be practiced on a commercial scale, it may be advantageous to immobilize the glucosyl transferase in a support. This immobilization facilitates the removal of the enzyme from the batch of product and subsequent use of the enzyme. The immobilization of glucosyl transferases can be carried out, for example, by removing the membrane binding domain from the transferase and fixing a cellulose binding domain instead. One skilled in the art will understand that other methods of immobilization can also be used and are described in the available literature. Because the acceptor substrates can essentially any monosaccharide or oligosaccharide having a terminal saccharide residue for which the particular glucosyl transferase has specificity, the substrate can be substituted at the position of its non-reducing end. In this manner, the glycoside acceptor can be a monosaccharide, an oligosaccharide, a saccharide with fluorescent labeling, or a saccharide derivative, such as an aminoglycoside antibiotic, a ganglioside, or a glycoprotein that includes antibodies and other proteins containing Fe. In a group of preferred embodiments, the glucoside acceptor is an oligosaccharide, preferably, Galβ (1 -3) GlcNAc, Galβ (1-4) GlcNAc, Galβ (1 -3) GalNAc, Galβ (1-4) GalNAc, Man <; x (1, 3) Man, Man 0 (1, 6) Man, or GalNAcß (1 -4) -mannose. In a particular preferred embodiment, the oligosaccharide acceptor is linked to the CH2 domain of an Fe-containing protein. The use of sugar-activated substrate, i.e., sugar-nucleoside phosphate, may be limited either by the use of a reaction regeneration concurrently with the glucotransferase reaction (also known as a recycling system). For example, as presented in, the US patent. 6,030,815, a CMP-sialic acid recycling system uses CMP-sialic acid synthetase to replenish CMP-sialic acid (CMP-NeuAc) as it reacts with a sialyltransferase acceptor in the presence of an α (2,3) sialyltransferase to form the sialyl saccharide. The CMP-sialic acid regenerative system useful in the invention comprises cytidine monophosphate (CMP), a nucleoside triphosphate (e.g., adenosine triphosphate (ATP), a phosphate donor (e.g., phosphoenolpyruvate or acetyl phosphate), a kinase ( example, pyruvate kinase or acetate kinase) capable of transferring phosphate from the phosphate donor to nucleoside diphosphates and a nucleoside monophosphate kinase (eg, myokinase) capable of transferring the terminal phosphate from a nucleoside triphosphate to CMP. 3) sialyltransferase and CMP sialic acid synthetase can also be seen as part of the regenerative system of sialic CMP acid while stirring the activated sialic acid which serves to maintain the upward rate of synthesis The synthesis and use of the sialic acid compounds in a sialization procedure using a phagemid comprising a gene for a modified sialic acid synthase CMP enzyme that is described in the international application WO 92/16640, published on October 1, 1992. An alternative method of preparing oligosaccharides is through the use of a glycosyltransferase and activated glucosyl derivatives as sugar donors as shown in the United States patent 5,952,203. Activated glycosyl derivatives act as variants for naturally occurring substrates which are expensive sugar nucleotides, usually nucleosidic diphosphates or nucleotide monophosphates in which the nucleotide phosphate is linked alpha to position 1 of the sugar. The activated glucoside derivatives which are useful include an activated leaving group, such as, for example, fluoro, chloro, bromo, tosylate ester, mesylate ester, triflate ester and the like. Preferred embodiments of the activated glucoside derivatives include glucosyl fluoride and glucosyl mesylates with glucosyl fluorides which are particularly preferred. Among the glucosyl fluorides, α-galactosyl fluoride, α-glucosyl fluoride, α-fucosyl fluoride, α-xylosyl fluoride, α-sialyl fluoride, alpha-N-acetylglucosaminyl fluoride, αN-acetylgalactosaminyl fluoride, β-galactosyl fluoride, β-mannosyl fluoride, β-glucosyl fluoride, β-fucosyl fluoride, β-xylosyl fluoride, beta-sialyl fluoride, β-N-acetylglucosaminyl fluoride, and β-N- fluoride Acetylgalactosaminyl are most preferred. The glucosyl fluorides can be prepared from free sugar by first acetylating the sugar and then treating it with HF / pyridine. The acetylated glucosyl fluorides may be deprotected by reaction with a mild (catalytic) base in methanol (e.g., NaOMe / MeOH). In addition, many glucosyl fluorides are commercially available. Other activated glucosyl derivatives may be prepared using conventional methods known to those skilled in the art. For example, glycosyl mesylates may be prepared by treating a fully benzylated hemioketal form of the sugar with mesyl chloride, followed by catalytic hydrogenation to remove the benzyl groups. Another component of the reaction is a catalytic amount of a nucleoside phosphate or its analogues. The nucleoside monophosphates which are suitable for use in this invention include, for example, adenosine monophosphate (AMP), cytidine monophosphate (CMP), uridine monophosphate (UMP), guanosine monophosphate (GMP), inosine monophosphate (IMP) and thymidine monophosphate (TMP). Suitable nucleoside triphosphates for use in accordance with this invention include adenosine triphosphate (ATP), cytidine triphosphate (CTP), uridine triphosphate (UTP), guanosine triphosphate (GTP), inosine triphosphate (ITP) and triphosphate of thymidine (TTP). A preferred nucleoside triphosphate is UTP. Preferably, the nucleoside phosphate is a nucleoside diphosphate, for example, adenosine diphosphate (ADP), cytidine diphosphate (CDP), uridine diphosphate (UDP), guanosine diphosphate (GDP), inosine diphosphate (IDP) and Thymidine diphosphate (TDP). A preferred nucleoside diphosphate is UDP. As noted above, this invention can also be practiced with a nucleoside phosphate analog. Suitable analogs include, for example, glucoside sulfates and sulphonates. Even other analogs include simple phosphates, for example pyrophosphate. A method for modifying the recombinant proteins produced, in for example murine cells in which the hydroxylated form of sialic acid predominates (NGNA), is to treat the protein with sialidase, to remove the sialic acid of the NGNA type, followed by an enzymatic galactosylation using the active agent UDP-Gal and beta 1, 4 galtransferase to produce highly homogeneous G2 glyco- forms. Then, the preparation can optionally be treated with the active agent CMP-NANA and alpha-2,3 sialitransferase to give highly homogenous G2S2 glyco- forms. For purposes of this invention, substantially homogeneity for a glycophor should be about 85% or greater than the glycophor and, preferably, about 95% or greater of the glycophor.

