Patent Grant
7897153
The present application claims priority to European Patent Application No. 0522582.0, filed Oct. 17, 2005, which is incorporated herein by reference in its entirety.
The instant application inter alia relates to chimeric fusion constructs comprising the extracellular portion of human FcsRIa and at least one avian constant immunoglobulin domain and the use thereof for in vitro diagnostic purposes.
More than 20% of the population suffer from type I allergic reactions. The prevalence of asthma, hay fever and other IgE-mediated diseases has increased dramatically in industrialized countries over the last decades, making allergies a very serious public health problem (Beasley et al., J Allergy Clin Immunol 105: 466-472, 2000).
For example, asthma prevalence rates increased by 75% in the United States between 1980 and 1994 (Doyle, Sci Am 282: 30, 2000).
The most important predisposing factor for the development of allergic diseases is atopy, the genetic predisposition to produce allergen-specific IgE. Childhood asthma is mainly found in patients who are atopic and sensitized to common environmental allergens including house dust mites, tree and grass pollens. Recent epidemiological studies suggest a close correlation between serum IgE levels and the presence or severity of asthma (Burrows et al., N Engl J Med 320: 271-277, 1989). In contrast, patients with a non-allergic asthma have negative skin tests to common allergens and normal total Ige serum levels. The association of increased IgE serum levels is also obvious for other allergic diseases such as seasonal allergic rhinitis, insect venom allergy or food allergies.
Symptoms of type I allergic reactions are due to the release of mediators (e.g. histamine) resulting from allergen-mediated crosslinking of IgE (immunoglobulin E) antibodies bound to IgE receptors on effector cells. The high-affinity IgE receptor FcεRI is mainly expressed on mast cells and basophils (Ravetch and Kinet, Ann Rev Immunol 9: 457, 1991), whereas the low-affinity IgE receptor FcεRII (CD23) is expressed on B cells (Conrad, Ann Rev Immunol 8: 623-645, 1990). Free serum IgE has a very short half-life of about two and a half days, while mast cells remain sensitized for up to 12 weeks following binding of IgE to high-affinity receptors.
Treatment modalities of type I allergies have substantially improved over the last decades. Allergen-specific immunotherapy (SIT) represents a curative approach. A rise in allergen-blocking IgG antibodies, particularly of the IgG4 class (Reid et al., J. Allergy Clin. Immunol. 78: 590-600, 1986), a reduction in the number of mast cells and eosinophils, and a decreased release of mediators (Varney et al., J. Clin. Invest. 92: 644-651, 1993) were found to be associated with successful SIT. However, for some patients including polysensitized and very young patients SIT is not applicable. The most advanced treatment modality that targets the pathophysiological cascade of allergen-mediated immune reactions earlier and in a more general way than SIT, is the inhibition of IgE responses by anti-IgE antibodies. The binding site of IgE for the high-affinity IgE receptor FcεRI is located within the third domain of the heavy chain, Cε3. A murine antibody, MAE1, was generated that recognizes the same residues in the Cε3 domain of IgE that are responsible for binding to FcεRI (Saban et al., J Allergy Clin Immunol 94: 836-843, 1994). To avoid sensitization to the murine antibody, a humanized version, containing 95% of a human IgG1 antibody and only 5% of the murine antibody, was constructed and named recombinant humanized monoclonal antibody (rhuMAb)-E25 (Presta et al., J Immunol 151: 2623-2632, 1993) or omalizumab (Xolair®). The main features of this anti-IgE antibody include a) recognition and binding to serum IgE, but not to IgG or IgA, b) inhibition of IgE binding to FcεRI, c) no binding to IgE bound to mast cells or basophils, thereby avoiding degranulation (‘non-anaphylactic antibody’), and d) its capability to block mast cell degranulation upon passive sensitization in vitro or challenge with allergen in vivo. Treatment with omalizumab (Xolair®) also reduces the number of FcεRI receptors on basophils in atopic patients. Since omalizumab (Xolair®) binds to any IgE molecule irrespective of its allergen specificity, this therapeutic approach provides a valuable alternative to SIT. Recent clinical trial have demonstrated that anti-IgE is an effective agent for the therapy of allergies including moderate to severe allergic asthma and seasonal allergic rhinitis (Hamelmann et al., Allergy 57: 983-994, 2002; Hamelmann et al., Curr Opin Allergy Clin Immunol 3: 501-510, 2003).
In order to assess the success of anti-IgE treatment, methods for the in vitro measurement of IgE antibodies are required that allow differentiation between complexed and non-complexed serum IgE.
Currently available methods for the in vitro determination of serum IgE include the radio-allergosorbent test (RAST), various enzyme-linked immunosorbent assays (ELISA) and other IgE-binding techniques such as immunoelectrophoresis, immunoblot and immunodotblotting. RAST and ELISA assays are performed in three steps including immobilization of the standard/reference allergens on a solid phase (e.g., the well of a microtiter plate), binding of the IgE antibodies in the serum of a patient to the immobilized allergens, and determination of coupled IgE antibodies by labelled anti-IgE antibodies. All of these techniques, however, utilize for the detection of bound IgE polyclonal or monoclonal anti-IgE antibodies that are not capable of differentiating between complexed and non-complexed serum IgE. Since these antibodies recognize other epitopes than those residues in the Cε3 domain of IgE that are responsible for binding to FcεRI, free IgE as well as anti-IgE/IgE complexes are detected.
A new strategy for detecting allergen-specific antibodies in serum is the bead array technology. These multiplex assays can be performed in the flow cytometer or in any other similar analytical equipment that allows for the discrimination of different particles on the basis of size and color. The bead array technology employs a series of particles with discrete fluorescence intensities to simultaneously detect multiple soluble analytes from a single serum, plasma, or tissue fluid sample. The analytes can also be allergen-specific antibodies of the IgE subclass. For example, for the detection of allergen-specific antibodies specific capture beads carrying immobilized standard/reference allergens, are mixed with phycoerythrin-conjugated detection antibodies and are then incubated with test samples to form sandwich complexes. However, the polyclonal or monoclonal anti-IgE capture antibodies utilized for this technique are not suitable for the determination of non-complexed IgE due to the lack of specificity for the residues in the Cε3 domain of IgE that are responsible for binding to FcεRI.
In addition to solid-phase technology, fluid-phase systems have been established for the in vitro measurement of allergen-specific antibodies in serum. In these immunoassays, modified allergens are employed. The modified allergens contain one or more residues (e.g., biotin) that allow for subsequent binding of allergen-serum antibody complexes to a solid phase coated with a corresponding binding protein (e.g., streptavidin). The soluble polymer/copolymer support systems utilized in these assays, increase the number of binding sites and, thereby, the detection sensitivity. The most advanced fluid-phase assays for the detection of allergen-specific IgE antibodies utilize an enzyme-enhanced chemiluminescent enzyme immunoassay technique for the quantification of complexed specific antibodies. For example, for the detection of allergen-specific IgE antibodies, streptavidin-coated beads, biotinylated liquid allergens, and the patient's sample are incubated, and after a spin wash coupled IgE antibodies are detected with an alkaline phosphatase-labelled monoclonal anti-human IgE antibody using a chemiluminescent substrate (e.g., phosphate ester of adamantyl dioxetane). Again, the labelled monoclonal anti-human IgE antibody utilized for these fluid-phase systems detects free IgE as well as anti-IgE/IgE complexes.
