Multispecific Antibodies Facilitating Selective Light Chain Pairing

Abstract

Provided herein are multispecific antibodies, e.g. bispecific antibodies, which are modified such that the desired chain pairing takes place and/or can be selected for. Specifically, this is achieved by using different dimerization domains for light chain pairing. Also disclosed herein are nucleic acids encoding for these antibodies, expression vectors comprising these nucleic acids, cells expressing them, and further to pharmaceutical compositions comprising the antibodies, as well as methods of isolating the antibodies.

Claims

1: An antibody heavy or light chain A or dimer thereof, wherein the heavy chain comprises a variable domain V.sub.HA linked to a dimerization domain CH2.sub.H and the light chain comprises a variable domain V.sub.LA linked to a dimerization domain CH2.sub.L, and wherein both CH2.sub.H and CH2.sub.L are an IgM constant domain MC.sub.H2 or an IgE constant domain EC.sub.H2.

2: An antibody or derivative thereof, comprising a heavy chain A and a light chain A according to claim 1.

3: The antibody or derivative of claim 2, further comprising a heavy chain B and a light chain B, wherein the heavy chain B comprises a variable domain V.sub.HB and the light chain B comprises a variable domain V.sub.LB.

4: The antibody or derivative of claim 3, wherein the antibody or derivative is multispecific.

5: The antibody or derivative of claim 3, comprising a first and a second heavy chain A and a first and a second light chain A each as defined in claim 1, wherein: (a) the CH2.sub.L domain of the first light chain A is linked to the V.sub.H1 or V.sub.L1 domain of an IgG antibody or derivative thereof and the CH2.sub.L domain of the second light chain A is linked to the other V.sub.H1 or V.sub.L1 domain of the IgG antibody or derivative thereof; or (b) the CH2.sub.H domain of the first light chain A is linked to the V.sub.H1 or V.sub.L1 domain of an IgG antibody or derivative thereof and the CH2.sub.H domain of the second light chain A is linked to the other V.sub.H1 or V.sub.L1 domain of the IgG antibody or derivative thereof.

6: The antibody or derivative of claim 3, wherein the antibody or derivative is asymmetric.

7: The antibody or derivative of claim 6, wherein V.sub.HB is linked to a dimerization domain DIM.sub.H and V.sub.LB is linked to a dimerization domain DIM.sub.L, and wherein DIM.sub.H and DIM.sub.L are dimerization domains different from CH2.sub.H and CH2.sub.L, respectively.

8: The antibody or derivative of claim 7, wherein: (a) DIM.sub.H is linked to CH2.sub.H or CH2.sub.L, or DIM.sub.L is linked to CH2.sub.H or CH2.sub.L; or (b) CH2.sub.H is linked via a hinge region to a constant domain C.sub.H2A, DIM.sub.H is linked via a hinge region to a constant domain C.sub.H2B.

9: The antibody or derivative of claim 8, wherein CH2.sub.H is linked via a hinge region to a constant domain C.sub.H2A, and DIM.sub.H is linked via a hinge region to a constant domain C.sub.H2B, wherein C.sub.H2A is linked to a constant domain C.sub.H3A and C.sub.H2B is linked to a constant domain C.sub.H3B.

10: The antibody or derivative of claim 8, wherein CH2.sub.H is linked via a hinge region to a constant domain C.sub.H2A, and DIM.sub.H is linked via a hinge region to a constant domain C.sub.H2B, wherein: (a) heavy chain A further comprises a further heavy chain variable domain V.sub.HX linked to V.sub.HA and light chain A further comprises a further light chain variable domain V.sub.LX linked to V.sub.LA; and/or (b) heavy chain B further comprises a further heavy chain variable domain V.sub.HY linked to V.sub.HB and light chain B further comprises a further light chain variable domain V.sub.LY linked to V.sub.LB.