Structural characterization of sialic acid variants For the structural characterization of sialic acid variants containing oligosaccharides, glycoprotein preparations include antibody preparations that were treated with peptide-N-glucosidase F to release bound N-oligosaccharides. The peptide-N-glucosidase F (PNGase F) enzyme breaks the oligosaccharides linked with asparigins. The released oligosaccharides are fluorescently labeled with anthranilic acid (2-aminobenzoic acid), purified and analyzed by HPLC as described in (see Anumula, K. R. and Dhume ST, Glicobiology, 1998 Jul; 8 (7): 685-94). Alternatively, the released oligosaccharides may be subject to MALDI-TOF-MS, as described herein or to Esl-MS. Separated oligosaccharides, as well as several separate molecular weights such as GO, G1, G2, G2S1 and G2S2, can be detected and quantified by these means.

Biological characterization of glycogen variants Proteins containing Fe can be compared for functionality by several known in vitro assays. In particular, the affinity for the elements of the Fc? RI, Fc? RII, and Fc? RIII families of the Fc? It is of interest. These measurements could also be made using recombinant soluble forms of the receptors or the forms associated with the cells of the receptors. In addition, the affinity for FcRn, which is the receptor responsible for the prolonged circulating half-life of IgGs, can be measured, for example by BIAcore using recombinant soluble FcRn. Cell-based functional assays, such as ADCC assays and CDC assays, provide explanation as to the possible functional consequences of particular variant structures. In one embodiment, the ADCC assay is configured to have NK cells that are the primary effector cell, with which the functional effects of the FcγRIIIA receptor are reflected. Phagocytosis assays can also be used to compare the immune effector functions of different variants, since assays that measure cell responses such as release of superoxide or inflammatory mediator. The in vivo models can be used as for example in the case of the variants using the anti-CD3 antibodies to measure the activation of the T cells in mice, an activity that depends on the Fe domains binding the specific ligands such as the Fc? .

Protein production procedures The different procedures involved with the production of Fe-containing proteins can cause an impact on the structure of the Fe oligosaccharide. For example, the host cells that secrete the Fe-containing protein are cultured in the presence of serum, example fetal bovine serum Fe that was not previously subjected to a treatment by high heating (for example, 56 ° C for 30 minutes). This can result in a protein that does not contain, or has very low amounts of, sialic acid due to the natural presence in the soil of active exit enzyme that can remove sialic acid from proteins containing Fe secreted from those cells. In another embodiment, the cells that secrete the Fe-containing protein are cultured either in the presence of serum that was subjected to a treatment by high heating, thereby activating the exit enzymes, or in the absence of serum or other media components which may contain exit enzymes, such as Fe-containing proteins that have higher or lower levels of glycosylation or glycosylation variants. In another embodiment, the conditions used to further purify and process the Fe-containing proteins establish and favor the optimal glucan content. In another embodiment, the conditions produce maximum or minimum oligosaccharide content or cause the transformation of the Fe-containing polypeptide expressed in a predominant glyco- form. For example, because sialic acid is acid-labile, prolonged exposure to a low pH environment, such as the following elution of the protein A chromatography column or viral inactivation efforts can lead to a reduction in the content of sialic acid.

Host cell selection or host cell processing As described he, the host cell chosen for the expression of the recombinant Fe-containing protein or the monoclonal antibody is an important contributor to the final composition, including, without limitation, the variation in composition of portions of oligosaccharide that decorates the protein in the immunoglobulin CH2 domain. Therefore, one aspect of the invention includes the selection of host cells suitable for the use and / or development of a production cell that expresses the desired therapeutic protein. In an embodiment in which the content of sialic acid is controlled, the host cell is a cell that is naturally deficient or devoid of sialyltransferases. In another embodiment, the host cell is genetically modified to be devoid of sialyltransferase. In another embodiment, the host cell is a derivatized host cell line selected to express the reduced or non-detectable levels of sialyltransferases. Even in another embodiment, the host cell is naturally devoid of, or is genetically modified to be devoid of, sialic CMP acid synthetase, the enzyme that catalyzes the formation of sialic CMP acid which is the source of sialic acid used by the sialyltransferase to Transfer the sialic acid to the antibody. In a related embodiment, the host cell may be naturally devoid of, or genetically modified to be devoid of, pyruvic acid synthetase, the enzyme that forms sialic acid from pyruvic acid. In a further embodiment, the host cell may be naturally devoid of, or genetically modified to be devoid of, galactosyl transferases, such as the antibodies expressed in said cells lacking galactose. Without galactose, sialic acid will not bind. In a separate embodiment, the host cell may be overexpressed naturally or may be genetically modified to overexpress, a sialidase enzyme that removes sialic acid from the antibodies during production. Such a sialidase enzyme can act intracellularly on antibodies before the antibodies are secreted or secreted into the culture medium and act on antibodies that have already been secreted into the medium and can also contain a galactase. Methods for selecting cell lines with altered glycosylases and which express glycoproteins with altered carbohydrate compositions have been described in (Ripka and Stanley, 1986. Somatic Cell Mol Gen 12: 51-62; US2004 / 0132140). Methods of making host cells to produce antibodies with altered glycosylation patterns result in improved ADCC which has been shown in, for example, U.S. Patent 6,602,864, whe the host cells harbor a nucleic acid encoding at least one transferase from glycoside modifying the glycoprotein, specifically β (1, 4) -N-acetylglucosaminyltransferase III (GnTIII).