One possibility to overcome these problems is the use of monoclonal antibodies with specificity for the residues in the Cε3 domain of IgE that are responsible for binding to FcεRI such as murine monoclonal antibody MAE1 or humanized monoclonal antibody E25 (omalizumab, Xolair®) as capture and/or detection antibody for the above listed IgE quantification procedures. However, the use of antibodies of mammalian origin raises a number of additional difficulties in practice due to the presence of rheumatoid factor (RF) and human anti-mouse IgG antibodies (HAMA) in serum samples. RF and HAMA are probably the most well known causes of false positive or false negative reactions in immunological assays (Boscato L M, Stuart M C, Clin Chem 34, 27-33, 1988). RF is an auto-antibody that reacts with the Fc part of IgG.
The disease most often associated with RF is rheumatoid arthritis, but RF can be found in serum from patients with many other diseases and also in 3-5% of healthy blood donors (Johnson P M, Faulk W P, Clin Immunol Immunopathol 6, 414-430, 1976). Production of HAMA is mainly the result of therapeutic approaches with mouse monoclonal antibodies, but HAMA may also be found in serum from patients who have not been treated with antibodies. RF or HAMA may react with both the capture antibody and the detection antibody in a sandwich assay, thereby mimicking antigen activity. A reaction with the detection antibody results in formation of an immune complex which may influence the activity of the detection antibody. HAMA may also react with the antigen-binding region of the detection antibody, thereby impairing or inhibiting antigen binding. The problem of RF and HAMA interference will increase as the sensitivity of the assay increases. Interference by anti-IgG antibodies and antibody-binding substances have been demonstrated in approximately 40% of serum samples from healthy individuals in an immunoradiometric assay (Boscato L, Stuart M, Clin Chem 32, 1491-1495; 1986). The prevalence of human anti-mammalian antibodies causing potential interferences in immunological assays varies from 1-80% in the general population (Kricka L J, Clin. Chem. 45, 942-956, 1999).
Furthermore, some mammalian IgG antibodies bound to a solid phase as well as antigen-antibody complexes comprising such antibodies, can activate the human complement system (Larsson A, Sjoquist J, J Immunol Methods 119, 103-109, 1989). Activated C4 molecules bind to the Fab region of IgG and may interfere with the antigen binding (Campbell R D, et al., Biochem J 189, 67-80, 1980). In clinical laboratories, most analyses are performed on serum samples. A newly obtained serum sample contains active complement, but the activity declines during storage and handling. Accordingly, the complement activity may vary between different patients and also between different samples from the same patient. To avoid activation of the complement cascade, EDTA is often included in tubes used for blood sampling.
EDTA prevents complement activation and coagulation by sequestering calcium ions. Most of the standards and controls used have been stored and contain an inactive complement system. This difference in activity between the samples and the standards will cause erroneous results. Complement activation was shown to interfere in an immunometric TSH assay and depressed the TSH values by up to 40% (Kapyaho K, et al., Scand J Clin Lab Invest 49, 211215, 1989).
In principle, the above mentioned problems, e.g. cross-reactivity and complement activation, could be avoided by using mammalian antibody fragments instead of complete antibodies. For example, IgG antibodies can be enzymatically cut by papain into Fab fragments (fragment antigen binding) and Fc fragments (fragment cristallizable). Furthermore, recombinant production of Fab fragments is possible. In a preferred form, light and heavy chain domains are formed by a single peptide chain, which can be recombinantly generated (scFv, single chain fragment variable). Libraries of scFv, in particular as phage display libraries, are available in the art, which facilitate generation of recombinant antibodies or scFv specific for a given antigen. One major limitation of scFv or Fab molecules, however, is their monovalent format, impairing the affinity of these molecules and, thereby, their applicability for analytical applications. Alternatively, bivalent Fab2 fragments could be used, which still contain the hinge region, wherein the two heavy chains are connected by a disulfide linkage. Fab2 can be produced by enzymatic digestion of antibodies with pepsin. However, especially for analytical applications, another major limitation of scFv, Fab and Fab2 molecules is the lacking Fc region of the heavy chain. As a result, recognition of theses antibody fragments by secondary antibodies to the Fc region is severely impaired.
As an alternative to the above listed assay systems, basophil granulocytes have been used for the detection of allergen-specific antibodies in serum. Blood basophils together with submucosal mast cells are primary effector cells in IgE-mediated immediate-type allergic reactions such as allergic rhinitis, allergic asthma, IgE-mediated urticaria or anaphylactic shock.
The principle of the method is to challenge sensitized basophils, believed to be sensitized analogously to skin mast cells, with allergen which will cross-link surface-bound specific IgE causing histamine to be released from the cells (in vitro mediator release assay; MRA). Released histamine is determined and a dose-response curve can be constructed. In order to ensure that the basophils are responding properly, anti-IgE antibodies are applied as positive control, whereas the test substance is applied on basophils carrying no specific IgE to exclude non-specific histamine release. The histamine release tests have been shown to correlate well with other methods for in vitro measurement of allergen-specific antibodies in serum and with skin prick tests (Østergaard et al., Allergy 45: 231, 1990). More recently, the synthesis and release of leukotrienes has been applied to monitor the response of basophils upon allergen stimulation. While this assay is suitable for the detection of allergen-specific antibodies in serum, the level of free IgE in serum cannot be determined with sensitized basophils since the binding sites of their Fcε receptors are already occupied.
Basophils could be used for the determination of free IgE in serum after stripping of their original IgE by a brief treatment at low pH (Pruzansky et al., J. Immunol. 131: 1949, 1983). The stripping step, however, can interfere with the biological function of these cells, which is likely to affect the reliability of the assay. Cord blood basophils which do not require the stripping step represent an alternative, but these cells are difficult to obtain. In principle, basophil cell lines such as the KU812 (Hara et al., Biochem. Biophys. Res. Commun. 247: 542, 1998) or animal cell lines transfected with the human FcεRI (Lowe et al., J. Immunol. Methods 184: 113, 1995) may be used as recipient cells for MRAs.
However, all cell-based assay systems pose important limitations for routine analyses since they are expensive, labor intensive, and difficult to standardize.
Application of a soluble derivative of the alpha chain of human FcεRI, also referred to as FcεRIα, as capture and/or detection reagent provides another possibility for establishing an IgE assay that allows differentiation between complexed and non-complexed serum IgE. The alpha chain of FcεRI binds IgE molecules with high affinity (KD of about 10−9 to 10−10 M). Prior investigators have disclosed the nucleic acid sequence for human FcεRIα as well as for a soluble fragment thereof (U.S. Pat. No. 4,962,035, by Leder et al.; U.S. Pat. No. 5,639,660, by Kinet et al.). By introduction of a stop codon before the single C-terminal transmembrane anchor, a soluble FcεRIα fragment of 172 amino acid residues has been expressed that mediates high affinity binding of IgE (Blank et al., J Biol Chem 266, 2639-2646, 1991). The dissociation rate of bound IgE from this soluble truncated receptor was comparable to that of FcεRI on intact cells. The extracellular portion of the human FcεRIα protein contains two immunoglobulin-like domains (D1 and D2) which belong to the truncated C2 subtype of the immunoglobulin superfamily (Kinet, Annu Rev Immunol 17: 931-972, 1999). There are seven N-linked carbohydrate attachment sites in the human FcεRIα molecule. They are distributed about the front and back of the molecule, but are not found on top of the molecule (Garman et al., Annu Rev Immunol 17: 973-976, 1999). Glycosylation of FcεRIα affects the secretion and stability of the receptor, but is not required for IgE-binding (Blank et al., J Biol Chem 266: 2639-2646, 1991). However, the deglycosylated receptor has a tendency to form oligomers and aggregates in solution (Robertson, J Biol Chem 268: 12736-12743, 1993; Scarselli et al., FEBS Lett 329: 223-226, 1993; Letourner et al., J Biol Chem 270: 8249-8256, 1995). Apparently, the carbohydrates reduce the affinity of the receptor for itself, thus preventing premature aggregation on the cell surface.
