Non-cross-reactive anti IgG antibodies

Abstract

Herein are reported the cell lines DSM ACC3006, DSM ACC3007, and DSM ACC3008, as well as the antibodies obtained from the cell lines and the use of an antibody obtained from the cell lines in an immunoassay. Also are reported antibodies binding to human or chimpanzee IgG and not binding to canine and marmoset IgG and antibodies specifically binding to an IgG 1 that comprises a kappa light chain constant domain.

Claims

1. A hybridoma cell line selected from the group consisting of DSM ACC3006, DSM ACC3007 and DSM ACC3008.

2. An antibody produced by one of the hybridoma cell lines of claim 1.

3. The antibody of claim 2, wherein the hybridoma cell line is DSM ACC3006.

4. The antibody of claim 2, wherein the hybridoma cell line is DSM ACC3007.

5. The antibody of claim 2, wherein the hybridoma cell line is DSM ACC3008.

6. A method for detecting a human or humanized therapeutic antibody comprising a human kappa light chain, or a Fab fragment thereof, in a sample obtained from an experimental animal comprising the steps of a) providing a sample to be analyzed, the sample obtained from an experimental animal selected from the group consisting of a canine, a rhesus-monkey, a marmoset, a baboon, and a cynomolgus monkey, b) incubating said sample with a capture antibody that is bound to a solid phase, wherein said capture antibody is the antibody of any one of claims 3 to 5, c) incubating said sample with a detection antibody that is conjugated to a detectable label so as to form a complex, wherein said detection antibody is the antibody of any one of claims 3 to 5, and d) detecting the detectable label on the detection antibody, thereby detecting the therapeutic antibody or Fab fragment thereof.

7. The method of claim 6, wherein the capture antibody is conjugated to biotin and immobilized to the solid phase via avidin or streptavidin.

8. The method of claim 6, wherein the detectable label is digoxigenin and the step of detecting the detectable label is performed via an antibody against digoxigenin.

9. The method of claim 6, further comprising correlating the detectable label that is conjugated to the detection antibody to a concentration of the therapeutic antibody, or Fab fragment thereof, in the sample.

10. The method of claim 6, further comprising washing the sample after incubating the sample with the capture antibody bound to the solid phase.

11. The method of claim 7, wherein the detectable label is digoxigenin and the step of detecting the detectable label is performed via an antibody against digoxigenin.

12. An antibody composition, comprising a mixture of antibodies produced by hybridoma cell line DSM ACC3006, hybridoma cell line DSM ACC3007, and/or hybridoma cell line DSM ACC3008.

Description

DESCRIPTION OF THE FIGURES

(1) FIGS. 1a)-1c) Fully generic Assay for quantification of human antibodies (human IgG) in an experimental animal: FIG. 1a) assay format; FIG. 1b) capture and detection reagent: antibody M-R10Z8E9; FIG. 1c) capture and detection reagent antibody M-1.7.10; therapeutic antibodies: empty triangles: anti-IL13R1 antibody, empty squares: anti-Abeta antibody, solid squares: anti-IL1R antibody, solid triangles anti-IL6R antibody.

(2) FIG. 2 Results obtained with an assay employing the antibodies as reported herein; antibodies used from left to right: M-R10Z8E9, M-1.3.2, M-1.5.8, M-1.7.10.

(3) FIGS. 3a)-3c) Exemplary surface plasmon surface resonance diagrams of antibodies: FIG. 3a) M-1.3.2, FIG. 3b) M-1.5.8, and FIG. 3c) M-1.7.10 as reported herein.

(4) FIG. 4 Dot Blot of anti-human IgG antibodies; as exemplary reference antibody an antibody against P-selectin has been chosen; the reference antibody is dotted is native (left column) and denatured (right column) onto a nitrocellulose membrane and detected by the respective digoxigenylated anti-human IgG antibodies; left to right: M-R10Z8E9, M-1.3.2, M-1.5.8, M-1.7.10.

(5) FIGS. 5a) and 5b) Assay for quantifying human antibody derivates in a sample obtained from an experimental animal: FIG. 5a) schematic assay set-up, FIG. 5b) calibration curve.

(6) FIGS. 6a) and b) Assay to proof structural integrity of human IgG in an experimental animal: FIG. 6a) schematic assay set-up, FIG. 6b) calibration curve.

(7) FIG. 7 Selection of antibodies with no detectable cross-reactivity to cynomolgus serum.