11: The antibody or derivative of claim 6, wherein the antibody or derivative thereof comprises one or more of: (a) a T366Y mutation and optionally further an S354C and T166W mutation in one heavy chain, and a Y407T mutation and optionally further a Y349C, T366S, L368A, and Y407V mutation in the other heavy chain; (b) a T366W mutation in one heavy chain, and a T366S, L368A and Y407V mutation in the other heavy chain; (c) a F405L mutation in one heavy chain, and a K409R mutation in the other heavy chain; (d) a T350V, L351Y, F405A, and Y407V mutation in one heavy chain, and a T350V, T366L, K392L, and T394W mutation in the other heavy chain; (e) a K409D and K392D mutation in one heavy chain, and a D399K and E356K mutation in the other heavy chain; (f) a D221E, P228E, and L368E mutation in one heavy chain, and a D221R, P228R, and K409R mutation in the other heavy chain; (g) a S364H and F405A mutation in one heavy chain, and a Y349T and T394F mutation in the other heavy chain; (h) an IgG/A chimera for one heavy chain, and an IgA/G chimera for the other heavy chain; (i) an Fc region or part thereof of one heavy chain from IgG3, and an Fc region or part thereof of the other heavy chain from IgG1, IgG2, or IgG4; (j) a H435R and Y436F mutation in one heavy chain, and a T407T mutation in the other heavy chain; and/or (k) a H435R mutation in one heavy chain, and no mutation in the other heavy chain.

12: The antibody or derivative of claim 11, wherein the heavy chains comprise knob-into-hole mutations and one or more mutations decreasing the binding to Protein A.

13: The antibody or derivative of claim 11, wherein the heavy chain comprising the hole mutation(s) of the knob-into-hole mutations comprises the one or more mutations decreasing the binding to Protein A.

14: One or more polynucleotides encoding for the heavy chain A and/or the light chain A of claim 1 or for an antibody or derivative comprising a heavy chain A and a light chain A according to claim 1.

15: One or more expression vectors comprising the one or more polynucleotides of claim 14.

16: A cell comprising the one or more polynucleotides of claim 14 or one or more expression vectors comprising the one or more polynucleotides of claim 14.

17: A pharmaceutical composition comprising a pharmaceutically acceptable carrier and a dimer of the heavy chain A and the light chain A of claim 1 or an antibody or derivative comprising a heavy chain A and a light chain A according to claim 1.

18: A method of isolating the antibody or derivative thereof of the second aspect, comprising the steps of: (a) providing a solution comprising a heavy chain A and a light chain A as defined in claim 1 and a heavy chain B and a light chain B; wherein the heavy chain B comprises a variable domain V.sub.HB and the light chain B comprises a variable domain V.sub.LB; and (b) purifying the antibody or derivative without a means for selecting for the pairing of heavy chain A with light chain A and/or the pairing of heavy chain B with light chain B.

19: The method of claim 18, wherein the antibody or derivative is purified with a means for selecting for a pairing of heavy chain A with heavy chain B.

20: The method of claim 18, wherein the means is: (a) selecting for Protein A binding; (b) selecting for the presence of a C.sub.H1 domain or a C domain, or for the absence of a CH2 domain; or (c) a combination of (a) and (b).

21: The method of claim 20, wherein in (c), (a) is carried out prior to (b).

22: The method of claim 18, wherein the antibody or derivative thereof comprises a heavy chain A, a light chain A, a heavy chain B, and a light chain B, wherein: (a) the heavy chain A comprises a variable domain V.sub.HA linked to a dimerization domain CH2.sub.H; (b) the light chain A comprises a variable domain V.sub.LA linked to a dimerization domain CH2.sub.L; (c) both CH2.sub.H and CH2.sub.L are an IgM constant domain MC.sub.H2 or an IgE constant domain EC.sub.H2; (d) the heavy chain B comprises a variable domain V.sub.HB linked to a dimerization domain DIM.sub.H, wherein DIM.sub.H is different from CH2.sub.H; (e) the light chain B comprises a variable domain V.sub.LB linked to a dimerization domain DIM.sub.L, wherein DIM.sub.L is different from CH2L; wherein the heavy chain comprising the hole mutation(s) of the knob-into-hole mutations comprises the one or more mutations decreasing the binding to Protein A; and wherein the antibody or derivative is purified by selecting for Protein A binding, characterized in that the method does not comprise selecting for the presence of a C.sub.H1 domain or a C domain, or for the absence of a CH2 domain.