The alpha chain of human FcεRI has been used for the detection of human IgE and IgE from other species including canine, feline, and equine IgE (WO 98/23964).
However, the molecular mass of the soluble truncated FcεRIα-construct (unglycosylated) is less than 20 kDa and immobilization or labelling of this small protein is likely to impair its capability as capture and detection reagent due to steric hindrance problems. IgE is a bulky ligand and high affinity binding to FcεRIα requires unrestricted interaction of all involved amino acid residues.
Steric hindrance problems can be minimized by fusion of FcεRIα With a protein suitable for immobilization and detection. Especially mammalian constant immunoglobulin domains are suitable for this purpose since a variety of established techniques are available for site-directed immobilization and detection of immunoglobulins by species-specific antibodies. The IgE binding capacity of FcεRIα is not affected by fusion to immunoglobulin domains. Using an ELISA format, chimeric molecules comprising the extracellular portion of human FcεRIα and constant domains of the heavy chain of human IgG1 have been demonstrated to bind efficiently to immobilized human IgE (Haak-Frendscho et al., J Immunol 151: 351-358, 1993). As described earlier, however, the presence of mammalian immunoglobulin domains in capture or detection reagents can lead to false positive or false negative reactions in immunological assays.
The person skilled in the art is therefore faced with the need of generating improved capture and/or detection reagents that allow differentiation between complexed and non-complexed serum IgE and avoid or minimize the above mentioned problems in the use of mammalian antibodies and fusion proteins comprising FcεRIα and constant immunoglobulin domains of mammalian origin.
This problem is solved by the subject matter of the claims. In particular, the present invention provides a chimeric fusion construct comprising the extracellular portion of human FcεRIα and at least one avian constant immunoglobulin domain, in particular, at least one IgY constant domain. The format of these constructs can be monomeric or homodimeric. The constructs are useful for the determination of the serum level of human IgE and, in particular, the level of non-complexed IgE in the presence of IgE/anti-IgE complexes which allows to assess the success of anti-IgE treatment.
The chimeric constructs allow to combine the advantageous properties of IgY constant domains with the IgE binding characteristics of human FcεRIα. Due to the phylogenetic difference between avian and mammalian proteins, constructs comprising IgY constant domains eliminate assay problems associated with cross-reactivity of secondary mammalian antibodies, RF, HAMA, and complement activation.
Additionally, the present invention provides a chimeric fusion construct comprising a non-glycosylated derivative of the extracellular portion of human FcεRIα and non-glycosylated avian IgY constant immunoglobulin domains. The absence of N-glycans guarantees a lack of interference by C-type lectins in analytical applications and, thereby, a higher degree of reliability.
In another embodiment of the present invention nucleic acid sequences and vectors encoding the FcεRIα constructs are provided, as well as mammalian and prokaryotic host cells transfected with these sequences.
B: chicken upsilon heavy chain constant region coding region CH1-4: SEQ ID NO:33 (domains of the heavy chain constant regions are defined according to a conserved domain architecture retrieval tool (Geer et al., 2002).
C: homodimeric FcεRIα construct comprising IgY constant domains CH1-CH4 coding region: corresponding to SEQ ID NO:34 (amino acid sequence: SEQ ID NO:35).
A: Detection of Omalizumab-bound human IgE
B: Detection of human Myeloma IgE Standard
The present invention relates to chimeric fusion constructs (also termed “chimeric antibody constructs” or “FcεRIα —IgY constructs”, respectively) comprising the extracellular portion of human FcεRIα and at least one avian constant immunoglobulin domain, in particular, at least one IgY constant domain. The format of these constructs can be monomeric or homodimeric.
In the context of the present invention “extracellular portion of human FcεRIα” refers to sequences—both on nucleic acid and amino acid level—which correspond to or are derived from the sequences shown in
Avian IgY, the equivalent of mammalian IgG, is the major low molecular weight serum immunoglobulin in oviparous (egg laying) animals. It is called IgY rather than IgG to distinguish it from its mammalian counterpart (Leslie G A, Clem L W, J Exp Med 130, 1337-1352, 1969). The overall structure of IgY is similar to mammalian IgG, with two light and two heavy chains. The light chain is lighter and the heavy chain of the IgY molecule is larger than its mammalian counterpart. The molecular mass has been reported to be approx. 167 kDa, slightly larger than IgG (approx. 160 kDa) (Sun S, et al., Rapid Commun Mass Spectrom 15, 708-712, 2001). The H chain (approx. 65 kDa), called upsilon (υ, capital letter Y) has one variable (V) region and four constant (C) regions. The light chain (approx. 18 kDa) is composed of one variable and one constant domain. In place of the hinge region of mammalian IgG, IgY has a more rigid structure, giving IgY limited flexibility. The Cυ3 and Cυ4 of the IgY are most closely related to the Cγ2 and Cγ3 of IgG, respectively, and the Cυ2 domain is absent in the γ-chain. The Cυ2 region was probably condensed to form the hinge region of IgG as studies have shown that IgY is an ancestor to mammalian IgG and IgE and also to IgA (Warr G W, et al., Immunol Today 16, 392-398, 1995).
An IgY constant domain in the context of the invention refers to an immunoglobulin domain that has a homology of at least 50%, preferably at least 70%, more preferably at least 80% or at least 90%, most preferred at least 95% amino acid homology to the corresponding IgY domain. Thus, e.g., one, two or three amino acids can be deleted, inserted or substituted. In particular, conservative substitutions are contemplated. The boundaries between the domains are preferably defined as by the conserved domain retrieval tool (Geer et al., 2002), however, they can also be shifted.
Due to the phylogenetic difference between avian IgY and mammalian IgG, human FcεRIα constructs comprising IgY constant domains provide several important advantages for analytical applications. One important advantage is the lack of cross-reactivity with mammalian antibodies. For example, utilizing polyclonal mammalian antibodies as secondary anti-IgE antibodies for the detection of captured human serum IgE, these antibodies can cross-react with mammalian capture antibodies. Because chicken IgY constant domains are so different from mammalian IgG constant domains, potential cross-reactivity of secondary mammalian anti-IgE antibodies can be minimized by using constructs of human FcεRIα and IgY constant domains as capture reagent.
Another advantage of the phylogenetic difference between avian and mammalian species is the lack of human complement activation by IgY antibodies. Because the constructs of the invention do not activate the human complement system, they can be used in analytical assays to reduce interference by complement activation (Larsson A, et al., J Immunol Methods 156, 79-83, 1992).
Still another important advantage of the phylogenetic difference between avian and mammalian species is the lack or minimisation of interaction of the construct of the invention with the rheumatoid factor (RF), human anti-mouse IgG antibodies (HAMA), and other anti-mammalian antibodies potentially present in human sera.
The IgY constant domains do not interact with RF or HAMA and can thus be used to avoid interference due to these factors (Larsson A, et al., Clin Chem 37, 411-414, 1991; Larsson A, Mellstedt H., Hybridoma 11, 33-39, 1992).
Thus, the phylogenetic difference of avian IgY antibodies to IgG antibodies of mammalian origin, and the use of IgY constant domains in the FcεRIα constructs of the invention provides several important advantages for analytical applications including the lack of cross-reactivity with mammalian antibodies, the lack of human complement activation, and the lack of interaction with human anti-mouse antibodies and the rheumatoid factor.
