HK1036286B - Enhancing the circulating half-life of antibody-based fusion proteins - Google Patents

Description

Method for increasing circulating half-life of antibody-based fusion proteins

Cross reference to related applications

U.S. provisional patent application serial No. 60/075,887 filed on 25/2/1998 and claiming priority and benefits thereof is hereby incorporated by reference.

Technical Field

The present invention relates generally to fusion proteins. More specifically, the invention relates to methods for increasing the circulating half-life of antibody-based fusion proteins.

Background

The use of antibodies for the treatment of human diseases is well established and has been closely associated with the introduction of genetic engineering. Several techniques have been developed to increase the utility of antibodies. They include: (1) creating "hybridomas" by cell fusion or generating monoclonal antibodies by molecular cloning of antibody heavy (H) and light (L) chains from antibody-producing cells; (2) conjugation of other molecules such as radioisotopes, toxic drugs, protein toxins and cytokines to antibodies for their transport to preferred in vivo sites; (3) manipulating antibody effector functions so as to increase or decrease biological activity; (4) combining other proteins such as toxins and cytokines with antibodies at the genetic level to produce antibody-based fusion proteins; and (5) combining one or more sets of antibody binding regions at the genetic level to produce a bispecific antibody.

When proteins are bound to each other by chemical or genetic manipulation, it is often difficult to predict what properties the end product retains from the parent molecule. With chemical conjugation, the binding process can occur at different sites on the molecule and generally results in molecules with varying degrees of alteration that can affect the function of one or both proteins. On the other hand, the use of genetic fusions allows a more consistent ligation process and results in the production of a stable end product that retains the function of both component proteins. See, e.g., gilles et al, proceedings of the american national academy of sciences (proc.natl.acad.sci.usa) 89: 1428-1432(1992) and U.S. Pat. No. 5,650,150.

However, the use of recombinantly produced antibody-based fusion proteins can be limited by their rapid clearance from the circulation in vivo. For example, antibody-cytokine fusion proteins have been shown to have significantly lower circulating half-lives in vivo than free antibodies. When testing various antibody-cytokine fusion proteins, Gillies et al reported that all tested fusion proteins had an alpha phase (distribution phase) half-life of less than 1.5 hours. In fact, by 2 hours, most antibody-based fusion proteins were cleared to a free antibody serum concentration of 10%. See Gillies et al, bioligand chemistry (biocoj. chem.) 4: 230-235(1993). Accordingly, there is a need in the art for methods of increasing the circulating half-life of antibody-based fusion proteins in vivo.

Summary of the invention

A novel method of increasing the circulating half-life of antibody-based fusion proteins in vivo has now been discovered. In particular, the present invention provides a method for producing a fusion protein between an immunoglobulin with reduced binding affinity for an Fc receptor and a second non-immunoglobulin protein. Antibody-based fusion proteins with reduced binding affinity for Fc receptors have significantly longer circulating half-lives in vivo than the unlinked second non-immunoglobulin protein.

IgG molecules interact with three classes of Fc receptors (fcrs) specific for IgG class antibodies, namely Fc γ RI, Fc γ RII and Fc γ RIII. In a preferred embodiment, the immunoglobulin (Ig) component of the fusion protein has a constant region of at least a portion of IgG that has reduced binding affinity for at least one of Fc γ RI, Fc γ RII, and Fc γ RIII.

In one aspect of the invention, the binding affinity of the fusion protein to the Fc receptor is reduced by using the heavy chain isotype as a fusion partner with reduced binding affinity to the Fc receptor on the cell. For example, human IgG1 and IgG3 were reported to bind Fc γ RI with high affinity, whereas IgG4 bound well to it by a factor of 10 or less, and IgG2 did not bind to it at all. An important sequence for IgG binding to Fc receptors is reported to be located in the CH2 domain. Thus, in a preferred embodiment, an antibody-based fusion protein with an increased circulating half-life in vivo is obtained by linking at least the CH2 domain of IgG2 or IgG4 to a second non-immunoglobulin protein.

In another aspect of the invention, the constant region that reduces the binding affinity of the fusion protein to the Fc receptor is reduced by introducing a genetic modification of one or more amino acids into the constant region of the IgG1 or IgG3 heavy chainBinding affinity of these isoforms to Fc receptors. Such modifications include changes to residues necessary for contact with the Fc receptor or other factors that affect contact between other heavy chain residues and the Fc receptor by induced conformational changes. Thus, in a preferred embodiment, an antibody-based fusion protein with an increased circulating half-life in vivo is obtained by the following steps: first introducing a mutation, deletion or insertion into one or more amino acids selected from Leu in the IgG1 constant region₂₃₄、Leu₂₃₅、Gly₂₃₆、Gly₂₃₇、Asn₂₉₇And Pro₃₃₁(ii) a And then linking the resulting immunoglobulin or portion thereof to a second non-immunoglobulin protein. In another preferred embodiment, the IgG3 constant region is mutated, deleted or inserted at one or more amino acids selected from the group consisting of Leu₂₈₁、Leu₂₈₂、Gly₂₈₃、Gly₂₈₄、Asn₃₄₄And Pro₃₇₈(ii) a And linking the resulting immunoglobulin or portion thereof to a second non-immunoglobulin protein. The resulting antibody-based fusion protein has a longer circulating half-life in vivo than the unlinked second non-immunoglobulin protein.

In a preferred embodiment, the second non-immunoglobulin component of the fusion protein is a cytokine. The term "cytokine" as used herein is used to describe proteins, analogs and fragments thereof that are produced and secreted by a cell and which elicit a specific response in the cell that has a receptor for the cytokine. Preferably, the cytokines include: interleukins such as interleukin-2 (IL-2); hematopoietic factors such as granulocyte-macrophage colony stimulating factor (GM-CSF); tumor Necrosis Factors (TNF) such as TNF α and lymphokines such as lymphotoxin. Preferably, the antibody-based fusion protein of the present invention exhibits the biological activity of a cytokine.

In another preferred embodiment, the second non-immunoglobulin component of the fusion protein is a biologically active ligand binding protein. For example, such ligand binding proteins may: (1) blocking receptor-ligand interactions at the cell surface; or (2) counteract the biological activity of a molecule (e.g., a cytokine) in the liquid phase of the blood, thereby preventing it from reaching its cellular target. Preferably, the ligand binding protein comprises CD4, CTLA-4, TNF receptor or interleukin receptors such as the IL-1 and IL-4 receptors. Preferably, the antibody-based fusion proteins of the present invention exhibit the biological activity of a ligand binding protein.