Description

DESCRIPTION OF THE FIGURES

[0212] FIG. 1: Examples of antibody or derivative configurations. Different dimerization domains ensure correct pairing of the variable domains. A: IgG configuration (linkers not shown except Fc-Fab linking by the hinge region), B: Bivalent Fab configuration (linkers not shown except Fab-Fab linker; this linker may alternatively link CH2.sub.H with DIM.sub.L, CH2.sub.L with DIM.sub.H or CH2.sub.L with DIM.sub.L), C: TBTI configuration (linkers not shown except Fc-Fab linking by the hinge region), D: CODV configuration showing two possible linker configurations on the left and on the right arm (both arms can also have the same linker configuration, i.e. that of the left or of the right arm shown), E: Tetravalent spider configuration (linkers not shown except Fc-Fab linking by the hinge region and Fab-Fab linkers), F: trivalent configuration (example for the CODV chimera configuration) containing three binding sites allowing for mono-, bi- and tri-specificity. The right arm shown is a CODV arm having the linker configuration of the right arm of the CODV configuration of FIG. 1 D, but it may instead also have the linker configuration of this CODV configuration. The left arm is a conventional IgG arm. Left and right arms are exchangeable, i.e. the left arm can be the CODV arm and the right arm the IgG arm. Further, the MCH2 domains are shown to be in the IgG arm, but they can instead be in the CODV arm (independent of its linker configuration) with the DIM domains then being in the IgG arm.

[0213] FIG. 2: Schematic representation of the architecture of the asymmetric antibody. A: On one half of the antibody like structure the CH1/kappa region is replaced by the MCH2 domains. The Fc of the Fab containing half is chimeric, harbouring the CH3 from IgG3 whereas all other Fc domains are of IgG1 origin. A two step purification including protein A and kappa select chromatography allows the isolation of the correct assembled heterodimeric molecule. B: SDS-PAGE showing the chain composition under reducing conditions. C: SEC profile of the purified heterodimeric antibody.

[0214] FIG. 3: Schematic representation of the architecture of the asymmetric antibody. A: On one half of the antibody like structure the CH1/kappa region is replaced by the MCH2 domains. The heterodimerization is driven by knobs-into-holes mutations within the Fc parts. The Fc part with the hole mutations further contains the RF mutation. A two-step purification including protein A and kappa select chromatography allows the isolation of the correct assembled heterodimeric molecule. B: SDS-PAGE showing the chain composition under reducing conditions. C: SEC profile of the purified heterodimeric antibody.

[0215] FIG. 4: Tetravalent spider configuration. A: Schematic representation of the symmetric IgG like structure. A single step protein A chromatography allows the isolation of the correctly assembled homodimeric molecule. B: SDS-PAGE showing the chain composition under reducing conditions. C: SEC profile of the purified heterodimeric antibody. D and E: Biacore sensograms showing binding to the corresponding antigens.

[0216] FIG. 5: Schematic representation of the architecture of the asymmetric antibody. A: On one half of the antibody like structure the CH1/kappa region is replaced by the ECH2 domains. The heterodimerization is driven by knobs-into-holes mutations within the Fc parts. The Fc part with the knob mutations further contains the RF mutation. B: SDS-PAGE showing the chain composition under reducing conditions. C: SEC profile of the purified protein. D: Biacore sensograms showing binding to IL4 and IL13.

[0217] FIG. 6: Bivalent Fab configuration. A: schematic representation of the bivalent bispecific Fab fragment. B: SEC profile of the purified protein. C and D: Biacore sensograms showing binding to CD3 and CD123 respectively.

[0218] FIG. 7: CODV bivalent Fab configuration (EC.sub.H2). A: schematic representation of the protein architecture. B: SEC profile of the purified protein. C: Biacore sensograms showing binding to both IL4 and IL13.

[0219] FIG. 8: F(ab)2 like configuration. A: schematic representation of the bivalent bispecific F(ab)2 like fragment. B: SEC profile of the purified heterodimeric antibody. C: SDS-PAGE showing the chain composition under reducing conditions. D: Biacore sensograms showing binding to IL4 and IL13.

DESCRIPTION OF THE EXAMPLES

[0220] Material & Methods

[0221] Generation of Expression Plasmids

[0222] All described constructs were built from gene synthesis fragments (GeneArt), cleaved with restriction enzymes and ligated into mammalian expression vector pXL. The vector was used to transform E. coli and clones were selected and amplified to isolate DNA for transient transfection using Qiagen Kits. The DNA sequence was verified by sequencing of the relevant ORF and sequence context. The resulting coding ORFs are in accordance with the before listed and described sequences.