Preferred Constructs
In the context of the invention “FcεRIα-IgY construct” refers to constructs comprising the extracellular portion of human FcεRIα (see definition above) and at least one avian constant immunoglobulin domains, in particular, an IgY constant domain or a combination of IgY domains, which are capable of forming (homo)dimeric molecules (
A preferred construct of the instant invention has the sequence shown in SEQ ID NO: 35.
For homodimeric constructs, the IgY domain or a combination of IgY domains selected from the group consisting of CH1, CH2, CH3, and CH4 IgY domains, are capable of forming dimeric molecules. The construct can comprise one, two, three or four IgY constant domains. According to one embodiment, the IgY domain is a CH2 constant domain. Preferably, the construct comprises a combination of IgY domains selected from the group consisting of (a) CH1 and CH2, (b) CH1, CH2 and CH3, (c) CH1, CH2, CH3 and CH4, (d) CH2 and CH3, (e) CH2, CH3 and CH4 IgY constant domains. Constructs comprising truncated IgY heavy chains provide a lower molecular mass than complete antibodies, which in many cases has been shown to be associated with increased expression rates. Instead of the conventional domains, fragments or variants thereof that do not inhibit folding or dimerisation of the constructs can be used. The boundaries between the domains, and thus the species origin of fragments at these boundaries can also be varied, as long as folding and dimerisation of the constructs is ensured.
In homodimeric constructs comprising the extracellular portion of human FcεRIα and the CH1 and CH2 IgY domains, or only the CH2 IgY domain, the amino acids E and F are fused to the carboxyterminal end of the IgY CH2-domain (3′ boundary). Constructs lacking the CH3 and CH4 domains require a stop codon at the 3′ boundary of the CH2 domain.
While a crystal structure of IgY constant domains is not available, the crystal structure of constant domains of mammalian IgE which is known to be close to chicken IgY in terms of the number of CH domains, the number of exons and exon lengths as well as the organization of intradomain and interchain disulfide bonds (Warr G W, et al., Immunol. Today 16, 392-398, 1995), has been published recently (Wan T, et al., Nat. Immunol. 3, 681-686, 2002). The analysis revealed an asymmetrically bent conformation of IgE with the CH2 domains forming highly bent structures at CH2-CH3 junctions. Therefore, truncation of mammalian IgE heavy chains at the CH2 domain is considered likely to fundamentally impair the folding and assembly process of IgE. As a result of the close relationship of mammalian IgE and chicken IgY antibodies, similar considerations apply to IgY antibody constructs lacking the CH3 and CH4 domains. On the other hand, ducks express a truncated version of IgY lacking the CH3 and CH4 domains, but the CH2 domain of this truncated duck antibody is followed by a terminal exon consisting of nucleic acids encoding EF and a stop codon. Utilizing single chain variable fragments (scFv) as model for the construction of FcεRIα-fusion proteins comprising the IgY CH2-domain followed at its carboxyterminal end by the amino acids E and F, it could be demonstrated that such constructs are expressed as homodimeric molecules.
Constructs Comprising the Extracellular Fragment of Human FcεRIα and Mono-Glycosylated or Non-Glycosylated IgY Constant Domains
Another limitation for analytical applications is the interaction of mammalian immunoglobulins and IgY antibodies via their oligosaccharide chains with C-type lectins in human sera and other human body fluids. C-type lectins bind sugars in a calcium-dependent manner via highly conserved carbohydrate recognition domains (CRD). C-Type lectins are either produced as transmembrane proteins, e.g., on dendritic cells and Langerhans cells (for a review, see Figdor C G, et al., Nature Rev. Immunol. 2, 77-84, 2002), or secreted as soluble proteins. Examples of soluble C-type lectins include the mannan-binding lectin (MBL), the surfactant protein A (SP-A) and D (SP-D), and the conglutenins CL-46 and CL-43, all of which belong to the collectin family (for a review, see Van de Wetering J K., et al., Eur. J. Biochem. 271, 1229-1249, 2004). MBL is secreted into the blood stream. The presence of substantial amounts of MBL in the small intestine suggests that this protein is acting as a humoral immune factor in the intestine, similar to secretory IgA. SP-A and SP-D are secreted at the luminal surface of pulmonary epithelial cells, and recently an increased expression of SP-D in the gastric mucosa during Heliobacter pylori infection has been reported, pointing to a role of the surfactant proteins in mucosal defense systems. The serum collectins conglutenin CL-46 and CL-43 have so far only be detected in bovidae, where the liver is their main site of production. The basic functional unit of collectins is a trimer, but the number of trimeric units per collectin molecule differs among the collectins. MBL and SP-A form octadecamers of six trimeric subunits, with their overall structure resembling a bouquet of flowers, whereas SP-D and the bovine conglutein proteins are assembled into dodecamers of four trimeric subunits. In addition, SP-D can form even higher-order multimers with a mass of several million kDa. The size of fully assembled collectins ranges from 13 nm for MBL to about 100 nm for SP-D. In order to recognize a variety of cell surface saccharides, collectins provide a broad monosaccharide specificity including mannose, galactose, and glucose.
Although the Kd of the binding of a single CRD with a monosaccharide ligand is in the order of 10−3 M., the Kd of binding of higher-order multimers of collectins to polyvalent ligands is in the order of 10−8 to 10−11 M. As a result, IgY and mammalian immunoglobulins can be bound tightly via their oligosaccharide chains by collectins. For example, MBL has been shown to bind agalactosylated glycoforms of IgG (Malhotra R., et al. Nat. Med. 1, 237-243, 1995) and to polymeric forms of serum IgA (Roos A., et al. J. Immunol. 167, 2861-2868, 2001). In chicken IgY, each H-chain contains two potential N-glycosylation sites located on CH2 (Asn308) and CH3 (Asn407) domains. The CH2 domain contains biantennary complex-type N-glycans (neutral, 29.9%; monosialyl, 29.3%; disialyl, 3.7%), whereas the CH3 domain contains high-mannose-type oligosaccharides (monoglucosy-lated, 26.8%; others, 10.5%) (Suzuki N, Lee Y C, Glycobiology 14, 275-292, 2004). The structural properties of complex-type N-glycans from chicken IgY resemble those of human IgG (Suzuki N, et al., J. Biol. Chem. 278, 46293-46306, 2003). Both human IgG and chicken IgY possess biantennary complex-type oligosaccharides with and without core alpha 1-6 Fuc and/or bisecting GlcNAc. Both antibodies have a mono-galactosylated branch predominantly on the GlcNAc beta 1-2 Man alpha 1-6 Man arm. Monosialylation in chicken IgY occurs on the Gal beta 1-4 GlcNAc beta 1-2 Man alpha 1-3 Man arm, which is also the case in normal human IgG (Takahashi N, et al., Anal. Biochem. 226, 139-146, 1995). Due to these similarities, N-glycans of IgY antibodies pose similar problems as those of human and other mammalian antibodies. In ELISA-type assays including those for the determination of free serum IgE, the presence of soluble C-type lectins in sera of patients can lead to false positive results as described for RF and HAMA (Boscato L M, Stuart M C, Clin Chem 34, 27-33, 1988). For example, monoclonal capture antibodies with specificity for human IgE produced by hybridoma technology or by expression in mammalian cells are glycoproteins which will bind soluble C-type lectins upon exposure to patient's sera. Since many C-type lectins such as MBL form oligomers of trimeric subunits, bound C-type lectins are likely to provide additional binding sites for glycosylated detection antibodies which could result in false positive data. Therefore, there is a need in the field for monoclonal antibodies or fragments thereof which do not interact with C-type lectins.