In another alternative preferred embodiment, the non-immunoglobulin component of the second said fusion protein is a protein toxin. Preferably, the antibody-toxin fusion proteins of the present invention exhibit toxic activity of the protein toxin.

In a preferred embodiment, the antibody-based fusion protein comprises a variable region specific for a target antigen and a constant region linked by a peptide that binds to a second non-immunoglobulin protein. The constant region may be a constant region normally associated with a variable region or a different region, e.g., from different kinds of variable and constant regions. The heavy chain may comprise CH1, CH2 and/or CH3 domains. The term "fusion protein" also includes within its meaning constructs having binding domains comprising framework and variable regions (i.e., complementarity determining regions) from different species such as those disclosed in GB2,188,638 by Winter et al. Antibody-based fusion proteins that include variable regions better exhibit antigen binding specificity. In another preferred embodiment, the antibody-based fusion protein further comprises a light chain. The invention thus provides fusion proteins that combine antigen binding specificity and antibody activity with the potent biological activity of a second non-immunoglobulin protein, such as a cytokine. The fusion proteins of the invention can be used to selectively transport a second non-immunoglobulin protein to a target cell in vivo so that the second non-immunoglobulin protein can exert a localized biological effect.

In an alternative preferred embodiment, the antibody-based fusion protein comprises a heavy chain constant region linked by a peptide that binds to a second non-immunoglobulin protein and does not comprise a heavy chain variable region. The invention thus further provides fusion proteins that retain the potent biological activity of a second non-immunoglobulin protein but lack antigen-binding specificity and antibody activity.

In a preferred embodiment, the antibody-based fusion proteins of the present invention comprise sequences necessary to bind an Fc protective receptor (FcRp), such as the neonatal intestinal transit receptor (FcRn) comprising beta-2 microglobulin.

In a preferred embodiment, the two fusion proteins comprising a chimeric chain comprising at least part of a heavy chain and a second non-Ig protein are linked by a disulfide bond.

The invention also features DNA constructs encoding the above fusion proteins and cell lines transfected with these constructs, such as myeloma.

These and other objects, advantages and features of the invention disclosed herein will become more fully apparent from the following description, the accompanying drawings and the claims.

Brief description of the drawings

The above and other objects, features and advantages of this invention, as well as the invention itself, will be more fully understood from the following description of the preferred embodiments, when read in conjunction with the accompanying drawings, wherein:

FIG. 1 is a drawing of a homologous alignment of the amino acid sequences of the C.gamma.1 and C.gamma.3 constant regions, the alignment being performed to maximize amino acid identity, and wherein the non-conserved amino acids are boxed;

FIG. 2 is a homology alignment of the amino acid sequences of the C.gamma.1, C.gamma.2 and C.gamma.4 constant regions, aligned to maximize amino acid identity, and in which non-conserved amino acids are boxed;

FIG. 3 is a schematic representation of a map of genetic constructs encoding antibody-based fusion proteins showing relevant restriction sites;

FIG. 4 is a bar graph depicting binding of antibody hu-KS-1/4 and antibody-based fusion proteins hu-KS γ 1-IL-2 and hu-KS γ 4-IL-2 to Fc receptors on mouse J774 cells in the presence (solid bars) or absence (dotted bars) of excess mouse IgG;

FIG. 5 is a linear graph depicting plasma concentrations of total antibody (free antibody and fusion protein) as a function of time for hu-KS γ 1-IL-2 (closed diamonds) and hu-KS γ 4-IL-2 (closed triangles) and hu-KS γ 1-IL-2 (open diamonds) intact fusion proteins in vivo;

FIG. 6 is a diagrammatic representation of a protocol for constructing antibody-based fusion proteins with mutations that reduce binding affinity to Fc receptors;

FIG. 7 is a diagram depicting in vivo hu-KS γ 1-IL-2 (. diamond.), mutated hu-KS γ 1-IL-2And hu-KS γ 4-IL-2 (. DELTA.) as a function of time.

Detailed Description

It has now been found that fusing a second protein, such as a cytokine, to an immunoglobulin can alter the antibody structure, resulting in increased binding affinity to one or more cell-bound Fc receptors and resulting in rapid clearance of the antibody-based fusion protein from the circulation. The present invention describes antibody-based fusion proteins with increased circulating half-life in vivo and includes methods for producing antibody-based fusion proteins with reduced binding affinity for one or more Fc receptors by recombinant DNA techniques.

First, an antibody-based fusion protein having an increased circulating half-life in vivo can be obtained by: fusion proteins are constructed using isoforms having reduced binding affinity for Fc receptors and avoid the use of sequences from antibody isoforms that bind Fc receptors. For example, of the 4 well-known IgG isotypes, IgG1(C γ 1) and IgG3(C γ 3) are known to bind FcR γ I with high affinity, whereas IgG4(C γ 4) binds with 10-fold lower affinity and IgG2(C γ 2) does not bind FcR γ I. Thus, antibody-based fusion proteins with reduced binding affinity for Fc receptors can be obtained by constructing fusion proteins with C γ 2 constant regions (Fc regions) or C γ 4 Fc regions and avoiding the use of constructs with C γ 1 Fc regions or C γ 3 Fc regions.