[0223] Expression of Recombinant Proteins

[0224] All protein production campaigns were realized via transient expression. Freestyle HEK293 (Life) cells growing in F17 serum free suspension culture (Life) were transfected with the expression plasmid. Transfections were performed using Cellfectin transfection reagent (Life). The cells were cultured at 37 C. for 7 days at 8% CO.sub.2 in shaker flasks. The culture supernatant containing recombinant protein was harvested by centrifugation and was clarified by filtration (0.22 m). Recombinant IgG1 constructs were purified by affinity chromatography on Protein A (HiTrap Protein A HP Columns, GE Life Sciences). After elution from the column with 100 mM acetate buffer and 100 mM NaCl, pH 3.5, the CODV-IgG1 constructs were desalted using HiPrep 26/10 Desalting Columns, formulated in PBS at a concentration of 1 mg/mL and filtered using a 0.22 m membrane. Fab-like constructs were purified by IMAC on HiTrap IMAC HP Columns (GE Life Sciences). After elution from the column with a linear gradient (Elution buffer: 20 mM sodium phosphate, 0.5 M NaCl, 50-500 mM imidazole, pH 7.4), the protein containing fractions were pooled and desalted using HiPrep 26/10 Desalting Columns, formulated in PBS at a concentration of 1 mg/mL and filtered using a 0.22 m membrane. For purification of asymmetric heterodimeric constructs the protein samples were additionally applied on His-Trap column (GE) and/or captured by Kappa select (GE) after capture on protein A and desalting step. The protein was polished by SEC using a Superdex 200 (GE). After a final ultrafiltration concentration step the protein was used for different assays. This strategy was used to isolate heterodimers from homodimers. Protein concentration was determined by measurement of absorbance at 280 nm. Each batch was analyzed by SDS-PAGE under reducing and non-reducing conditions to determine the purity and molecular weight of each subunit and of the monomer.

[0225] Characterization of MCH2 and ECH2 Variants

[0226] To determine whether the MCH2 and ECH2 antibody-like protein heavy and light chains were pairing and folding properly, the aggregation level was measured by analytical size-exclusion chromatography (SEC). Analytical SEC was performed on assembled pairs using an KTA explorer 10 (GE Healthcare) equipped with a TSKgel G3000SWXL column (7.8 mm30 cm) and TSKgel SWXL guard column (Tosoh Bioscience). The analysis was run at 1 ml/min. using 250 mM NaCl, 100 mM Na-phosphate, pH 6.7, with detection at 280 nm. 30 l of protein sample (at 0.5-1 mg/ml) were applied onto the column. For estimation of the molecular size, the column was calibrated using a gel filtration standard mixture (MWGF-1000, SIGMA Aldrich). Data evaluation was performed using UNICORN software v5.11.

[0227] Binding Analysis by SPR

[0228] Two pairs of heavy and light chains were selected for full kinetic analysis. Recombinant human IL13 and IL4 was purchased from Chemicon (USA). Kinetic characterization of purified antibodies was performed using surface plasmon resonance technology on a BIACORE 3000 (GE Healthcare). A capture assay using a species-specific antibody (e.g., human-Fc specific MAB 1302, Chemicon) for capture and orientation of the investigated antibodies was used. The capture antibody was immobilized via primary amine groups (11000 RU) on a research grade CM5 chip (GE Life Sciences) using standard procedures. The analyzed antibody was captured at a flow rate of 10 L/min with an adjusted RU value that would result in maximal analyte binding of 30 RU. Binding kinetics were measured against recombinant human IL4 and IL13 over a concentration range between 0 to 25 nM in HBS EP (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, and 0.005% Surfactant P20) at a flow rate of 30 l/min. Chip surfaces were regenerated with 10 mM glycine, pH 2.5. Kinetic parameters were analyzed and calculated in the BIAevaluation program package v4.1 using a flow cell without captured antibody as reference.