Other approaches to genetically engineer the glycosylation properties of a host cell through manipulation of the host cell glycosyltransferase include eliminating or suppressing the activity, as shown in EP1, 176,195, specifically, alpha 1, 6 fucosyltransferase (FUT8 gene product). . Any person skilled in the art would know how to practice the methods of preparing host cells in other specific examples cited above. In addition, the elaborated host cell may be of mammalian origin or selected from COS-1, COS-7, HEK293, BHK21, CHO, BSC-1, Hep G2, 653, SP2 / 0, 293 HeLa, myeloma, lymphoma, yeast, insect or plant cells, or any derivative, cell immortalized or transformed thereof. In another embodiment, the method for suppressing or eliminating the activity of the enzyme required for oligosaccharide binding may be selected from the group consisting of quenching the gene, such as by the use of siRNA, genetic knock-out or adding an enzyme inhibitor such as by the coexpression of an intracellular Ab or a peptide specific for the enzyme that binds and blocks its enzymatic activity and other known genetic manufacturing techniques. In another embodiment, a method for improving the expression or activity of an enzyme that blocks the binding of saccharides, or a saccharidase enzyme that removes sugars that have already been bound can be selected from the groups consisting of: transfections with genes from recombinant enzyme, transcription transcription factors that improve the synthesis of enzymatic RNA and genetic modifications that improve the stability of enzymatic RNA, all leading to the enhanced activity of enzymes such as sialidases, which result in lower levels of sialic acid in the product purified. In another embodiment, the specific enzyme inhibitors can be added to the cell culture medium. Alternatively, the host cell may be selected from a species or organism that is not capable of glycosylating the polypeptides, for example a prokaryotic cell or organism such as and from the native or elaborated E. coli spp., Klebsiella spp., Or Pseudomonas spp.

Antibodies An antibody described in this application may include or may be derived from any mammal, such as, but not limited to, a human, a mouse, a rabbit, a rat, a rodent, a primate or any combination thereof and includes isolated antibodies of human, primate, rodent, mammalian, chimeric, humanized and / or grafted with CDR, immunoglobulins, breaking products and other specific portions and variants thereof. The invention also relates to an antibody encoding or complementary nucleic acids, vectors, host cells, compositions, formulations, devices, transgenic animals, transgenic plants and methods for making and using same, as described herein combined as what is known In the art The described Fe fusion protein antibodies can be derived in different ways known in the art. In one aspect, the antibodies are conveniently obtained from hybridomas prepared by immunizing a mouse with the target peptides. The antibodies can therefore be obtained using any of the hybridoma techniques known in the art, see for example, Ausubel, et al., Ed., Current Protocols in Molecular Biology, John Wiley & Sons, Inc., NY, NY (1987-2001); Sambrook, et al., Molecular Cloning: A Laboratory Manual, 2nd Edition, Cold Spring Harbor, NY (1989); Harlow and Lane, antibodies, to Laboratory Manual, Cold Spring Harbor, NY (1989); Colligan, et al., Eds., Current Protocols in Immunology, John Wiley & Sons, Inc., NY (1994-2001); Colligan et al., Current Protocols in Protein Science, John Wiley & Sons, NY, NY, (1997-2001), each one completely incorporated as a reference. The Fe fusion protein antibodies or the components and domains thereof can also be obtained by selecting from libraries of such domains or components, for example a phage library. A phage library can be created by inserting a library of random oligonucleotides or a library of polynucleotides containing sequences of interest, such as from the B cells of an immunized animal or human (Smith, GP 1985, Sciences 228: 1315-1317 ). Antibody phage libraries contain heavy (H) and light chain (L) variable region pairs in a phage that allows expression of the Fv fragments of a chain or Fab fragments (Hoogenboom, et al., 2000, Immunol. 21 (8) 371-8). The diversity of a phagemid library can be manipulated to increase and / or alter the immunospecificities of the monoclonal antibodies of the library to produce and subsequently identify additional and desired human monoclonal antibodies. For example, the heavy chain (H) and light chain (L) immunoglobulin molecule encoding the genes can be randomly mixed (shuffled to create new HL pairs in an assembled immunoglobulin molecule). Additionally, each or both of the H and L chains encoding the genes can be mutagenized in a complementarity determining region (CDR) of a variable region of the immunoglobulin polypeptide and subsequently examined for desirable affinity and neutralization capabilities. Antibody libraries can also be created synthetically by selecting one or more human framework sequences and introducing CDR cassette collections derived from human antibody repertoires or through designated variations (Kretzschmar and von Ruden 2000, Current Opinion in Biotechnology, 13 : 598-602). The diversity positions are not limited to CDR, but may also include the framework segments of the variable regions or may include other variable regions of antibodies such as peptides. Other libraries of target link components which may include other variable regions of antibodies are ribosome displays, yeast deployments and bacterial displays. Ribosome display is a method of translating mRNAs into their cognate proteins while keeping the protein bound to the RNA. The nucleic acid encoding the sequence is coated by RT-PCR (Mattheakis, L.C. et al., 1994. Proc. Nati.

Acad. Sci. USA 91, 9022). The yeast deployment is based on the construction of fusion proteins of the associated membrane alpha-agglutinin yeast adhesion receptor, agal and aga2, a part of the mating type system (Broder, et al., 1997. Nature Biotechnology, 15 : 553-7). The deployment of bacteria is based on the fusion of the target to export the bacterial proteins that are associated with the cell membrane or the cell wall (Chen and Georgiou 2002. Biotechnol Bioeng 79: 496-503). Compared with hybridoma technology, phage and other methods of antibody deployment provide the opportunity to manipulate the selection against the target antigen in vitro and without limiting the possibility of the host effects of the antigen or vice versa. Also disclosed is a method for producing an antibody or Fe fusion protein comprising translating the encoding nucleic acid under conditions in vitro, in vivo or in situ, such as the peptide or antibody that has been expressed in detectable or recoverable amounts. Since the invention has been described in general terms, the embodiments of the invention will be described in more detail in the following examples.