The present invention provides constructs comprising the extracellular portion of human FcεRIα and avian (IgY) constant immunoglobulin domains with one or more deleted glycosylation sites. Glycosylation of only one potential site on IgY constant domains reduces interference by C-type lectins in diagnostic and therapeutic applications, and is likely to help folding, assembling, and secreting such constructs in eukaryotic cells. One line of evidence for this assumption is the observation that glycosylation site Asn407 of IgY antibodies is well conserved in mammalian IgG (Asn297) and mammalian IgE (Asn394) although the number and position varies among the different species (Suzuki N, Lee Y C, Glycobiology 14, 275-292, 2004). Structural analyses of constant domains of mammalian IgE and IgG have demonstrated that N-glycans on Asn394 (mammalian IgE) and Asn297 (mammalian IgG) are buried in a cavity between the two heavy chains. Although the function of N-glycans in the cavity between the two heavy chains of mammalian IgE, IgG, and avian IgY remains to be elucidated, they probably play a role in folding, assembly, and stabilization of the immunoglobulin structures. Human IgE lacking N-glycosylation at Asn394 by point mutation has been reported to tend to self-aggregation (Basu M, et al., J. Biol. Chem. 268, 13118-13127, 1993), suggesting that N-glycans at this position are involved at least in stabilization of the protein structure by conferring suitable hydrophilicity to the cavity (Suzuki N, Lee Y C, Glycobiology 14, 275-292, 2004). Therefore, mono-glycosylation of IgY constant domains can be expected to help folding, assembling, and secreting such constructs in eukaryotic cells.
However, utilizing single chain variable fragments (scFv) as model for the construction of fusion proteins comprising avian constant domains, it has surprisingly been found that such constructs are correctly folded and secreted, even if both glycosylation sites are deleted. Therefore, in one embodiment the avian constant domains of the FcεRIα-fusion protein constructs are thus non-glycosylated. The absence of N-glycans guarantees a lack of interference by C-type lectins in IgE quantitation procedures. Compared to constructs comprising glycosylated IgY constant domains, constructs comprising non-glycosylated IgY constant domains provide a higher degree of reliability in diagnostic assay procedures.
Preferred constructs with non-glycosylated IgY constant domains include homodimeric constructs comprising the extracellular portion of human FcεRIα and avian constant immunoglobulin domains, in particular, IgY constant domains capable of forming dimeric molecules (
Other preferred fusion protein formats of this part of the invention include constructs comprising the extracellular portion of human FcεRIα and mono-glycosylated avian constant immunoglobulin domains, in particular, mono-glycosylated IgY constant domains (
Constructs Comprising Non-Glycosylated Derivatives of the Extracellular Fragment of Human FcεRIα and Non-Glycosylated IgY Constant Domains
There are seven N-linked carbohydrate attachment sites in the human FcεRIα molecule and in the truncated extracellular fragment expressed in CHO cells, all seven sites proved to be glycosylated (Letourner et al., J Biol Chem 270: 8249-8256, 1995). Glycosylation of FcεRIα is not required for IgE-binding (Blank et al., J Biol Chem 266: 2639-2646, 1991), but mutations of the carbohydrate attachment sites had an additive effect on the folding and secretion. Mutation of all seven N-glycosylation sites resulted in misfolding and retention of the fragment in the endoplasmic reticulum (Letourner et al., J Biol Chem 270: 8249-8256, 1995). Tunicamycin treatment reduced the folding efficiency in CHO cells substantially, but small quantities of non-glycosylated FcεRIα were expressed (Blank et al., J Biol Chem 266: 2639-2646, 1991). These data indicate that glycosylation facilitates correct folding in eukaryotic cells, but is not absolutely required for secretion.
The instant invention thus also relates to antibody constructs, wherein one or more glycosylation site(s) of the extracellular portion of human FcεRIα is(are) deleted. In addition, of course, one or more glycosylation site(s) of the immunoglobulin domain(s) may be deleted. According to a preferred embodiment, both the extracellular portion of human FcεRIα and the immunoglobulin domain(s) are non-glycosylated.
A functional, non-glycosylated extracellular fragment of FcεRIα can be expressed in reasonable yields in E. coli (Robertson, J Biol Chem 268: 12736-12743, 1993; Letourner et al., J Biol Chem 270: 8249-8256, 1995). In contrast to non-glycosylated FcεRIα fragments expressed in mammalian cells, those expressed in E. coli do not show a tendency to form oligomers and aggregates. Therefore, prokaryotic expression appears to be most appropriate for the production of non-glycosylated FcεRIα fragments and chimeric constructs of the invention comprising non-glycosylated FcεRIα fragments and non-glycosylated IgY constant domains.
Preferred fusion protein formats of this part of the invention comprise a non-glycosylated derivative of the extracellular portion of human FcεRIα and non-glycosylated avian constant immunoglobulin domains selected from the group consisting of CH1, CH2, CH3, and CH4 IgY domains. The construct can comprise one, two, three or four IgY constant domains.
FcεRIα constructs comprising only one or two IgY constant domains provide a lower molecular mass, require less disulfide bond formation and, therefore, are associated with increased expression rates. Instead of the conventional domains, fragments or variants thereof that do not inhibit folding of the constructs can be used. The boundaries between the domains, and thus the species origin of fragments at these boundaries can also be varied, as long as folding of the constructs is ensured.
Nucleic Acids and Expression System
The present invention also provides nucleic acid sequences and vectors encoding the above listed constructs of the invention. The invention thus relates to nucleic acid molecules encoding the antibody constructs of the invention, as well as to expression vectors comprising said nucleic acid molecules under control of appropriate promoters. The expression vectors include procaryotic as well as eukaryotic expression vectors.
According to a preferred embodiment, the invention relates to a nucleic acid encoding the chimeric fusion construct having the amino acid sequence represented as SEQ ID NO:35. Specifically the nucleic acid encoding said homodimeric FcεRIα construct comprising IgY constant domains CH1-CH4 has the nucleic acid sequence shown as SEQ ID NO:34. In a particular embodiment, vectors are included, which comprise the aforementioned specific nucleic acid sequences.
To produce the constructs, one having ordinary skill in the art can prepare a DNA molecule encoding the construct using well known techniques. The coding sequence can be obtained from natural sources or synthesized or otherwise constructed using widely available starting materials by routine methods. In some embodiments, a DNA molecule that includes a nucleotide sequence that encodes a construct of the invention may be synthesized using the amino acid sequence information herein and the genetic code. When the coding DNA is prepared synthetically, advantage can be taken of known codon preferences of the intended host where the DNA is to be expressed. One having ordinary skill in the art can insert that DNA molecule into a commercially available expression vector for use in well known expression systems, employing well known methods and readily available starting materials (see e.g., Sambrook et al., Molecular Cloning. A Laboratory Manual, second ed., Cold Spring Harbor Press, 1989).
The present invention thus also relates to nucleic acid sequences and expression vectors comprising this nucleic acid molecule under control of an appropriate promoter. The potential glycosylation sites can be eliminated by site-directed mutation using well known techniques.
In another embodiment, the present invention provides expression systems suitable for the production of the above listed constructs. Homodimeric constructs and those comprising glycosylated components or those requiring complex disulfide bond formation, have to be expressed in eukaryotic systems which provide a variety of processing mechanisms in the endoplasmatic reticulum (ER), including chaperone-assisted folding, glycosylation and oxidation. Several eukaryotic hosts including commonly used yeast, fungal cells, insect cells, mammalian cells, and cells of higher plants, are available for the production of the construct of the invention. The particulars for the construction of expression systems suitable for the desired host are known to those in the art. As shown in
The present invention provides also prokaryotic expression systems suitable for the production of non-glycosylated constructs. One advantage of prokaryotic systems is their ability to produce protein at low costs and in large quantities. Preferably, the expression is directed to the periplasmic space of Gram negative bacteria due to its oxidizing environment. The extracellular portion of human FcεRIα has been expressed as functional molecule in E. coli using the pel-B leader from the pectate lyase gene for directing secretion of the receptor fragment to the periplasmic space (Letourner et al., J Biol Chem 270: 8249-8256, 1995).