Second, antibody-based fusion proteins with increased circulating half-life in vivo can be obtained by modifying the sequences necessary for binding to Fc receptors in isoforms with binding affinity for Fc receptors, thereby reducing or eliminating binding. As mentioned above, IgG molecules interact with three classes of Fc receptors (FcR), namely Fc γ RI, Fc γ RII and Fc γ RIII. C γ 1 and C γ 3 bind Fc γ RI with high affinity, while C γ 4 and C γ 2 have reduced or no binding affinity for Fc γ RI. Comparison of C γ 1 and C γ 3 shows that: the amino acid sequence homology between these two isoforms is high except for the extended hinge segment in C γ 3. This is true even in those regions that have been shown to interact with complement Clq fragments and various Fc γ R classes. FIG. 1 provides a sequence alignment of the amino acid sequences of C.gamma.1 and C.gamma.3. The other two isotypes of human IgG (C γ 2 and C γ 4) have sequence differences associated with Fc γ R binding. FIG. 2 provides a sequence alignment of the amino acid sequences of C.gamma.1, C.gamma.2 and C.gamma.4. An important sequence for Fc γ R binding is Leu-Gly (residues 234 to 237 in cy 1) located in the CH2 domain adjacent to the hinge. Canfield and Morrison journal of experimental drugs (j.exp.med.) 173: 1483-1491(1991). Motifs of these sequences are conserved in C γ 1 and C γ 3, consistent with their similar biological properties; and may be related to the similarity of pharmacokinetic properties when they are used to construct IL-2 fusion proteins. Has already been used forA number of mutational analyses were performed to demonstrate that specific mutations were binding to FcR, including residues 234-237 and the hinged proximal bent residue Pro substituted by Ser in IgG4₃₃₁The effects of those in (1). Another important structural element necessary for efficient FcR binding is the presence of a covalently bonded Asn₂₉₇Bound N-linked carbohydrate chains. Such a structure or mutation that enzymatically removes Asn residues may effectively eliminate or at least significantly reduce all types of binding to Fc γ R.

Brumbell et al postulate that there is a protective receptor (FcRp) that slows the rate of catabolism of circulating antibodies by binding to the Fc portion of the antibody and can redirect them back into circulation after their pinocytosis into the cell. Brumbell et al Nature (NATURE) 203: 1352-135(1964). The beta-2 microglobulin-containing neonatal intestinal transit receptor (FcRn) has recently been identified as FcRp. See Junghans et al, "proceedings of the national academy of sciences of the united states (proc.natl.acad.sci.usa) 93: 5512-5516(1996). The sequences necessary for binding to this receptor are conserved in all four classes of human IgG and they are located at the interface between the CH2 and CH3 domains. See Medesan et al journal of immunology (J.IMMUBOL.) 158: 2211-2217(1997). These sequences have been reported to be important for the circulating half-life of the antibody in vivo. See international PCT application WO 97/34631. Thus, preferred antibody-based fusion proteins of the invention have sequences necessary for binding FcRp.

Methods of embodiments of the present synthesis and assays for detecting pharmacokinetic activity thereof are described, including the use of in vitro and preclinical in vivo animal models. Preferred genetic constructs encoding the chimeric chains comprise a DNA segment encoding at least a portion of an immunoglobulin in the 5 'to 3' direction and a DNA encoding a second non-immunoglobulin protein. Another preferred genetic construct comprises a DNA segment encoding a second non-immunoglobulin protein in the 5 '-3' direction and a DNA encoding at least a portion of an immunoglobulin protein. The fusion gene is assembled and inserted into an expression vector for transfection of a suitable recipient cell, where it is expressed.

The invention is further illustrated by the following non-limiting examples:

example 1 increasing antibodies by class switching from C γ 1 to C γ 4 IgG constant regions-

Circulating half-life of IL2 fusion protein in vivo

According to the present invention, an antibody-based fusion protein having an improved in vivo circulating half-life can be obtained by constructing an antibody-based fusion protein using sequences from antibody isotypes that have reduced or no binding affinity to an Fc receptor.

To assess whether the in vivo circulating half-life of antibody-based fusion proteins could be increased by using sequences from antibody isotypes that have reduced or no binding affinity for Fc receptors, an antibody-IL 2 fusion protein with a human C γ 1 constant region (Fc region) was compared to an antibody-IL 2 fusion protein with a human C γ 4 Fc region.

1.1 construction of antibody-IL 2 fusion proteins with C.gamma.4 IgG constant regions

The construction of antibody-IL 2 fusion proteins with C γ 1 constant regions has been described in the prior art. See, e.g., gilles et al, "proceedings of the national academy of sciences of the united states (proc.natl.acad.sci.usa) 89: 1428-1432(1992) and U.S. Pat. No. 5,650,150, which are incorporated herein by reference.

To construct an antibody-IL 2 fusion protein with a C γ 4 constant region, a plasmid vector capable of expressing a humanized antibody-IL 2 fusion protein with a variable (V) region specific for human whole cancer antigen (KSA) and a human C γ 1 heavy chain fused to human IL-2 was modified by removing the C γ 1 gene fragment and replacing it with the corresponding sequence from the human C γ 4 gene. A map of certain relevant restriction sites and insertion sites of C γ 4 gene fragments is provided in fig. 3. These plasmid constructs contain the Cytomegalovirus (CMV) early promoter for transcription of mRNA encoding the light (L) and heavy (H) chain variable (V) regions derived from the mouse antibody KS-1/4. Mouse V regions are humanized by standard methods and the DNA sequences encoding them are synthesized chemically. A functional splice donor site was added to the end of each V region so that it could be used in vectors containing H and L chain constant region genes. The human C γ light chain gene is inserted downstream of the cloning site for the VL gene and is followed by its endogenous 3' untranslated region and polyadenylation site. This transcription unit is followed by a second independent transcription unit for the heavy chain-IL 2 fusion protein. It is also initiated by the CMV promoter. The VH coding sequence is inserted upstream of the DNA encoding the particular C γ heavy chain gene fused to the human IL-2 coding sequence. Such C γ genes contain a splice acceptor site for the first heavy chain exon (CH1), just downstream from Hind III, which is the only one common to all human C γ genes. The 3' untranslated and polyadenylation site from the SV40 virus was inserted at the end of the IL-2 coding sequence. The remainder of the vector contains the bacterial plasmid DNA necessary for propagation in E.coli and a selectable marker gene (dihydrofolate reductase-dhfr) for selection of mammalian cell transfectants.

The interchange of C.gamma.1 and C.gamma.4 fragments was accomplished by digesting the plasmid DNA containing the original C.gamma.1 with Hind III and Xho I and purifying the large 7.8kb fragment by agarose gel electrophoresis. The second plasmid DNA containing the C.gamma.4 gene was digested with HindIII and Nsi I and the 1.75kb fragment was purified. A third plasmid containing the human IL-2 cDNA and SV40 poly A site fused to the carboxy terminus of the human C γ 1 gene was digested with Xho I and Nsi I and a small 470bp fragment was purified. All three fragments were ligated to each other in approximately equimolar amounts and the ligation products were used to transform competent E.coli. This ligation product was used to transform competent E.coli and clones were selected by growth on ampicillin-containing plates. Correctly assembled recombinant plasmids were identified by restriction analysis of plasmid DNA preparations from isolated transformants and digests using Fsp I were used to distinguish gene inserts of C γ 1 (no Fsp I) from C γ 4 (one site). The final vector containing the C.gamma.4-IL 2 heavy chain substitution was introduced into mouse myeloma cells and transformants were selected by growth in medium containing methotrexate (0.1. mu.M). Cells expressing high levels of antibody-IL 2 fusion protein were clonally expanded and the fusion protein was purified from culture supernatants using protein asecharose chromatography. The purity and integrity of the C γ 4 fusion protein was determined by SDS-polyacrylamide gel electrophoresis. I1-2 activity was measured in a T-cell proliferation assay and was found to be the same as that of the C γ 1 construct.