EXAMPLE 1 Isolation of the Fe domain from IGG Papain was obtained from Sigma and PNGase F (peptide N-glycosidase F) was obtained from New England Biolabs. The sinapic acid was obtained from Fluka. The MALDI-TOF-MS analyzes were carried out using a Voyager DE biospectrometry workstation (Applied BioSystems, Foster City, CA). The antibody samples were deglycosylated by treating them with PNGase F in pH buffer Tris-HCl 20 mM, pH 7.0. The deglycosylated antibody samples were purified on a protein A column (HiTrap protein A cartridges were obtained from Amersham Biosciences) and analyzed by MALDI-TOF-MS for purity. Antibody samples (at ~ 1 mg / ml, before and after deglycosylation) were treated with papain (1: 50, w / w) in pH buffer Tris-HCl 20 mM, pH 7.0, containing 2 mM L -cysteine and aliquots were discarded at fixed time intervals (0, 15, 30, 60, 90 minutes followed by 2, 3, 4, 5, 6, 8 and 24 hours). The aliquots (approximately 2 μl) were immediately mixed with 2 μl of the matrix solution (the matrix solution was prepared by dissolving 10 ml of sinapic acid in 1.0 ml of 50% acetonitrile in water containing 0.1% trifluoroacetic acid ) and 2 μl of this solution was loaded onto a MALDI target plate and allowed to air dry before analysis.

The MALDI-TOF-MS data indicate that deglycosylation of IgG under natural conditions with PNGase F is complete and non-destructive (see Figures 4A to 4G). MALDI-TOF-MS analysis of discarded aliquots at fixed time intervals showed that most of the deglycosylated IgG was digested by papain within one hour while complete digestion of the control (glycosylated) IgG took more than 4 hours. The Fe fragments (at 10 KDa and 23.7 KDa) of deglycosylated IgG were observed in 30 minutes in the papain digestion and the majority of deglycosylated Fe was digested into fragments within 4 hours. The Fe fragments of glycosylated IgG were observed after only 4 hours of digestion with papain at 1: 50 (w / w) enzyme with substrate ratio and required more than 24 hours to completely convert the Fe into smaller fragments at 10 KDa and 23.7 KDA. During attempts to separate the Fe fragments from different IgG antibodies it was observed that the previous shedding of CH2 domain glycans increased the rate of papain mediated degradation of the Fe domain making it harder to obtain intact Fe from deglycosylated IgG (or aglycosylated). Subsequent time course experiments compared with glycosylated and deglycosylated sprays of IgG antibodies and purified Fe domains showed that, in both cases, the Fe degradation rate in the deglycosylated molecules was at least 4 to 8 times faster compared to their glycosylated counterparts. These results indicate that the presence of CH2 domain glycans increase the resistance to papain-mediated degradation of Fe domains. It is also suggested that, since IgG lacking glycosylation does not appear to have a defined Fe structure and does not link Fe receptors. , the sensitivity of papain can be an additional means to assess the appropriate Fe structure.

EXAMPLE 2 Digestion of papain from homogeneous glycoforms To assess the papain digestion parameters of substantially homogenous antibody preparations with respect to their glycosylation patterns, the antibody samples are enzymatically modified to produce such preparation for testing as described below. To galactosylate the purified antibody samples by means of the enzymatic method, the ß-1,4-galactosyltransferase (β1, 4GT) and bovine UDP-Gal obtained from Sigma Chemical Co. (St. Louis, MO) were added to the samples of antibody. Recombinant rat liver -2,3-sialitransferase (a2.3ST), recombinant a-1,3-galactositransferase (a1, 3GT) and CMP-Sia were obtained from Calbiochem (San Diego, CA). PNGase F was obtained from New England Biolabs (Beverly, MA) or from Prozyme (San Leandro, CA) or from Selectin BioSciences (Pleasant Hill, CA). β-Galactosidase and ß-glucosaminidase from Diplococcus pneumoniae was obtained from ProZyme or Selectin BioSciences. β-Galactosidase from the bovine kidney and other enzymes were obtained from ProZyme or Selectin BioSciences. Protein columns A NAP-5 and HiTrap were obtained from Pharmacia Biotech (Piscataway, NJ). All other reagents were analytical grade. Test antibodies for these assays include monoclonal IgG Abs with human IgG1 and kappa constant regions expressed in Sp2 / 0 mouse myeloma cells. The MAbs are completely human or a specific mouse / human chimeric MAb for, for example, human TNF. Fully galactosylated but not sialylated biantennary structures are designated G2 glyco- forms. G2 antibodies were prepared by attaching IgG samples in a 100 mM MES pH buffer (pH 7.0) (-10 mg in 1.0 ml of pH buffer) to 50 milliunits of β1, 4GT, 5 μmol of UDP-Gal and 5 μmol of MnCl2 at 37 ° C for 24 hours. Another aliquot of the enzyme was added and the mixture was incubated for an additional 24 hours at 37 ° C. The delelactosylated UDP-Gal samples were purified using a protein A IgG column. The oligosaccharides were released by HiTrap and characterized by PNGase F and by MALDI-TOF-MS as described below. It was found that the resulting preparations of Mabs contain 0% sialic acid. Fully sialized and galactosylated antibodies are designated G2S2. The G2S2 glyco- form was made by taking IgG samples in a pH buffer MES 100 mM, pH 7.0, (or 10 mg in 1.0 ml of pH buffer) using NAP-5 columns according to the manufacturers' suggested protocol. To this solution were added 50 milliunits each of β1 4GT and a2.3ST and 5 μmol each of UDP-Gal, CMP-Sia (isomer NSNA) and MnCl2. The mixture was incubated at 37 ° C. After 24 hours, another aliquot of enzyme was added together with the nucleotide sugars and the mixture was incubated for an additional 24 hours at 37 ° C. The glycoprotein G2S2 of sample IgG was purified as described above. Using this method, the sialation of an antibody preparation reached 90 to 98% of G2S2 glyco- form. The test Abs were analyzed structurally by different methods. To carry out the MALDI-TOF-MS analysis of samples of IgG Abs, intact IgG were taken in pH buffer Tris-HCl 10 mM, pH 7.0 and concentration adjusted to pH regulator of ~ 1 mg / ml. Approximately 2 μl of IgG solution was mixed with 2 μl of the matrix solution (the matrix solution was prepared by dissolving 10 mg of sinapinic acid in 1.0 ml of 50% acetonitrile in water containing 0.1% trifluoroacetic acid) and 2 ml of this solution was loaded onto the objective and allowed to air dry. MALDI-TOF-MS was purchased using a Voyager DE instrument from Applied BioSystems (Foster City, CA). To perform the MALDI-TOF-MS analysis of released Fe glycans, the IgG samples (-50 μg), before and after in in vitro glycosylation reactions, were digested with PNGase F in a pH regulator 10 mM Tris-HCl (50 μl), pH 7.0 for 4 hours at 37 ° C. Digestion was stopped by acidifying the reaction mixture with 50% acetic acid (-5 μl) and then passing it through a column of cation exchange resin as described above (Papac et al., 1996; Papac et al., 1998; Raju et al., 2000). These samples containing a mixture of acidic and neutral oligosaccharides were analyzed by MALDI-TOF-MS in positive and negative ion modes, as described (Papac et al., 1996; Papac et al., 1998; Raju et al., 2000) using a Voyager DE instrument from Applied BioSystems (Foster City, CA). HPLC analysis of Fe glycans was performed by ingesting IgG samples (50 μg) in pH buffer 10 mM Tris-HCl (50 μl) pH 7.0 with PNGase F at 37 ° C for 4 to 8 hours. Derivatization of liberated oligosaccharides with anthranilic acid (2-aminobenzoic acid) was carried out as described in (Anumula KR, Anal Biochem 2000 Jul 15; 283 (1): 17-26). Briefly, a solution of 4% sodium acetate 3H20 (w / v) and 2% boric acid (w / v) in methanol was prepared first. The active derivatization agent was prepared fresh by dissolving -30 mg of anthranilic acid (Aldrich) and -20 mg of sodium cyanoborohydride (Aldrich) in 1.0 ml of methanol-sodium-sodium-borate solution. Oligosaccharides derived from IgG- (> 3 nmoles in 20-50 μl of water) were mixed with 0.1 ml of anthranilic acid (AA) reactive solution in 1.6 ml of polypropylene screw cap freezing bottles with "O" ring (Sigma) and tightly capped. The bottles were heated to 80 ° C in a heating oven or blocker (Reacti-Therm, Pierce) for 1 to 2 hours. After cooling the flasks to room temperature, the samples were diluted with water to give a volume of -0.5 ml. Derivatized oligosaccharides were purified using NAP-5 columns. Using a preparation that is at least 90% in the G2 or G2S2 glycoform, a papain digestion experiment is performed and analyzed as described in example 1 by demonstrating the effect of the sialic acid content on the breakage ratio and specificity of papain. Additional breakdown analyzes were performed with other proteolytic enzymes using the samples as prepared in this example.