The same leader sequence has been utilized successfully for periplasmic expression of antibody fragments (for a review, see Verma et al., J Immunol Meth 216: 165-181, 1998). The particulars for the construction of expression systems suitable for the desired prokaryotic host are known to those in the art.
Analytical Applications
The construct of the invention can advantageously be used for in vitro diagnostic purposes, such as for the in vitro quantitation of human serum IgE, and in particular for the in vitro quantitation of non-complexed human serum IgE in the presence of IgE/anti-IgE complexes, wherein the monoclonal anti-IgE antibody specifically recognizes the residues in the Cε3 domain of IgE that are responsible for binding to FcεRI (e.g., murine monoclonal antibody MAE1 or humanized monoclonal antibody E25 designated omalizumab or Xolair®). Utilization of the construct of the invention for such applications minimizes cross-reactivity and false positive and negative results. Particular advantages are provided by use of those constructs that comprise a non-glycosylated fragment of human FcεRIα and non-glycosylated avian constant domains. Further applications are found as research tools or for biotechnological methods, e.g., in purification of IgE antibodies. The invention thus also provides a diagnostic agent or kit comprising the antibody construct disclosed herein.
In summary, the constructs of the invention offer great opportunities for diagnostic applications in atopic diseases.
The invention is described below by means of examples and figures, representing preferred embodiments of the invention, which however, shall not be considered to limit the invention.
A variety of constructs are suitable for the present invention. The structure of some of these formats described in the following examples are shown in
The human FcεRIα extracellular domain (FcεRIα-ecd) was synthesized from cDNA derived from human peripheral mononuclear cells. FcεRIα-ecd was amplified using one PCR primer containing a Pf123 II site (gatc cgtacg tgt ggg GCA GTC CCT CAG AAA CCT AAG G, SEQ ID NO:1) and another primer containing an Asc I site (gatc ggcgcgcc cggagcttttattacagtaatgttgag, SEQ ID NO:2). Utilizing Pwo-Polymerase (PeqLab, Germany) in combination with a mastercycler gradient (Eppendorf, Germany), following protocol was applied for PCR amplifications: 1×90 s 95° C., 30×(20 s 95° C., 1 min 60° C., 1 min 72° C.), 1×5 min 72° C.
All IgY immunoglobulin constant domains were synthesised from cDNA derived from chicken splenocytes. The 5′ end of the chicken υ1 heavy chain was amplified using the PCR primer (accgaagtcatcgtctcctc, SEQ ID NO:3) and (cctcagtttggcgtctaagc, SEQ ID NO:4). The 3′ end was amplified using the PCR primer (gaggatcacgtcaagggatg, SEQ ID NO:5) and (gcacccccaatcctttattt, SEQ ID NO:6). After hybridization of these two fragments the complete chicken υ1 heavy chain was amplified using the PCR primer (accgaagtcatcgtctcctc, SEQ ID NO:7) and (gcacccccaatcctttattt, SEQ ID NO:8). Amplification by PCR was performed as described in example 1.
2.1. Amplification υ1 CH1-4 domains. The chicken υ1 CH1-4 domains were amplified using one PCR primer containing an Asc I site (gatcggcgcgcccgcgagccccacatcgcc, SEQ ID NO:9) and another PCR primer containing a Xba I site and a 4×His overhang (gatctctagatcagtgatggtgatgtttaccagcctgtttctg, SEQ ID NO:10).
2.2. Amplification υ1 CH2-4 domains. The υ1 CH2-4 domains were amplified using one PCR Primer containing an Asc I site (gatcggcgcgccgcctgtagccccagag, SEQ ID NO:11) and another Primer containing a Xba I site and a 4×His overhang (gatctctagatc agtgatggtgatgtttaccagcctgtttctg, SEQ ID NO:12).
2.3. Amplification υ1 CH3-4 domains. The υ1 CH3-4 domains were amplified using one PCR primer containing an Asc I site (gatcggcgcgcccggcgctcagagctgc, SEQ ID NO:13) and another PCR primer containing a Xba I site and a 4×His overhang (gatctctagatcagtgatggtgatgtttaccagcctgtttctg, SEQ ID NO:14).
2.4. Amplification υ1 CH1-2 domains. The υ1 CH1-2 domains were amplified using one PCR primer containing an Asc I site (gatcggcgcgcccgcgagccccacatcgcc, SEQ ID NO:15) and another PCR primer containing a Xba I site and a 4×His overhang (gatctctagatcagtgatggtgatggaactccgggcatcccttgacgtgatc, SEQ ID NO:16).
2.5. Amplification υ1 CH2 domain. The υ1 CH2 domain was amplified using one PCR primer containing an Asc I site (gatcggcgcgccgcctgtagccccagag, SEQ ID NO:17) and another PCR primer containing a Xba I site and a 4×His overhang (gatctctagatcagtgatggtgatggaactccgggcatcccttgacgtgatc, SEQ ID NO:18).
The signal sequence of a rodent κ light chain was assembled by 2 PCR primers containing a Nhe I site (gtacaagcttgctagcaagatggaatcacagacccaggtcctcatgtccctgctgctc, SEQ ID NO:19) and another primer (atgtccctgctgctctggatttctggtacctgtggggtccctcagaaacctaag, SEQ ID NO:20). Amplification by PCR was performed as described in example 1.
For convenient expression of the constructs of the invention a set of modular cassettes containing IgY constant immunoglobulin domains, and restriction sites for the incorporation of FcεRIα-ecd was generated.
The individual IgY constant immunoglobulin domains immunoglobulin were inserted via Asc I and Xba I into the mammalian expression vector pcDNA3.1-zeo (Invitrogen life technologies, Karlsruhe, Germany) containing the rodent K light chain leader sequence. Introduction of FcεRIα-ecd into the vector was performed by introduction of a BsiW I site at the N-terminus (gatccgtacgtgtggggccgtgacgttggacg, SEQ ID NO:21) and an Asc I site at the C-terminus (gatcggcgcgccacctaggacggtcaggg, SEQ ID NO:22) of the receptor fragment by PCR. Subsequently, the DNA was ligated into the vector pcDNA3.1-zeo (Invitrogen life technologies, Karlsruhe, Germany) containing the signal sequence and the particular constant regions.
The site-directed mutagenesis was performed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, USA). For substitution of the glycosylation site in position 407 the expression vector pcDNA3.1-zeo (Invitrogen life technologies, Karlsruhe, Germany) modified as described in example 4 was amplified using the PCR primer (ggtcctccaagaacactt ccagggcacctacagcgccagc, SEQ ID NO:23) and the PCR primer (gctggcgctgtaggtgccctggaagtgtccttggaggacc, SEQ ID NO:24). For substitution of the glycosylation site in position 308 the expression vector pcDNA3.1-zeo (Invitrogen life technologies, Karlsruhe, Germany) modified as described in example 4 was amplified using the PCR primer (gcctgagcagccgcgtcca ggtcagcggcaccgattgg, SEQ ID NO:25) and the PCR primer (ccaatcggtgccgctgacctggacgcggctgctcaggc, SEQ ID NO:26). Correctness of the mutations was verified by DNA sequencing.