1.2 binding to Fc receptors by antibodies and antibody-IL 2 fusion proteins with C.gamma.1 and C.gamma.4 IgG constant regions

Various mouse and human cell lines may express one or more Fc receptors. For example, the mouse J774 macrophage-like cell line expresses FcR γ I capable of binding a suitable subclass of mouse or human IgG. Likewise, the human K562 erythroleukemia cell line expresses FcR γ II but not FcR γ I. To assess the potential contribution of Fc receptor binding to antibody-based fusion proteins cleared from circulation, the binding affinities of antibody, C γ 1-IL2 fusion protein, and C γ 4-IL2 fusion protein to FcR γ I were compared in the mouse J774 cell line.

The two antibody-IL 2 fusion proteins hu-KS γ 1-IL2 and hu-KS γ 4-IL2 described in example 1 were combined in a mixture of 0.1% Bovine Serum Albumin (BSA) and 2X 10 in a final volume of 0.2ml⁵J774 cells were diluted to 2. mu.g/ml in PBS. After incubation for 20 min on ice FITC conjugated anti-human IgG Fc antibody (Fab) was added₂) And incubation continued for another 30 minutes. Unbound antibody was removed by washing twice with PBS-BSA and cells were analyzed with a Fluorescence Activated Cell Sorter (FACS). Control reactions contained identical cells mixed with either just FITC labeled secondary antibody or with humanized KS γ 1 antibody (without IL-2).

As expected, the binding of the C gamma 4-IL2 fusion protein to J774 cells was significantly lower than the binding of the C gamma 1-IL2 fusion protein to J774 cells. See fig. 4. However, it was surprising that the binding of the C γ 1-IL2 and C γ 4-IL2 fusion proteins to J774 cells was significantly higher than the KS γ 1 antibody (without IL-2). This suggests that fusing a second protein, such as a cytokine, to the immunoglobulin can alter the structure of the antibody, resulting in increased binding affinity to one or more cell-bound Fc receptors, thereby resulting in rapid clearance from the circulation.

To determine whether the stronger binding observed with the IL-2 fusion protein was due to the presence of IL-2 receptor or FcR γ I receptor on the cells, an excess of mouse IgG (mIgG) was used to compete for binding on the Fc receptor. As illustrated in FIG. 4, background levels of binding were observed using the antibody and two antibody IL2 fusion proteins in the presence of a 50-fold molar excess of mIgG. This suggests that the increased signal binding of the antibody-IL 2 fusion protein is due to increased binding to Fc receptors.

Fc receptor expressing cell lines are used to test the binding affinity of candidate fusion proteins to Fc receptors in order to identify antibody-based fusion proteins with improved half-life in vivo. Candidate antibody-based fusion proteins can be detected by the methods described above. Candidates for antibody-based fusion proteins having greatly reduced binding affinity for Fc receptors are identified as antibody-based fusion proteins having increased in vivo half-life.

1.3 determination of circulating half-life of antibody-IL 2 fusion proteins with C.gamma.1 and C.gamma.4 IgG constant regions

To assess whether the use of an Fc region of an IgG isotype with reduced affinity for Fc receptors would increase circulating half-life in vivo, a fusion protein containing the heavy chain of the C γ 1 isotype (i.e., hu-KS γ 1-IL2) was compared to a fusion protein containing the heavy chain of the C γ 4 isotype (i.e., hu-KS γ -IL 2).

Purified humanized KS-1/4-IL2 fusion protein containing heavy chains of C γ 1 or C γ 4 isotype was buffer exchanged by diafiltration into Phosphate Buffered Saline (PBS) and further diluted to a concentration of-100 μ g/ml. Approximately 20. mu.g of antibody-based fusion protein (0.2ml) was injected into the tail vein of 6-8 week old Balb/c mice using a slow bolus injection method. Each group was injected with 4 mice. At various time points, a small blood sample was taken by bleeding from the back of the orbit of the anesthetized animal and collected into a tube containing citrate buffer to prevent clotting of the blood. Cells were removed by centrifugation for 5 minutes using an Eppendorf high speed tabletop centrifuge. Plasma was removed with a micropipette and frozen at-70 ℃. The concentration of human antibody determinants in the blood of mice was determined by ELISA. Capture antibodies specific for human H and L antibody chains were used to capture the fusion proteins from the diluted plasma samples. After 2 hours incubation in antibody coated 96 well plates, unbound material was removed by three washes with ELISA buffer (0.01% Tween 80 in PBS). The second culture step uses anti-human Fc antibodies (for detection of antibodies and intact fusion proteins) or anti-human IL-2 antibodies (for detection of intact fusion proteins only). Both antibodies were conjugated to horseradish peroxidase (HRP). After 1 hour of incubation, unbound detection antibody was removed by washing with ELISA buffer and the amount of bound HPR was determined by incubation with substrate and measurement with a spectrophotometer.

As depicted in FIG. 5, the alpha phase half-life of the hu-KS γ 4-IL2 fusion protein was significantly lower than the alpha phase half-life of the hu-KS γ 1-IL2 fusion protein. The increased half-life is best typified by a concentration of hu-KS γ 4-IL2 fusion protein (3.3 μ g/ml) that is significantly higher than the concentration of hu-KS γ 1-IL2 fusion protein (60ng/ml) found in mice after 24 hours.