EXAMPLE 3 Production of antibody fragments using matrix metalloproteinase 3 The metalloproteinases (MMPs) were purified from the supernatant of cell clones expressing recombinant human MMPs. The enzyme was activated with 1 mM of 4-aminophenylmercuric acetate (APMA, Sigma) for 1 hour at 37 ° C or when treated with chymotrypsin. The activated enzyme was stored at -70 ° C. Immunoglobulin preparations (0.5-1.0 mg / ml) were incubated with the pH regulator of digestion (250 mM Tris-HCl, pH 7.4, containing 1.5 M NaCl, 50 mM CaCl2 containing 15-60 μg / ml activated MMP ) for 0 to 24 hours at 37 ° C. The aliquots were removed at 0, 15, 30, 45, 60, and 120 minutes followed by 3, 4, 5, 6, 8, 12, and 24 hours. The aliquots were (approximately 2 microliters) with matrix solution (approximately 2 microliters) and 2 microliters of this mixture was loaded onto the MALDI-TOF-MS target plate and then analyzed by MALDI-TOF-MS using a Voyager DE spectrometer.

EXAMPLE 4 Proteolytic cleavage of purified FC Papain was obtained from Sigma and PNGase F (peptide N-glycosidase F) was obtained from New England Biolabs. The sinapic acid was obtained from Fluka. The MALDI-TOF-MS analyzes were carried out using a Voyager DE Bioespectrometry workstation (Applied BioSystems, Foster City, CA). The antibody samples (IgG) were deglycosylated by treating them with PNGase F in a pH buffer of 20 mM Tris-HCl, pH 7.0. The deglycosylated antibody samples were purified on a protein A column (HiTrap cartridges of Protein A were obtained from Amersham Biosciences) and analyzed by MALDI-TOF-MS for purity. The Fe fragments of IgG were separated and purified as described. Samples of Fe fragment (approximately 1 mg / ml, before and after deglycosylation) were treated with papain (1: 50, w / w) in pH buffer 20 mM Tris-HCl, pH 7.0, containing 2 mM L -cysteine and the aliquots were removed at fixed time intervals (0, 15, 30, 60, 90, minutes followed by 2, 3, 4, 5, 6, 8 and 24 hours). The aliquots (approximately 2 μl) were immediately mixed with 2 μl of matrix solution (the matrix solution was prepared by dissolving 10 mg of sinapic acid in 1.0 ml of 50% acetonitrile in water containing 0.1% trifluoroacetic acid) and 2 μl of this solution was loaded onto the MALDI target plate and allowed to air dry before analysis. The MALDI-TOF-MS data of digested glycosylated and deglycosylated IgG were analyzed to obtain the peak area data of relative% of intact IgG and fragments of Fe at all points in time (Figure 5A and Figure 5B). For IgG samples, time-course experiments indicated that in 15 minutes of digestion, more than 70% of the deglycosylated IgG were converted into Fab, Fe, and small fragments of Fe (in m / z 10.5 KDa and 12 KDa fragments) whereby less than 50% of the glycosylated IgG was converted into fragments. After 60 minutes of incubation, no deglycosylated IgG was detectable, so that approximately 80% of the glycosylated IgG was still intact in accordance with the MALDI-TOF-MS analyzes. For the fragments of Fe, after 4 hours of digestion, more than 95% of the deglycosylated Fe fragment became small fragments of 10.5 and 12 KDa, so no more than 10% of glycosylated Fe became these small fragments at the point of time. In fact, almost 50% of glycosylated Fe remained undigested even after 24 hours. These data indicate that glycosylated IgG is significantly more resistant to digestion with papain than deglycosylated IgG, and that glycosylated Fe is much more resistant than deglycosylated Fe. The time-course experiment also revealed that the amount of deglycosylated and glycosylated IgG Fab fragments were equivalent, suggesting that only the Fe fragments suffer overdigestion and conversion in the 10.5 KDa and 12 KDa fragments.