[0229] Redirected Cell Killing

[0230] Peripheral blood mononuclear cells (PBMCs) were isolated from 200 ml peripheral blood of healthy donors treated with EDTA by Ficoll density centrifugation. 15 ml Histopaque (Sigma-Aldrich) was preloaded on a 50 ml Leucosep-Tube (Greiner bio-one). Blood was diluted with autoMACS Rinsing Buffer+1% BSA (Miltenyi Biotec) and loaded on the membrane of a total of ten prepared tubes. Tubes were centrifuged without brake for 10 min at 1000 xg. PBMCs were collected and washed with autoMACS Rinsing Buffer+1% BSA three times. Finally, PBMCs were resuspended in autoMACS Running Buffer (Miltenyi Biotec) for isolation of T lymphocytes by autoMACSpro technology using the Pan T Cell isolation Kit (Miltenyi Biotec) according to manufacturer's instructions. Purity of separated T cells was analyzed by MACSQuant flow cytometry using the human 7-Color Immunophenotyping Kit (Miltenyi Biotec). T-cell engaging effect of bispecific antibodies was analyzed by a flow cytometry based cytotoxic assay. Target cells (i.e. THP-1 cell line) were stained for 15 min at 37 C. with 1 M CFSE in 1 ml RPMI+GlutaMAX I (Gibco) per 1E7 cells. Afterwards, cells were washed twice and resuspended in RPMI+GlutaMAX I+10% FCS (Invitrogen). 2.5E4 target cells were seeded in 96-well U-bottom suspension culture plates (Greiner bio-one) in 50 l medium per well. Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX I+10% FCS and were added at indicated effector-to-target ratio in 50 l per well to the target cells (e.g. E:T=10:1). Bispecific antibodies were diluted 1:3 in serial in PBS (Invitrogen) and 5 l each were added to the cells at a final maximum concentration of 3000 ng/ml. The assay was incubated for 20 h at 37 C. in 5% CO2. To detect dead target cells, all cells were stained with 7-AAD. Therefore, 5 g/ml 7-AAD diluted in Stain Buffer with FBS (BD Pharmingen) were added to each well and were incubated for 15 min at 4 C. in the dark. Cells were measured using the MACSQuant (Miltenyi Biotec) or LSRII (BD) flow cytometer, respectively. Further data analyses were performed using the FlowJo software (Tree Star, Inc.). Read out was percentage of CFSE and 7-AAD double positive cells. The results of these investigations demonstrate the ability of the CD123xCD3 Fabs to mediate redirected killing of tumor cells.

Example 1: IgG Configuration (MC.SUB.H.2), RF Mutations

[0231] As described in the Material & Methods section, an antibody in the IgG configuration comprising an IgG3-CH3 domain in one Fc chain and an IgG1-CH3 domain in the other and variable domains being IL13 and IL4 was generated, wherein the CH2.sub.H and CH2.sub.L domains are an IgM constant domain MC.sub.H2 (see FIG. 2A). It was purified with the following steps:

[0232] 1. Protein A chromatography

[0233] 2. Kappa select chromatograpy

[0234] 3. Desalting on HiPrep 26/10

[0235] The yield was 4 mg/l. SDS-PAGE on NuPAGE Novex 4-12% Bis-Tris (FIG. 2B) showed the expected number and size of fragments under reducing conditions. SEC analysis of the purified protein showed a monomer level of 96% (FIG. 2C). Biacore analysis showed binding to the corresponding antigens and binding kinetics as expected from the parental antibodies (Table 2).

TABLE-US-00002 TABLE 2 Binding kinetics of the purified bispecific antibody against IL4 and IL13 Analyte ka (1/Ms) Kd (1/s) KD (M) IL13 1.49E+06 8.21E05 5.50E11 IL4 1.51E+08 2.53E04 1.68E12

Example 2: IgG Configuration (MC.SUB.H.2), RF and Knob-into-Hole Mutations

[0236] As described in the Material & Methods section, an antibody in the IgG configuration comprising mutations H435R and Y436F (RF mutations) in one CH3 domain as well as know-into-hole mutations and variable domains binding IL13 and IL4 was generated (see FIG. 3A). It was purified with the following steps:

[0237] 1. Protein A chromatography

[0238] 2. Kappa select chromatograpy

[0239] 3. Desalting on HiPrep 26/10

[0240] The yield was with 10 mg/l more than double as high as for the variant described in example 1, indicating a higher efficiency of correct chain association driven by the knob-into-hole modification of the Fc fragments. SDS-PAGE on NuPAGE Novex 4-12% Bis-Tris (FIG. 3B) showed under reducing conditions the expected number and size of fragments SEC analysis of the purified protein showed a monomer level of 96% (FIG. 3C). Biacore analysis showed binding to the corresponding antigens and binding kinetics as expected from the parental antibodies.