As reported earlier for the chimeric 14, 18-IL2 fusion protein, the hu-KS γ 1-IL2 protein had a fast distribution (. alpha.) phase followed by a slower catabolic (. beta.) phase. See Gillies et al, bioligand chemistry (bioconj. chem.) 4: 230-235(1993). In the Gillies et al study, only antibody determinants were determined, and it is not clear whether clearance represents a cleared intact fusion protein or a cleared antibody component of the fusion protein. In this example, samples were tested using (1) an antibody-specific ELISA and (2) a fusion protein-specific ELISA (i.e., an ELISA that requires physically linked antibody and IL-2 components). . As illustrated in FIG. 5, in animals injected with the hu-KS γ 1-IL2 fusion protein, the amount of circulating fusion protein was lower than the total amount of circulating antibody, particularly at the 24 hour time point. This suggests that the fusion protein is proteolytically cleaved in vivo and the released antibody continues to circulate. Surprisingly, there was no significant difference between the amount of circulating fusion protein and the total amount of circulating antibody in the animals injected with the hu-KS γ 4-IL2 fusion protein. This suggests that the hu-KS γ 4-IL2 fusion protein did not proteolytically cleave in these animals over the measured 24 hour period.

As described above, C γ 1 and C γ 3 have binding affinity for Fc receptors, while at this time C γ 4 has reduced binding affinity and C γ 2 has no binding affinity for Fc receptors. This example describes methods for producing antibody-intact fusion proteins (an IgG isotype with reduced affinity for Fc receptors) using the C γ 4 Fc region and identifies such antibody-based fusion proteins with improved in vivo circulating half-life. Thus, one skilled in the art can use these methods to produce antibody-based fusion proteins having a C γ 2 Fc region in place of a C γ 4 Fc region in order to increase the circulating half-life of the fusion protein. Using the same restriction fragment substitutions and the above methods for the C.gamma.4-IL 2 fusion protein, hu-KS-IL2 fusion proteins employing the human C.gamma.2 region can be constructed and tested using the methods described herein to demonstrate increased circulating half-life. An antibody-based fusion protein having a C γ 2 Fc region or any other Fc region having binding affinity or lacking binding affinity for an Fc receptor has an increased circulating half-life in vivo compared to an antibody-based fusion protein having binding affinity for an Fc receptor.

Example 2 introduction of human C.gamma.1 or C.gamma.3 in antibody-based fusion protein constructs

The gene is mutated to increase the circulating half-life in vivo

IgG molecules interact with several molecules in the circulation, including members of the complement system of proteins (e.g., Clq fragments) as well as three classes of FcR. An important residue for Clq binding is residue Glu located in the CH2 domain of human heavy chain₃₁₈、Lys₃₂₀And Lys₃₂₂. Tao et al J.J.Experienct drugsMed.) 178: 661-667(1993). To differentiate the difference between FcR and Clq binding as a rapid clearance mechanism, we replaced the more completely altered C γ 2 hinge proximal fragment with the C γ 1 heavy chain. Such mutations are expected to affect FcR binding without affecting complement binding.

Mutation was accomplished by cloning and using the overlap Polymerase Chain Reaction (PCR) using a small region between the hinge and the initiator of the CH2 exon of the germline C γ 1 gene. The PCR primers were designed to replace the new sequence at the junction of two adjacent PCR fragments spanning the Psi I to Drd I fragments (see FIG. 6). In the first step, two independent PCR reactions using primers 1 and 2 (SEQ ID NOS: 5 and 6, respectively) or primers 3 and 4 (SEQ ID NOS: 7 and 8, respectively) were prepared using the C.gamma.1 gene as a template. The cycling conditions for the primary PCR were 35 cycles of the following conditions: 94 ℃ for 45 seconds; annealing at 48 ℃ for 45 seconds; and primer extension at 72 ℃ for 45 seconds. The product of each PCR reaction was used as a template for the secondary ligation reaction step. One tenth of each primary reaction was mixed with each other and combined with primers 1 and 4 to amplify only the combined products of the two initial PCR products. The conditions for the secondary PCR were: 94 ℃ for 1 minute; annealing at 51 ℃ for 1 minute; and primer extension at 72 ℃ for 1 minute. After denaturation and annealing, ligation reactions occur as a result of overlap between two separate fragments paired with the other fragment end. As shown in FIG. 6, fragments of the hybrid amplified by Taq polymerase are formed and the fully mutated product is selectively amplified by the priming of the outer primer. The final PCR product was cloned in a plasmid vector and its sequence was identified by DNA sequence analysis.

The assembly of the mutant gene is performed in multiple steps. In the first step, the cloning vector containing the human C.gamma.1 gene was digested with Psi I and XhoI to remove the unmutated hinge-CH 2-CH3 coding sequence. Drd I to Xho I fragments encoding part of CH2, all of CH3 and the fused human IL-2 coding sequence were prepared from the Cy 1-IL2 vector as described above. A third fragment was prepared from the PCR product of the subclone by digestion with Psi I and DrdI. All three fragments were purified by agarose gel electrophoresis and ligated to each other in a single reaction mixture. The ligation products were used to transform competent E.coli and clones were selected by growth on ampicillin containing plates. Correctly assembled recombinant plasmids were identified by restriction analysis of plasmid DNA preparations from isolated transformants and the mutated genes were confirmed by DNA sequence analysis. Hind III to Xho I fragments from mutant C.gamma.1-IL 2 were used to reassemble the complete hu-KS antibody-IL 2 fusion protein expression vector.

To assess the improvement in circulating half-life in vivo induced by important amino acid mutations for FcR binding and to distinguish between FcR and Clq binding as a mechanism of rapid clearance, plasma concentrations of in vivo mutated hu-KS γ 1-IL2 were compared to plasma concentrations of hu-KS γ 1-IL2 at various specific times. As illustrated in FIG. 7, the in vivo clearance of the mutated hu-KS γ 1-IL2 and hu-KS γ 4-IL2 was significantly lower than the clearance of hu-KS γ 1-IL 2. These results suggest that antibody-based fusion proteins with increased circulating half-life in vivo can be obtained by modifying the sequences necessary to bind Fc receptors within isoforms that have binding affinity for Fc receptors. In addition, these results suggest that the mechanism of rapid clearance involves FcR binding rather than Clq binding.