EXAMPLE 5 Preparation of mabs with specific glucoforms To better understand the role of oligosaccharides in increasing antibody resistance to papain, IgG preparations with homogeneous oligosaccharides GO, G2 and G2S2, were prepared using in vitro methods.

Preparation of GO qlucoform In order to prepare homogeneous GO glycogen, IgG samples with sialidase A were first treated to remove minor amounts of terminal sialic acid residues followed by treatment with β-galactosidase to remove the terminal β-galatose residues. IgG samples (10 mg in 1.0 ml) in pH buffer of 100 mM MES (pH 7.0) were treated with 100 milliunits each of sialidase A. { A. ureafaciens) and β-galactosidase (D. pneumoniae) for 24 hours at 37 ° C. After 24 hours, another aliquot of enzymes was added and incubated for an additional 24 hours at 37 ° C.

After purification on a protein A column, the resulting GO glucoform was characterized by MALDI-TOF-MS for intact mass (Figure 6A). The mass spectrum contained an individually charged molecular ion at m / z 147.7 KDa, a doubly charged molecular ion at m / z 73.9 KDa and a triple charged molecular ion at m / z 49.3 KDa. The mass spectrum also contained a 23.4 KDa ion representing the free light chain produced during laser desorption ionization. Mass spectral data indicated that the antibody was intact after enzyme treatment to modify the Fe glycans in homogeneous GO oligosaccharide. The modified oligosaccharide chain, released by treating the IgG samples with PNGase F, was analyzed by MALDI-TOF-MS in a positive mode using sDHB as a matrix (after purification using a cation exchange column) and also by HPLC (after of derivatizing with anthranilic acid using a reductive amination procedure as described by Anumula (1998 supra) The MALDI-TOF-MS analysis showed a molecular ion at m / z 1486.8 which corresponds to the molecular weight of a biantennary oligosaccharide of the fucosylated complex of the nucleus sodiated with GIcNAc residues Normal-phase HPLC analysis of AA-oligivaccharide from rivatization produced a single peak eluting at 20.5 minutes and corresponding to the elution time of the AA-labeled standard nucleus fucosylated core bicarbonate oligosaccharide terminating with GIcNAc residues (data not shown) indicating that the glycogen IgG GO sample contained more than 99% of the GO oligosaccharide.

Preparation of the glycoprotein G2 The IgG samples were first treated with sialidase A and purified as described above. IgG samples treated with sialidase A (10 mg in 1.0 ml) in pH buffer of 100 mM MES (pH 7.0) were treated with 50 milliunits of β1, 4GT, 5 μmol of UDP-Gal, and 5 μmol of MnCl2 at 37 ° C for 24 hours. Another aliquot of enzyme and UDP-Gal was added and the mixture was incubated for an additional 24 hours at 37 ° C. After purification on a protein A column, the antibody sample was analyzed by MALDI-TOF-MS for intact mass. The mass spectrum showed an individually charged molecular ion at m / z 148.7 KDa, a doubly charged molecular ion at m / z 74.2 KDa and a triple charged molecular ion at 49.5 KDa. The mass spectrum also contained a molecular ion individually charged at m / z 23.4 KDa due to the free light chain produced during laser desorption ionization. The glycogen G2 was then subjected to PNGase F treatment to release bound N oligosaccharides and the released oligosaccharides were analyzed by MALDI-TOF-MS in the positive mode using sDHB as the matrix. The mass spectrum showed a molecular ion at m / z 1812.1 (Figure 6B) and this molecular ion at m / z 1812.1 corresponds to the molecular weight of the biantennary oligosaccharide of fucosylated complex of the sodiated core terminated with galactose residues. Normal-phase HPLC analysis of the oligosaccharides released with PNGase F after derivatization with AA showed a single peak and the elution time of this peak corresponds to the elution time of the standard G2 oligosaccharide indicating that the G2 glycophor preparation contained an intact IgG molecule with more than 99% of the G2 oligosaccharide.

Preparation of the G2S2 glyco- form For the preparation of the G2S2 glyco- form, the antibody samples were treated with a mixture of β-galactosyltransferase and a2,3-sialyltransferase in the presence of UDP-Gal, CMP-NANA and MnCl2 in a single step. The IgG samples were taken in a pH buffer of 100 mM MES (pH 7.0) (10 mg in 1.0 ml) using columns of NAP-5 in accordance with the protocol suggested by the manufacturer. To this solution, 50 milliunits each of ß1, 4GT and a2.3ST and 5 μmol each of UDP-Gal, CMP-Sia and MnCl2 were added. The mixture was incubated at 37 ° C. After 24 hours, another aliquot of enzymes was added together with the nucleotide sugars and the mixture was incubated for an additional 24 hours at 37 ° C. MALDI-TOF-MS analysis of the enzyme-treated sample, after purification on a protein A column, showed an individually charged molecular ion at m / z-148.8 KDa, a doubly charged molecular ion at -74.3 KDa and an ion molecularly triply charged at -49.5 KDa m / z (Figure 6C) suggesting that the antibody was intact and the enzymatic treatment did not alter the primary structure of the antibody. The MALDI-TOF-MS in negative mode and the HPLC analysis of the oligosaccharide released by PNGase F (FIGS. 7A-7D) showed that the antibody contained more than 85% of the G2S2 glyco- form together with minor amounts of the G2S1 structure (monosialated structures ). In addition, 99% of the Fe glycans contained at least one sialic acid residue. Figure 7A shows the control (Voyager Spec # 1 => NF0.7 => NR (2.00) [BP = 1485.0, 6636]), Figure 7B shows the glyoform G2 (Voyager Spec # 1 => NF0. 7 => NR (2.00) [BP = 1811.4, 5403]), Figure 7C shows the GO glycogen (Voyager Spec # 1 => NF0.7 => NF0.7 => NR (2.00) [BP = 1486.5, 7006]) and Figure 7D shows the glyco- form G2S2 (mode-ve) (Voyager Spec # 1 => NF0.7 [BP = 2385.4, 2769]).