For prokaryotic expression the human FcεRIα extracellular domain was amplified using one PCR primer containing an Nde I site (gatccatatggcagtccctcagaaacctaag, SEQ ID NO:27) and another primer containing a Not I site (gatcgcggccgccggagcttttattacagtaatgttgag, SEQ ID NO:28), and introduced into the procayotic expression vector pET26b+ (Novagen, Schwalbach, Germany).
The plasmids encoding the FcεRIα-ecd was transformed in the E. coli K12 strain BL21 Ril DE3. Small-scale expressions were performed at 30° C. using 100 ml 2YT medium (16 g/liter tryptone, 10 g/liter yeast extract, 5 g/liter NaCl) containing 50 μg/ml kanamycin. Cultures were inoculated from a 10 ml preculture to OD550=0.1. Expression was induced with 1 mM IPTG at an OD550 between 1.0 and 1.5. Cells were harvested 3 h after induction by centrifugation and resuspended in PBS buffer (20 mM NaH2PO4, 500 mM NaCl, 0.1 mM EDTA, pH 7.2) using 1 ml buffer per mg cells. Whole cell extracts were prepared by sonification and crude extract was centrifuged in an Eppendorf tube for 60 min at 20.000×g and 4° C. The supernatants containing the soluble material were passed through a 22 μm filter (Millipore, Germany).
HEK-293 cells (ATCC number CRL-1573) were cultivated in DMEM (Dulbecco's modified Eagle Medium) supplemented with 10% (v/v) heat-inactivated fetal calf serum, 100 IU/ml penicillin, and 100 μg/ml streptomycin. Tissue culture reagents were obtained from Invitrogen life technologies (Karlsruhe, Germany). HEK-293 cells growing in DMEM supplemented with 10% (v/v) fetal calf serum were transfected with 2 μg of the particular expression vector using PEI (polyethylenimine, Sigma, Taufkirchen, Germany). Stable transfectants then were selected in DMEM supplemented with 10% (v/v) fetal calf serum and 100 μg/ml of zeocin (Invitrogen life technologies, Karlsruhe, Germany).
For expression of the constructs, transfected cells were grown for 3 days as an adhesion culture. The constructs secreted by transfected HEK-293 cells were purified from the culture medium by affinity chromatography using Ni-NTA-agarose (Qiagen, Hilden, Germany) according to the manufacturers recommendations.
As an example,
FcεRIαecd and IgY Constant Domains
Purified recombinant human IgE was diluted in phosphate-buffered saline-2% (w/v) milk powder (MPBS; Sigma, Taufen-stein, Germany) containing 10% (w/v) fetal calf serum (FKS, Biochrom, Germany), applied to microtiter plates at 4° C. overnight, and blocked with 10% (w/v) FKS at room temperature for 2 h. After incubation of the microtiter plates for 90 min at room temperature on a rocker platform, the wells were rinsed 3 times each with Tween (0.1% v/v)-phosphate-buffered saline (TPBS) and PBS and further incubated with 100 μl each of cell culture supernatant FcεRIα-IgYCH1-4His (construct I) (diluted in 10% (w/v) FKS) for 60 min at room temperature on a rocker platform. The wells were rinsed again 3 times each with 0.1% (v/v) TPBS and PBS and further incubated with 100 μl each of polyclonal anti-chicken-IgG-AP conjugate (diluted 1:1000 in 10% FKS, Sigma, Taufenstein, Germany) for 60 min at room temperature on a rocker platform. The wells were rinsed again 3 times each with 0.1% (v/v) TPBS and PBS and bound antibodies were visualized by the addition of 75 μl of a 4-nitrophenyl disodium orthophosphate substrate solution (Sigma, Taufenstein, Germany). Absorbance was determined at 405 nm after 20 min of incubation. Data of the experiment are shown in
The assay principle is illustrated in
Purified FcεRIα-IgYCH1-4His (construct I) (diluted with phosphate buffered saline-milk powder (2% (w/v), MPBS) was applied to microtiter plates at 4° C. overnight and blocked with MPBS at room temperature for 2 h. After incubation of the microtiter plates for 90 min at room temperature on a rocker platform, the wells were rinsed 3 times each with) Tween (0.1% v/v)-PBS (TPBS) and PBS and further incubated with 100 μl each of recombinant human IgE (diluted 100 nM to 0.1 nM in 2% (w/v) MPBS) for 60 min at room temperature on a rocker platform. The wells were rinsed again 3 times each with 0.1% (v/v) TPBS and PBS and further incubated with 100 μl each of polyclonal anti-human IgE-HRP conjugate (diluted 1:1000 in 2% (w/v) MPBS, Sigma, Taufenstein, Germany) for 60 min at room temperature on a rocker platform. The wells were rinsed again 3 times each with 0.1% (v/v) TPBS and PBS and bound antibodies were visualized by the addition of 75 μl of a 4-nitrophenyl disodium orthophosphate substrate solution (Sigma, Taufenstein, Germany). Absorbance was determined at 405 nm after 20 min of incubation. Data of the experiment are shown in
The assay principle is illustrated in
To prepare IgE standards, a human serum containing a very low total IgE concentration (<1 U/ml) was spiked with predefined amounts of purified human myeloma IgE (range: 0-4,000 μg/ml). As an alternative, the same standard range of IgE was prepared in assay diluent buffer, which is part of a set of ready-to-use ELISA reagents (Pharmingen, 550534).
ELISA plates (Greiner) were coated overnight at 4° C. with 100 μl/well of FcεRIα-IgYCH1-4His (construct I) containing coating buffer solution (part of Pharmingen kit 550534) at coating concentrations of 1 or 10 μg/ml. The coating solution was removed and the plates were washed three times with 300 μl washing buffer (part of Pharmingen kit). Subsequently, unoccupied sites were blocked by incubating for 1 h at room temperature with 200 μl/well of assay diluent buffer. After washing three times with 300 μl/well of washing buffer, IgE standard range solutions and controls (diluted 1:10 in assay diluent buffer) were added to the appropriate wells (100 μl/well) in duplicate and incubated for 2 h at room temperature. The plates were washed again five times with 300 μl/well of washing buffer. Subsequently, 100 μl of a detector mix solution were added per well, containing a biotinylated murine anti-human IgE antibody (BD, 555858) and an avidin-horseradish peroxidase (HRP) complex (BD), both diluted 1:250 in assay diluent buffer, and incubated for 1 h at room temperature. Plates were washed seven times (300 μl/well), followed by the addition of 100 μl/well of substrate solution (part of the Pharmingen kit) and incubation for 30 min at room temperature in the dark. The substrate reaction was terminated by the addition of 50 μl/well of stop solution (Pharmingen kit). The optical density was read at 450 nm using a Dynex MRX II ELISA reader. The working range of the assay was between 31 and 4,000 μg/ml. Data of the experiment are shown in
The assay principle is illustrated in
Human myeloma IgE standard solutions (range: 0-4,000 μg/ml) were prepared asw described in example 10. ELISA plates (Greiner) were coated overnight at 4° C. with 100 μl/well of omalizumab (Xolair®) containing coating buffer solution (part of Pharmingen kit 550534) at coating concentrations of 1 or 10 μg/ml. The coating solution was removed and the plates were washed three times with 300 μl washing buffer (part of Pharmingen kit). Subsequently, unoccupied sites were blocked by incubating for 1 h at room temperature with 200 μl/well of assay diluent buffer. After washing three times with 300 μl/well of washing buffer, IgE standard range solutions and controls (diluted 1:10 in assay diluent buffer) were added to the appropriate wells (100 μl/well) in duplicate and incubated for 2 h at room temperature. The plates were washed again five times with 300 μl/well of washing buffer. Subsequently, 100 μl of a detector mix solution were added per well, containing a biotinylated murine anti-human IgE antibody (BD, 555858) and an avidin-horseradish peroxidase (HRP) complex (BD), both diluted 1:250 in assay diluent buffer, and incubated for 1 h at room temperature. Plates were washed seven times (300 μl/well), followed by the addition of 100 μl/well of substrate solution (part of the Pharmingen kit) and incubation for 30 min at room temperature in the dark. The substrate reaction was terminated by the addition of 50 μl/well of stop solution (Pharmingen kit). The optical density was read at 450 nm using a Dynex MRX II ELISA reader. The working range of the assay was between 31 and 4,000 μg/ml. Data of the experiment are shown in
The assay principle is illustrated in
For the preparation of complexed IgE, human myeloma IgE was added to IgE-negative normal human serum (total serum IgE concentration <1 U/ml) at a final concentration of 10 ng/ml.