Example 3: Tetravalent Spider Configuration

[0241] As described in the Material & Methods section, an antibody in the Tetravalent spider configuration (see FIG. 4A) was generated. It was purified with the following steps:

[0242] 1. Protein A chromatography

[0243] 2. Desalting on HiPrep 26/10

[0244] The yield was 5 mg/l. SDS-PAGE on NuPAGE Novex 4-12% Bis-Tris (FIG. 4B) showed all expected fragments with the expected size under non-reducing conditions. SEC analysis showed a monomer level of 98% (FIG. 4C). Biacore analysis (FIGS. 4D and E) showed binding against the corresponding antigens.

Example 4: IgG Configuration (EC.SUB.H.2), RF and Knob-into-Hole Mutations

[0245] This bivalent bispecific antibody (see FIG. 5A) contains a CH2 domain of human IgE (Uniprot P01854) from amino acid position 104 to 210. The domain contains mutation of cysteine 105 to alanine to avoid unwanted cys reactivity and mutation of asparagine 146 to glutamine to avoid glycosylation. In the following examples called CH2E. To fuse the CH2E part to the HC of the construct, a flexible linker (GSGSGS) was introduced.

[0246] Example using variable domains huBB13 and hu8D4-8 recognizing antigens: IL4 and IL13 described before (US 20100226923 A1) composed in an IgG like structure consisting of two different Fabs each representing a different target VD fused to a modified IgG1-Fc domain. To enhance the expression and formation of heterodimers KIH (knob into hole) mutations were introduced into the CH3 domains. L234A, L235A mutations (Hezareh et al. 2001, J. Virol., 75: 12161) were incorporated to prevent effector function of the Fc backbone. The ECH2 domain is replacing CH1/kappa.

[0247] The antibody comprises the chains

[0248] anti-IL4(hu8D4-8-VL1)-CL1gk

[0249] anti-IL4(hu8D4-8-VH1)-CH1g-Fc(IgG1 LALA knob RF)

[0250] anti-IL13-VL(huBB13-VL3)-CH2e

[0251] anti-IL13-(huBB13-VH2)-CH2e-Fc(IgG1 LALA hole)

[0252] A two-step affinity chromatography was performed to ensure enrichment of heterodimeric chain pairs. The protein was captured by HiTrap Protein A 5 ml (GE Healthcare) and eluted by pH shift. Protein fractions were collected and buffer was instantly exchanged to PBS by a HiPrep 26/10 Desalting 53 ml desalting column (GE Healthcare). The protein solution was applied to HiTrap KappaSelect 5 ml (GE Healthcare) and eluted via pH shift. Protein containing fractions were collected and buffer was instantly exchanged to PBS by a HiPrep 26/10 Desalting 53 ml desalting column (GE Healthcare). Fractions were pooled and sample volume was reduced via ultrafiltration. The protein was polished by SEC using a Superdex 200 (GE Healthcare) column. After a final ultrafiltration and 0.22 m filtration step the protein solution was used for further assays.

[0253] The yield after purification was 18 mg/l. SDS-PAGE on NuPAGE Novex 4-12% Bis-Tris (FIG. 5B) showed the expected number and size of fragments under reducing conditions. SEC analysis of the purified protein showed a monomer level of 99% (FIG. 5C). Biacore analysis showed binding to the corresponding antigens IL4 and IL13 (FIG. 5D).