Size exclusion chromatography To evaluate the amounts of aggregates present in the antibody samples before and after the in vitro modification of the glucan structures, the Ab samples were analyzed by size exclusion chromatography using an Agilent 1100 HPLC system ( Agilent). A TASO HAAS TSK 3000SWXL (Tosoh Biosep LLC) 7.8 mm x 30 cm, 5 μm column was used at room temperature. The mobile phase was phosphate buffered saline (PBS), and the flow rate was 0.5 ml / min. Modified IgG glucoform preparations presented similar chromatographic profiles for the control antibody profile indicating that the modification procedures did not create any aggregation and / or additional loading on the primary structure of the protein.

EXAMPLE 6 Digestion with papain from mabs with specific glucoforms To evaluate the relative resistance of the glycophodies GO, G2 and G2S2 for papain cleavage, IgG glucoforms and a control sample were treated with papain in the presence of cysteine at 37 ° C for a period of 24 hours and the digested was analyzed by MALDI-TOF-MS.

Time-course analysis of digestion with papain The GO, G2 and G2S2 glyco- form together with the IgG control samples were treated with papain at 1: 50, ratio of enzyme to substrate at 37 ° C. Of these reactions, the aliquots at 0, 15, 30, 60 and 90 minutes and 2, 3, 4, 5, 6, 8 and 24 hours were examined by MALDI-TOF-MS. These three IgG glyco- forms together with the control antibody sample were divided into Fab and Fe fragments, as evidenced by the presence of molecular ions at m / z ~ 47.3 and -52.5 KDa for Fab and Fe fragments, respectively (Figure 8). A comparison of the peak height of intact IgG observed at m / z ~ 147.5 KDa is shown in Figure 9. The intact IgG peak at m / z ~ 147.5 KDa was measured from 0 to 120 minutes, however, after 2 hours, each small intact IgG peak was observed. At minute 0, it was observed that the peak height of all IgG glyco- forms together with the control IgG was equal.

At 15 minutes, approximately 50% of GO, 45% of control and 35% of glycogen G2 remained undigested. In contrast, only about 25% of the G2S2 glyco- form remained undigested. At 30 minutes, approximately 45% of the GO glycogen, 40% of the control antibody and -20% of the G2 glycogen remained undigested, so that only about 10% of the G2S2 glycogen remained undigested. Therefore, the data presented in Figure 9 suggest that the G0 glyco- form is more resistant to the breakdown of papain in the CH1 domain, which produces Fab and Fe fragments as the primary products, than the other glucoforms. In addition to the primary papain cleavage site in the CH1 domain, IgGs can also undergo secondary cleavage in the CH2 domain of Fe by reduced papain. To examine the resistance of IgG gofoforms to digestion with papain at the secondary breaking site, the peak heights of the Fc fragments observed at m / z ~ 52.5 KDa were compared. The peak height data of the Fe fragments of G0, G2 G2S2 and control IgG is shown in Figure 10. The relative peak height of the Fe fragments of G2 and G2S2 from 0.25 hours (15 minutes) to 1 hour was about 5% greater than the relative peak height of the G0 glyco- form; the peak heights of the Fe fragments of the glycoprotein G0 and control IgG were almost similar. Thus, both intact IgGs such as G2 and G2S2 glyco- forms and the F2 product itself comprising the glyco- forms are more sensitive to digestion with papain. The data shown in Figure 10 therefore exemplifies the proportions of training and degradation competence of the Fe product: at a time point of 1.5 hours, the peak heights of the Fe fragments of all the glyco- forms and the control IgG They were almost the same. After a 1.5 hour time point, the peak heights of the Fe fragments of the G2 and G2S2 glyco- forms were gradually less than the peak height of the Fc fragments of the GO glycogen and control IgG. At the 6-hour time point, the peak height of the G2S2 glyco- form was about 30% less than the peak height of the GO glyco- form; The peak height of the G2 glyco- form was about 25% lower than the peak height of the GO glyco- form. At the 8-hour time point, the peak height of the Fe fragment of the G2S2 glyco- form was about 60% less than the peak height of the Fe fragment of the glyoform G0; the peak height of the Fe fragment of the glyco- form G2 was approximately 50% less than the peak height of the Fe fragment of the G0 glyco- form. After a 0.5 hour digestion, at all points in time the peak height of the G0 Fe fragment was higher than those of G2, G2S2 and control IgG. At a 24-hour time point, no appreciable Fe fragment was observed for the G2 and G2S2 glyco- forms, so approximately 70% of the G0 Fe fragments and control IgG were observed. These data indicate that the Fe fragment of the glycoprotein G0 is more resistant to papain digestion at the secondary cleavage site present in the CH2 domain of Fe. In addition, the data also suggest that the G2S2 glucoform is the most sensitive of the glucoforms to primary digestion in the CH1 domain and secondary digestion in the CH2 domain. These data suggest that the G2S2 glyco- form may be more sensitive to digestion with papain and the GO glyco- form may be more resistant to digestion with papain than the G2 glyco- form. Glucoform G2 was more resistant to digestion with papain than G2S2 glyco- form, but approximately 50% less resistant than GO glyco- form. These results suggested that there was a differential sensitivity of glycoforms IgG to digestion with papain. These differences in sensitivity appear to be both in a primary breakthrough site as well as in the secondary breakthrough site in the Faith. It will be clear that the invention can be practiced in other ways than those particularly described in the description and examples above.

Numerous modifications and variations of the present invention are possible in view of the foregoing teachings and, therefore, within the scope of the appended claims.

Claims (47)

NOVELTY OF THE INVENTION CLAIMS

1. - The use of a protein preparation containing glycosylated Fe, for the manufacture of a useful drug the treatment of a human disease determined by the release of a protease, wherein the antibody preparation is substantially homogeneous for a single glucoform.

2. The use as claimed in claim 1, wherein the Fe-containing protein is an antibody.

3. The use as claimed in claim 2, wherein the antibody is a therapeutic monoclonal antibody.

4. The use as claimed in claim 1, wherein the protease is selected from the group consisting of pepsin, a matrix metalloproteinase, trypsin, chymotrypsin, and a glycosylation modification enzyme.