From the teachings of the present invention, those skilled in the art will appreciate that several other mutations to the C γ 1 or C γ 3 genes can be introduced to reduce FcR binding and increase the in vivo circulating half-life of antibody-based fusion proteins. In addition, mutations may also be introduced into the C γ 4 gene in order to further reduce the binding of the C γ 4 fusion protein to FcR. For example, additional possible mutations include mutations in the hinge proximal amino acid residue: pro bringing about₃₃₁By mutating the N-linked glycosylation sites in all IgGFc regions. The latter being positioned at Asn as part of the canonical sequence₂₈₇: Asn-X-Thr/Ser, where the second position can be any amino acid (possibly except Pro) and the third position is Thr or Ser. For example, conservative mutations in the amino acid Gln have little effect on the protein, but may prevent the attachment of any carbohydrate side chain. Strategies for mutating this residue may be employedThe general procedure for hinging the proximal region was as described above. Methods for generating point mutations in cloned DNA sequences are well established in the art and commercial kits are available from several vendors for this purpose.

Example 3 increasing the circulating half-life of receptor-antibody based fusion proteins

Several references have reported that the Fc portion of human IgG can be used as a useful carrier for a number of biologically active ligand binding proteins or receptors. Certain of these ligand binding proteins, such as CD4, CTLA-4, and TNF receptors, have been fused to the N-terminus of the Fc portion of Ig. See, e.g., Capon et al Nature (NATURE) 337: 525-531 (1989); linsley et al, J.EXP.MED., 174: 561-; wooley et al, J.Immunol., 151: 6602-6607(1993). Increasing the circulating half-life of the receptor-antibody based fusion protein may make the ligand binding protein partner (i.e., the second non-Ig protein) more effective: (1) blocking receptor-ligand interactions at the cell surface; or (2) counteract the biological activity of a molecule (e.g., a cytokine) in the blood fluid phase, thereby preventing it from reaching its cellular targets. To assess whether decreasing the ability of a receptor-antibody based fusion protein to bind an IgG receptor would increase its circulating half-life in vivo, a receptor-antibody based fusion protein having a human C γ 1 Fc region was compared to an antibody based fusion protein having a human C γ 4 Fc region.

To construct CD 4-antibody based fusion proteins, PCR from human Peripheral Blood Mononuclear Cells (PBMCs) was used to clone the extracellular domain of the human CD4 cell surface receptor. The cloned CD4 acceptor includes compatible restriction sites and splice donor sites as described in example 1. The expression vector contains the unique Xba I cloning site downstream of the CMV early promoter and the human C gamma 1 or C gamma 4 gene downstream of the endogenous HindIII site. The remainder of the plasmid contains the bacterial genetic information for propagation in E.coli and the dhfr selectable marker gene. The ligated DNAs were used to transform competent bacteria and recombinant plasmids were identified based on restriction analysis from individual bacterial clones. Two plasmid DNA constructs were obtained: CD4-C γ 1 and CD4-C γ 4.

The expression plasmid was used to transfect murine myeloma cells by electroporation and transformants were selected by growth in medium containing methotrexate (0.1. mu.M). Transfectants expressing the fusion protein were identified by ELISA analysis and amplified in culture medium to generate the fusion protein for purification by binding and elution from protein a Sepharose. The protein purified with chromatographic elution buffer was diafiltered into PBS and diluted to a final concentration of 100. mu.g/ml. Balb/C mice were injected with 0.2ml (20. mu.g) of CD 4-C.gamma.1 or CD 4-C.gamma.4 fusion protein and pharmacokinetic experiments were performed as described in example 1.3. The CD4-C gamma 4 fusion protein has a half-life which is obviously higher than that of the CD4-C gamma 1 fusion protein.

Equivalent mode

The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The above-described embodiments are therefore to be considered in all respects as illustrative and not restrictive of the invention described herein. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein.