Example 5: Bivalent Fab Configuration

[0254] As described in the Material & Methods section, an antibody in Bivalent Fab configuration was generated, wherein the two Fabs are linked via a linkage of a MCH2 domain of one Fab-like fragment (CD3) to a variable domain of the other Fab (CD123), see FIG. 6A. The yield was 7 mg/l. SEC analysis showed a monomer level of 97% (FIG. 6B). Biacore analysis showed binding to the corresponding antigens (FIGS. 6C and D) and binding kinetics as expected from the parental antibodies (Table 3)

TABLE-US-00003 TABLE 3 Binding kinetics of the prurified bispecific bivalent Fab against CD3 and CD123 Analyte ka (1/Ms) Kd (1/s) KD (M) CD123 4.5E+05 3.77E05 8.38E11 CD3 1.581E+05 9.00E04 5.71E09

Example 6: CODV Bivalent Fab Configuration (EC.SUB.H.2)

[0255] The bispecific F-like fragment (see FIG. 7A) contains a CH2 domain of human IgE (Uniprot P01854) from amino acid position 104 to 210. The domain contains mutation of cysteine 105 to alanine to avoid unwanted cys reactivity and mutation of asparagine 146 to glutamine to avoid glycosylation. In the example, variable domains huBB13 and hu8D4-8 are used recognizing antigens IL4 and IL13 described before (US 20100226923 A1). The molecule is composed in the CODV format. In this example, the ECH2 domain is replacing the CH1 and CL domains in some embodiments located at the C-terminal end of both chains to connect HC and LC via disulfide bridges. An 8x histidine tag was added for purification.

[0256] The antibody comprises 2 unique chains:

[0257] Chain 1: anti-IL13(huBB13-VH2)-anti-IL4(hu8D4-8-VH1)-ECH2-His

[0258] Chain 2: anti-IL4(hu8D4-8-VL1)-anti-IL13(huBB13-VL2)-ECH2

[0259] The His-tagged protein was captured on HisTrap High Performance 5 ml (GE Healthcare) and eluted by an imidazole gradient. The protein was polished by SEC using a Superdex 200 (GE Healthcare) By a final ultrafiltration concentration step the protein was concentrated and used for further assays. FIG. 7B shows the SEC profile of the Fab-like fragment after purification. Biacore analysis showed binding of both IL4 and IL13 (FIG. 7C).

Example 7: F(Ab)2 Like Configuration (EC.SUB.H.2 Replacing the Hinge Region)

[0260] This F(ab)2-like antibody (see FIG. 8A) contains a CH2 domain of human IgE (Uniprot P01854) from amino acid position 104 to 210. The domain contains mutation of cysteine 105 to alanine to avoid unwanted cys reactivity and mutation of asparagine 146 to glutamine to avoid glycosylation. Also in this example, an inclusion of flexible linker connecting the CH2E domain on the heavy chain was introduced. Example using variable domains huBB13 and hu8D4-8 recognizing antigens IL4 and IL13 described before (US 20100226923 A1) composed in a Fab.sub.2 like format. In this example, the ECH2 domain is replacing the hinge region to connect two Fab modules via HC parts. Generation of a bispecific molecule is done via heterodimerization of two different Fabs representing two separate target specifities. One Fab arm contains a MCH2 domain the other corresponds to a native Fab structure. In this example, anti-IL4 VD is connected to a CH1/kappa domain and the anti-IL13 VD is fused to a modified MCH2 domain. An 8x histidine tag was added at the Cter of one ECH2 domain for purification. The antibody comprises 4 unique chains:

[0261] anti-IL4 (hu8D4-8-VH1)-CH1g-CH2e

[0262] anti-IL4 (hu8D4-8-VL1)-Ck

[0263] anti-IL13 (huBB13-VH2)-MCH2-ECH2-8xHis

[0264] anti-IL13 (huBB13-VL3)-MCH2

[0265] A two-step affinity chromatography was performed to ensure enrichment of heterodimeric chain pairs. The protein was captured by HiTrap KappaSelect 5 ml (GE Healthcare) and eluted following manufacturer's instructions by pH shift. By a desalting column the buffer was immediately exchanged to PBS. After the protein solution was applied to HisTrap High Performance 5 ml (GE Healthcare) and eluted by an imidazole gradient. Afterwards the protein containing fractions were pooled and further polished by SEC using a Superdex 200 (GE Healthcare). After a final ultrafiltration concentration step the protein was used for further assays. The SEC profile after purification revealed a monomeric fraction of 99% (FIG. 8B). A reducing SDS PAGE (Novex 4-12% Bis-Tris) showed the expected 4 fragment sizes (FIG. 8C). The construct is able to bind both IL4 and IL13 as demonstrated in Biacore analysis (FIG. 8D).