5. The use as claimed in claim 4, wherein the matrix metalloproteinase is selected from the group consisting of matrix metalloproteinase-7 (MMP-7), neutrophil elastase (HNE), stromelysin (MMP-3) ), and elastase macrophage (MMP-12).

6. - The use as claimed in claim 4, wherein the glycosylation modification enzyme comprises β-galactosidase or sialidase A.

7. The use as claimed in claim 2, wherein the glucoform of the antibody is substantially in the form of GO glucoforma.

8. The use as claimed in claim 2, wherein the glucoform of the antibody is substantially in the glyco- form G2.

9. The use as claimed in claim 2, wherein the glucoform of the antibody is substantially in the G2S2 glyco- form.

10. The use as claimed in claim 1, wherein the disease to be treated is determined by the invasion of neutrophils at an affected site in the body.

11. The use as claimed in claim 1, wherein the disease to be treated is an autoimmune disease.

12. The use as claimed in claim 11, wherein the autoimmune disease is rheumatoid arthritis.

13. A method for altering the stability of a protein containing Fe to divide by a protease, which comprises modifying the amount of sialylated glyco- forms in the Fe-containing protein.

14. - The method according to claim 13, further characterized in that the Fe-containing protein comprises an antibody in an antibody preparation.

15. The method according to claim 14, further characterized in that the step of altering comprises modifying the antibody preparation such that the antibody is substantially free of sialylated glucoforms and the stability of the antibody increases.

16. The method according to claim 15, further characterized in that the step of modifying the antibody preparation is selected from the group consisting of culturing an antibody host cell with serum, preparing the antibody at a low pH, using a cell specific host, and treat with glycosylation modification enzyme.

17. The method according to claim 16, further characterized in that it comprises the step of modifying the antibody preparation so that the antibody is substantially homogeneous for a GO glyco- form.

18. The method according to claim 16, further characterized in that the modification step comprises treating the preparation of antibody with sialidase A.

19. The method according to claim 18, further characterized in that it also comprises the step of treating the preparation of antibody with β-galactosidase after treatment with sialidase A.

20. The method according to claim 15, further characterized in that the protease is selected from the group consisting of papain, ficin, bromolein, pepsin, metalloproteinase-7. of matrix (MMP-7), neutrophil elastase (HNE), stromelysin (MMP-3), macrophage elastase (MMP-12), trypsin, chymotrypsin and glycosylation modification enzymes.

21. The method according to claim 20, further characterized in that the protease makes contact with the antibody preparation in vitro.

22. The method according to claim 20, further characterized in that protease is papain.

23. The method according to claim 20, further characterized in that the protease makes contact with the antibody preparation in vivo.

24. The method according to claim 23, further characterized in that the protease is related to a pathological condition.

25. The method according to claim 24, further characterized in that the pathological condition is cancer.

26. A method for detecting or diagnosing a disease state in a cell or subject, comprising: determining the glycosylation state of Fe-containing proteins in the cell or in a biological sample of said subject; correlate the state of glycosylation with the presence or levels of protease; and correlating the presence or levels of protease with a disease state indicated by the presence or levels of the protease.

27. The method according to claim 26, further characterized in that the Fe-containing protein is an antibody.

28.- A method to evaluate the glycosylation of an antibody, which comprises: contacting the antibody with an enzyme, and monitoring the activity of the enzyme.

29. The method according to claim 28, further characterized in that it further comprises comparing the activity of the enzyme to a known activity of the enzyme in connection with a known antibody composition.

30. The method according to claim 28, further characterized in that the activity monitored is the resistance to breakage by the antibody.

31. The method according to claim 30, further characterized in that the resistance is determined by the presence of Fab, F (ab ') 2, Fv, facb, or fragments of Fe.

32.- The method according to claim 28, further characterized in that the enzyme is a protease.

33. - The method according to claim 32, further characterized in that the protease is selected from the group consisting of papain, pepsin, a matrix metalloproteinase which includes MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3), macrophage elastase (MMP-12), trypsin, chymotrypsin, and glycosylation modification enzymes.

34. The method according to claim 28, further characterized in that the evaluated glycosylation comprises the amount of glycosylation.

35. The method according to claim 28, further characterized in that the evaluated glycosylation comprises the glucoform content.

36.- A method for rapidly dividing an antibody into a Fab, F (ab ') 2, Fv, facb, or Fe, which comprises: preparing a substantially deglycosylated antibody; and contacting the substantially deglycosylated antibody with a protease.

37. The method according to claim 36, further characterized in that the protease is papain.

38.- A method for rapidly dividing an antibody into a Fab, F (ab ') 2, Fv, facb, or Fe, comprising: preparing an antibody with a specific glycophor composition; and contacting the antibody with a protease.

39.- The method according to claim 38, further characterized in that the protease is selected from the group consisting of papain, pepsin, a matrix metalloproteinase that includes MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3) , elastase macrophage (MMP-12), trypsin, chymotrypsin, and glycosylation modification enzymes.

40.- A method for rapidly targeting an antibody in multiple portions, comprising: preparing a substantially deglycosylated antibody; and contacting the substantially deglycosylated antibody with a protease.

41. The method according to claim 40, further characterized in that the protease is selected from the group consisting of papain, pepsin, a matrix metalloproteinase that includes MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3), macrophage elastase (MMP-12), trypsin, chymotrypsin, and glycosylation modification enzymes.

42.- A method for rapidly digesting an antibody in multiple portions, comprising: preparing an antibody with a specific glycophor composition; and contacting the antibody with a protease.

43.- The method according to claim 42, further characterized in that the protease is selected from the group consisting of papain, pepsin, a matrix metalloproteinase including MMP-7, neutrophil elastase (HNE), stromelysin (MMP-3) , elastase macrophage (MMP-12), trypsin, chymotrypsin, and glycosylation modification enzymes.

44.- The use of a protein containing deglycosylated Fe, for the manufacture of a drug useful for treating a human disease determined by a desire to treat with a protein containing Fe with a reduced half-life.

45. The use as claimed in claim 44, wherein the protein containing deglycosylated Fe is a protein preparation.

46. The use as claimed in claim 44, wherein the protein containing Fe is an antibody.

47. The use as claimed in claim 46, wherein the antibody is a therapeutic monoclonal antibody.