<110>GILLIES，Stephen D

LO，Kin-Ming

LAN，Yan

WESOLOWSKI，John

<120> improvement of antibody-based fusion protein

Method for circulating half-life

<130>LEX-003PC

<140>

<141>

<150>US 60/075,887

<151>1998-02-25

<160>8

<170>PatentIn Ver.2.0

<210>1

<211>447

<212>PRT

<213> human

<220>

<223> IGG-1 chain C region

<400>1

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

1 5 10 15

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

20 25 30

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

35 40 45

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

50 55 60

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

65 70 75 80

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

85 90 95

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

100 105 110

Xaa Xaa Xaa Xaa Xaa Ala Ser Thr Lys Gly Pro Ser Val Phe Pro Leu

115 120 125

Ala Pro Ser Ser Lys Ser Thr Ser Gly Gly Thr Ala Ala Leu Gly Cys

130 135 140

Leu Val Lys Asp Tyr Phe Pro Glu Pro Val Thr Val Ser Trp Asn Ser

145 150 155 160

Gly Ala Leu Thr Ser Gly Val His Thr Phe Pro Ala Val Leu Gln Ser

165 170 175

Ser Gly Leu Tyr Ser Leu Ser Ser Val Val Thr Val Pro Ser Ser Ser

180 185 190

Leu Gly Thr Gln Thr Tyr Ile Cys Asn Val Asn His Lys Pro Ser Asn

195 200 205

Thr Lys Val Asp Lys Lys Val Glu Pro Lys Ser Cys Asp Lys Thr His

210 215 220

Thr Cys Pro Pro Cys Pro Ala Pro Glu Leu Leu Gly Gly Pro Ser Val

225 230 235 240

Phe Leu Phe Pro Pro Lys Pro Lys Asp Thr Leu Mer Ile Ser Arg Thr

245 250 255

Pro Glu Val Thr Cys Val Val Val Asp Val Ser His Glu Asp Pro Glu

260 265 270

Val Lys Phe Asn Trp Tyr Val Asp Gly Val Glu Val His Asn Ala Lys

275 280 285

Thr Lys Pro Arg Glu Glu Gln Tyr Asn Ser Thr Tyr Arg Val Val Ser

290 295 300

Val Leu Thr Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu Tyr Lys

305 310 315 320

Cys Lys Val Ser Asn Lys Ala Leu Pro Ala Pro Ile Glu Lys Thr Ile

325 330 335

Ser Lys Ala Lys Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr Leu Pro

340 345 350

Pro Ser Arg Asp Glu Leu Thr Lys Asn Gln Val Ser Leu Thr Cys Leu

355 360 365

Val Lys Gly Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser Asn

370 375 380

Gly Gln Pro Glu Asn Asn Tyr Lys Thr Thr Pro Pro Val Leu Asp Ser

385 390 395 400

Asp Gly Ser Phe Phe Leu Tyr Ser Lys Leu Thr Val Asp Lys Ser Arg

405 410 415

Trp Gln Gln Gly Asn Val Phe Ser Cys Ser Val Met His Glu Ala Leu

420 425 430

His Asn His Tyr Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly Lys

435 440 445

<210>2

<211>443

<212>PRT

<213> human

<220>

<223> IGG-2 chain C region

<400>2

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

1 5 10 15

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

20 25 30

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

35 40 45

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

50 55 60

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

65 70 75 80

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

85 90 95

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

100 105 110

Xaa Xaa Xaa Xaa Xaa Ala Ser Thr Lys Gly Pro Ser Val Phe Pro Leu

115 120 125

Ala Pro Cys Ser Arg Ser Thr Ser Glu Ser Thr Ala Ala Leu Gly Cys

130 135 140

Leu Val Lys Asp Tyr Phe Pro Glu Pro Val Thr Val Ser Trp Asn Ser

145 150 155 160

Gly Ala Leu Thr Ser Gly Val His Thr Phe Pro Ala Val Leu Gln Ser

165 170 175

Ser Gly Leu Tyr Ser Leu Ser Ser Val Val Thr Val Pro Ser Ser Asn

180 185 190

Phe Gly Thr Gln Thr Tyr Thr Cys Asn Val Asp His Lys Pro Ser Asn

195 200 205

Thr Lys Val Asp Lys Thr Val Glu Arg Lys Cys Cys Val Glu Cys Pro

210 215 220

Pro Cys Pro Ala Pro Pro Val Ala Gly Pro Ser Val Phe Leu Phe Pro

225 230 235 240

Pro Lys Pro Lys Asp Thr Leu Met Ile Ser Arg Thr Pro Glu Val Thr

245 250 255

Cys Val Val Val Asp Val Ser His Glu Asp Pro Glu Val Gln Phe Asn

260 265 270

Trp Tyr Val Asp Gly Val Glu Val His Asn Ala Lys Thr Lys Pro Arg

275 280 285

Glu Glu Gln Phe Asn Ser Thr Phe Arg Val Val Ser Val Leu Thr Val

290 295 300

Val His Gln Asp Trp Leu Asn Gly Lys Glu Tyr Lys Cys Lys Val Ser

305 310 315 320

Asn Lys Gly Leu Pro Ala Pro Ile Glu Lys Thr Ile Ser Lys Thr Lys

325 330 335

Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr Leu Pro Pro Ser Arg Glu

340 345 350

Glu Met Thr Lys Asn Gln Val Ser Leu Thr Cys Leu Val Lys Gly Phe

355 360 365

Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser Asn Gly Gln Pro Glu

370 375 380

Asn Asn Tyr Lys Thr Thr Pro Pro Met Leu Asp Ser Asp Gly Ser Phe

385 390 395 400

Phe Leu Tyr Ser Lys Leu Thr Val Asp Lys Ser Arg Trp Gln Gln Gly

405 410 415

Asn Val Phe Ser Cys Ser Val Met His Glu Ala Leu His Asn His Tyr

420 425 430

Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly Lys

435 440

<210>3

<211>494

<212>PRT

<213> human

<220>

<223> IGG-3 chain C region

<400>3

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

1 5 10 15

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

20 25 30

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

35 40 45

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

50 55 60

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

65 70 75 80

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

85 90 95

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

100 105 110

Xaa Xaa Xaa Xaa Xaa Ala Ser Thr Lys Gly Pro Ser Val Phe Pro Leu

115 120 125

Ala Pro Cys Ser Arg Ser Thr Ser Gly Gly Thr Ala Ala Leu Gly Cys

130 135 140

Leu Val Lys Asp Tyr Phe Pro Glu Pro Val Thr Val Ser Trp Asn Sar

145 150 155 160

Gly Ala Lsu Thr Ser Gly Val His Thr Phe Pro Ala Val Leu Gln Ser

165 170 175

Ser Gly Leu Tyr Ser Leu Ser Ser Val Val Thr Val Pro Ser Ser Ser

180 185 190

Leu Gly Thr Gln Thr Tyr Thr Cys Asn Val Asn His Lys Pro Ser Asn

195 200 205

Thr Lys Val Asp Lys Arg Val Glu Leu Lys Thr Pro Leu Gly Asp Thr

210 215 220

Thr His Thr Cys Pro Arg Cys Pro Glu Pro Lys Ser Cys Asp Thr Pro

225 230 235 240

Pro Pro Cys Pro Arg Cys Pro Glu Pro Lys Ser Cys Asp Thr Pro Pro

245 250 255

Pro Cys Pro Arg Cys Pro Glu Pro Lys Ser Cys Asp Thr Pro Pro Pro

260 265 270

Cys Pro Arg Cys Pro Ala Pro Glu Leu Leu Gly Gly Pro Ser Val Phe

275 280 285

Leu Phe Pro Pro Lys Pro Lys Asp Thr Leu Mer Ile Ser Arg Thr Pro

290 295 300

Glu Val Thr Cys Val Val Val Asp Val Ser His Glu Asp Pro Glu Val

305 310 315 320

Gln Phe Lys Trp Tyr Val Asp Gly Val Glu Val His Asn Ala Lys Thr

325 330 335

Lys Pro Arg Glu Glu Gln Tyr Asn ser Thr Phe Arg Val Val Ser Val

340 345 350