Anti-IgE antibody omalizumab (Xolair®) was added to these IgE-containing sera at a final concentration of 0.002 to 200 μg/ml. Thus, concentration-dependent amounts of IgE-containing complexes and free IgE were generated.
Free IgE was measured using an ELISA technique with FcεRIα-IgYCH1-4His (construct I) as capture reagent. This avoids detection of IgE that is already complexed with omalizumab and thus measures both baseline IgE levels before anti-IgE treatment and remaining free (uncomplexed) IgE after anti-IgE treatment.
A biotinylated monoclonal murine anti-IgE which was specific for an epitope of the human IgE molecule different from that of omalizumab and FcεRIα was used as the secondary (revealing) antibody.
ELISA plates (Greiner) were coated overnight at 4° C. with 100 μl/well of FcεRIα-IgYCH1-4His (construct I) containing coating buffer solution (part of Pharmingen kit 550534) at coating concentrations of 1 or 10 μg/ml. The coating solution was removed and the plates were washed three times with 300 μl washing buffer (part of Pharmingen kit). Subsequently, unoccupied sites were blocked by incubating for 1 h at room temperature with 200 μl/well of assay diluent buffer. After washing three times with 300 μl/well of washing buffer, sera containing free IgE and IgE/anti-IgE-complexes and controls (diluted 1:10 in assay diluent buffer) were added to the appropriate wells (100 μl/well) in duplicate and incubated for 2 h at room temperature. The plates were washed again five times with 300 μl/well of washing buffer. Subsequently, 100 μl of a detector mix solution were added per well, containing the biotinylated murine anti-human IgE antibody (BD, 555858) and an avidin-horseradish peroxidase (HRP) complex (BD), both diluted 1:250 in assay diluent buffer, and incubated for 1 h at room temperature. Plates were washed seven times (300 μl/well), followed by the addition of 100 μl/well of substrate solution (part of the Pharmingen kit) and incubation for 30 min at room temperature in the dark. The substrate reaction was terminated by the addition of 50 μl/well of stop solution (Pharmingen kit). The optical density was read at 450 nm using a Dynex MRX II ELISA reader. The working range of the assay was between 31 and 4,000 μg/ml.
The assay principle is illustrated in
For the preparation of complexed IgE, human myeloma IgE was added to IgE-negative normal human serum (total serum IgE concentration <1 U/ml) at a final concentration of 10 ng/ml. FcεRIα-IgYCH1-4His (construct I) was added to these IgE-containing sera at a final concentration of 0.002 to 200 μg/ml. Thus, concentration-dependent amounts of IgE-containing complexes and free IgE were generated.
Free IgE was measured using an ELISA technique with omalizumab (Xolair®) as capture antibody. A biotinylated monoclonal murine anti-IgE which was specific for an epitope of the human IgE molecule different from that of omalizumab and FcεRIα was used as the secondary (revealing) antibody.
ELISA plates (Greiner) were coated overnight at 4° C. with 100 μl/well of omalizumab (Xolair®) containing coating buffer solution (part of Pharmingen kit 550534) at coating concentrations of 1 or 10 μg/ml. The coating solution was removed and the plates were washed three times with 300 μl washing buffer (part of Pharmingen kit). Subsequently, unoccupied sites were blocked by incubating for 1 h at room temperature with 200 μl/well of assay diluent buffer. After washing three times with 300 μl/well of washing buffer, sera containing free IgE and IgE/anti-IgE-complexes and controls (diluted 1:10 in assay diluent buffer) were added to the appropriate wells (100 μl/well) in duplicate and incubated for 2 h at room temperature. The plates were washed again five times with 300 μl/well of washing buffer. Subsequently, 100 μl of a detector mix solution were added per well, containing the biotinylated murine anti-human IgE antibody (BD, 555858) and an avidin-horseradish peroxidase (HRP) complex (BD), both diluted 1:250 in assay diluent buffer, and incubated for 1 h at room temperature. Plates were washed seven times (300 μl/well), followed by the addition of 100 μl/well of substrate solution (part of the Pharmingen kit) and incubation for 30 min at room temperature in the dark. The substrate reaction was terminated by the addition of 50 μl/well of stop solution (Pharmingen kit). The optical density was read at 450 nm using a Dynex MRX II ELISA reader. The working range of the assay was between 31 and 4,000 μg/ml. Data of the experiment are shown in
The assay principle is illustrated in
For the preparation of complexed IgE, human myeloma IgE was added to IgE-negative normal human serum (total serum IgE concentration <1 U/ml) at a final concentration of 10 ng/ml. Anti-IgE antibody omalizumab (Xolair®) was added to these IgE-containing sera at a final concentration of 0.002 to 200 μg/ml. Thus, concentration-dependent amounts of IgE-containing complexes and free IgE were generated.
Free IgE was measured using an ELISA technique with omalizumab (Xolair®) as capture antibody. A biotinylated monoclonal murine anti-IgE which was specific for an epitope of the human IgE molecule different from that of omalizumab and FcεRIα was used as the secondary (revealing) antibody.
ELISA plates (Greiner) were coated overnight at 4° C. with 100 μl/well of omalizumab (Xolair®) containing coating buffer solution (part of Pharmingen kit 550534) at coating concentrations of 1 or 10 μg/ml. The coating solution was removed and the plates were washed three times with 300 μl washing buffer (part of Pharmingen kit). Subsequently, unoccupied sites were blocked by incubating for 1 h at room temperature with 200 μl/well of assay diluent buffer. After washing three times with 300 μl/well of washing buffer, sera containing free IgE and IgE/anti-IgE-complexes and controls (diluted 1:10 in assay diluent buffer) were added to the appropriate wells (100 μl/well) in duplicate and incubated for 2 h at room temperature. The plates were washed again five times with 300 μl/well of washing buffer. Subsequently, 100 μl of a detector mix solution were added per well, containing the biotinylated murine anti-human IgE antibody (BD, 555858) and an avidin-horseradish peroxidase (HRP) complex (BD), both diluted 1:250 in assay diluent buffer, and incubated for 1 h at room temperature. Plates were washed seven times (300 μl/well), followed by the addition of 100 μl/well of substrate solution (part of the Pharmingen kit) and incubation for 30 min at room temperature in the dark. The substrate reaction was terminated by the addition of 50 μl/well of stop solution (Pharmingen kit). The optical density was read at 450 nm using a Dynex MRX II ELISA reader. The working range of the assay was between 31 and 4,000 μg/ml. Data of the experiment are shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 05022582 | Oct 2005 | EP | regional |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5945294 | Frank et al. | Aug 1999 | A |
| 6608172 | Chiou | Aug 2003 | B1 |
| 6737056 | Presta | May 2004 | B1 |
| 20050221424 | Utku | Oct 2005 | A1 |
| Number | Date | Country |
|---|---|---|
| WO 9823964 | Jun 1998 | WO |