Leu Thr Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu Tyr Lys Cys

355 360 365

Lys Val Ser Asn Lys Ala Leu Pro Ala Pro Ile Glu Lys Thr Ile Ser

370 375 380

Lys Thr Lys Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr Leu Pro Pro

385 390 395 400

Ser Arg Glu Glu Met Thr Lys Asn Gln Val Ser Leu Thr Cys Leu Val

405 410 415

Lys Gly Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser Ser Gly

420 425 430

Gln Pro Glu Asn Asn Tyr Asn Thr Thr Pro Pro Mer Leu Asp Ser Asp

435 440 445

Gly Ser Phe Phe Leu Tyr Ser Lys Leu Thr Val Asp Lys Ser Arg Trp

450 455 460

Gln Gln Gly Asn Ile Phe Ser Cys Ser Val Met His Glu Ala Leu His

465 470 475 480

Asn Arg Phe Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly Lys

485 490

<210>4

<211>444

<2l2>PRT

<213> human

<220>

<223> IGG-4 chain C region

<400>4

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

1 5 10 15

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

20 25 30

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

35 40 45

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

50 55 60

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

65 70 75 80

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

85 90 95

Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa

100 105 110

Xaa Xaa Xaa Xaa Xaa Ala Ser Thr Lys Gly Pro Ser Val Phe Pro Leu

115 120 125

Ala Pro Cys Ser Arg Ser Thr Ser Glu Ser Thr Ala Ala Leu Gly Cys

130 135 140

Leu Val Lys Asp Tyr Phe Pro Glu Pro Val Thr Val Ser Trp Asn Ser

145 150 155 160

Gly Ala Leu Thr Ser Gly Val His Thr Phe Pro Ala Val Leu Gln Ser

165 170 175

Ser Gly Leu Tyr Ser Leu Ser Ser Val Val Thr Val Pro Ser Ser Ser

180 185 190

Leu Gly Thr Lys Thr Tyr Thr Cys Asn Val Asp His Lys Pro Ser Asn

195 200 205

Thr Lys Val Asp Lys Arg Val Glu Ser Lys Tyr Gly Pro Pro Cys Pro

210 215 220

Ser Cys Pro Ala Pro Glu Phe Leu Gly Gly Pro Ser Vel Phe Leu Phe

225 230 235 240

Pro Pro Lys Pro Lys Asp Thr Leu Mer Ile Ser Arg Thr Pro Glu Val

245 250 255

Thr Cys Val Val Val Asp Val Ser Gln Glu Asp Pro Glu Val Gln Phe

260 265 270

Asn Trp Tyr Val Asp Gly Val Glu Val His Asn Ala Lys Thr Lys Pro

275 280 285

Arg Glu Glu Gln Phe Asn Ser Thr Tyr Arg Val Val Ser Val Leu Thr

290 295 300

Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu Tyr Lys Cys Lys Val

305 310 315 320

Ser Asn Lys Gly Leu Pro Ser Ser Ile Glu Lys Thr Ile Ser Lys Ala

325 330 335

Lys Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr Leu Pro Pro Ser Gln

340 345 350

Glu Glu Met Thr Lys Asn Gln Val Ser Leu Thr Cys Leu Val Lys Gly

355 360 365

Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser Asn Gly Gln Pro

370 375 380

Glu Asn Asn Tyr Lys Thr Thr Pro Pro Val Leu Asp Ser Asp Gly Ser

385 390 395 400

Phe Phe Leu Tyr Ser Arg Leu Thr Val Asp Lys Ser Arg Trp Gln Glu

405 410 415

Gly Asn Val Phe Ser Cys Ser Val Met His Glu Ala Leu His Asn His

420 425 430

Tyr Thr Gln Lys Ser Leu Ser Leu Ser Leu Gly Lys

435 440

<210>5

<211>17

<212>DNA

<213> Artificial sequence

<220>

<223> description of artificial sequences: primer 1

<400>5

catcggtctt ccccctg 17

<210>6

<211>35

<212>DNA

<213> Artificial sequence

<220>

<223> description of artificial sequences: primer 2

<400>6

cggtcctgcg acgggaggtg ctgaggaaga gatgg 35

<210>7

<211>45

<212>DNA

<213> Artificial sequence

<220>

<223> description of artificial sequences: primer 3

<400>7

tcttcctcag cacctcccgt cgcaggaccg tcagtcttcc tcttc 45

<210>8

<211>17

<212>DNA

<213> Artificial sequence

<220>

<223> description of artificial sequences: primer 4

<400>8

gaggcgtggt cttgtag 17

Claims (8)

1. A fusion protein having reduced binding affinity for one or more Fc receptors, comprising an immunoglobulin (Ig) heavy chain or a portion thereof, and a second non-immunoglobulin protein linked to said immunoglobulin heavy chain, wherein said immunoglobulin heavy chain comprises:

(i) an IgG2 or IgG4 constant region but not an IgG1 or IgG3 constant region, wherein the reduced binding affinity for an Fc receptor is compared to a corresponding fusion protein comprising an IgG1 or IgG2 constant region;

(ii) modified IgG1, wherein the modification is to the unmodified oneThe reduced binding affinity for Fc receptors of said fusion protein is responsible for the modified Fc receptor by mutation or deletion at one or more amino acid positions selected from Leu₂₃₄、Leu₂₃₅、Gly₂₃₆、Gly₂₃₇、Asn₂₉₇And Pro₃₃₁Of the group consisting of

(iii) Modified IgG3, wherein the modification is responsible for the reduced binding affinity of the fusion protein for Fc receptor compared to the unmodified fusion protein, the modification being a mutation or deletion at one or more amino acid positions selected from Leu₂₈₁、Leu₂₈₂、Gly₂₈₃、Gly₂₈₄、Asn₃₄₄And Pro₃₇₈Group (d) of (a).

2. The fusion protein of claim 1, wherein the IgG2 or IgG4 constant region comprises at least a CH2 domain.

3. The fusion protein of claim 1 or 2, wherein the immunoglobulin (Ig) heavy chain or portion thereof further has binding affinity for an immunoglobulin protection receptor.

4. The fusion protein of claim 1 or 2, wherein said second non-immunoglobulin albumin is selected from the group consisting of a cytokine, a ligand binding protein, and a protein toxin.

5. The fusion protein of claim 1 or 2, further comprising a variable region thereby displaying antigen binding specificity.

6. A method of increasing the in vivo circulating half-life of an antibody fusion protein comprising an IgG1 or IgG3 heavy chain constant region, or a portion thereof, linked to non-immunoglobulin protein, the method comprising the steps of:

(i) replacing the IgG1 or IgG3 heavy chain constant region with an IgG2 or IgG4 constant region;

(ii) to IgG1 heavy chainIntroducing mutation or deletion at one or more amino acid sites in the constant region, wherein the amino acid is selected from Leu₂₃₄、Leu₂₃₅、Gly₂₃₆、Gly₂₃₇、Asn₂₉₇And Pro₃₃₁Of the group consisting of

(iii) Introducing a mutation or deletion into the IgG1 heavy chain constant region at one or more amino acid positions selected from the group consisting of Leu₂₈₁、Leu₂₈₂、Gly₂₈₃、Gly₂₈₄、Asn₃₄₄And Pro₃₇₈Group (d) of (a).

7. The method of claim 6, wherein the IgG2 or IgG4 constant region comprises at least a CH2 domain.

8. The method of claim 6 or 7, wherein the non-immunoglobulin albumin is selected from the group consisting of a cytokine, a ligand binding protein, and a protein toxin.