EXAMPLE 1

(8) Preparation of the F(Ab).sub.2 Fragment of Human IgG (Immunogen)

(9) The full length human antibody of the class G (human IgG) in 100 mM sodium citrate buffer, pH 3.7 was incubated with pepsin (1 g pepsin per mg IgG). The fragmentation was analyzed by analytical gel filtration and stopped after 90 minutes by adjusting the pH value to 6.5 by the addition of potassium phosphate. After dialysis of the mixture against 10 mM sodium citrate buffer with 10 mM sodium chloride, pH 5.5, the solution was applied to an SP-Sepharose chromatography column and the isolated fractions eluted in a salt gradient were analyzed individually by analytical gel filtration. The pool containing the antibody F(ab).sub.2 fragments were applied to an affinity matrix with immobilized polyclonal antibodies against human Fc to eliminate trace amounts of Fc fragments. The flow through was pooled, concentrated to about 16 mg/ml and finally applied to a gel filtration column (Superdex 200).

EXAMPLE 2

(10) Generation of Monoclonal Anti-Human IgG Antibodies

(11) a) Immunization of Mice

(12) Female NMRI mice, 8-12 weeks of age, were each primarily immunized intraperitoneally with 100 g of the antibody F(ab).sub.2 fragments prepared according to Example 1 mixed with CFA (Complete Freund's Adjuvant). Two further intraperitoneal immunization steps followed after 6 and 10 weeks, each with 100 g of the antibody F(ab).sub.2 fragments per mouse mixed with IFA (Incomplete Freund's adjuvant). Subsequently, intravenous boost immunizations were done, each with 50 g of antibody F(ab).sub.2 fragments in PBS (phosphate buffered saline) three days before the fusion.

(13) b) Fusion and Cloning

(14) Spleen cells of the mice immunized according to a) were fused with myeloma cells according to Galfre and Milstein (Galfre, G. and Milstein, C, Methods Enzymol. 73 (1981) 3-46). Approximately 2.110.sup.8 splenocytes were mixed with 4.210.sup.7 myeloma cells (P3x63-Ag8.653, ATCC CRL1580) and centrifuged (10 min. at 300g and 4 C.). The cells were washed afterwards once with the culture medium PvPMI 1640 without FCS (fetal calf serum), and centrifuged again at 400g in a 50 ml pointed vial. Thereafter, 1 ml of PEG (poly (ethylene glycol), molecular weight 4,000 g/mol) was added, mixing was done by the pipetting. After 1 min. in a water bath at 37 C., 5 ml of RPMI 1640 without FCS were added drop wise, the suspension was mixed, RPMI 1640 with 10% (v/v) FCS was added to a final volume of 50 ml, and then centrifuged. The sedimented cells were resuspended in RPMI 1640 with 10% FCS, and plated in hypoxanthine-azaserine selection medium (100 mmol/1 hypoxanthine, 1 g/ml azaserine in RPMI 1640 with 10% FCS) containing the growth factor recombinant murine interleukin 6 (Peprotech, 0.5 ng/ml). After 11 days, the primary cultures were assayed for specific antibody synthesis (see Example 3). Primary cultures exhibiting binding to biotinylated antibody F(ab).sub.2 fragments as well as to biotinylated human normal IgG were individualized by single cell deposition into 96-well cell culture plates using a flow cytometer (FACSAria, BD Biosciences) in medium containing the growth factor recombinant murine interleukin 6 (Peprotech, 0.5 ng/ml). By following this protocol, the cell lines DSM ACC3006, DSM ACC3007, and DSM ACC3008 were obtained. Antibody M-1.7.10 is of the IgG2a class, antibodies M-1.5.8 and M-1.3.2 are of the IgG1 class.

(15) c) Production of Immunoglobulin

(16) The hybridoma cell lines obtained in b) were inoculated at initial cell densities (live cells) between 1.010.sup.5 and 2.210.sup.5 cells per ml in RPMI 1640 supplemented with 10% FCS, and commonly used supplements and expanded in a T-flask (Celline, IBS) for a period of approximately three weeks. In the harvested culture supernatants, concentrations between 0.7 mg/ml and 1.5 mg/ml of monoclonal antibody were obtained. Purification of the antibodies from the culture supernatants was done according to standard protein chemical methods, e.g. as those reported in Bruck, C., et al, Methods Enzymol. 121 (1986) 587-596.

EXAMPLE 3

(17) Screening Assays for Detection of Anti-Human IgG Antibodies

(18) a) Primary Screening for Antibodies Binding to Human IgG

(19) For the determination of the specificity of the antibodies in the culture supernatants of the hybridoma cells, MTPs (microtiter plates) pre-coated with recombinant streptavidin (MicroCoat, Bernried, lot MC 1098) were coated with biotinylated humanized IgG used for the immunization process, 250 ng/ml, or biotinylated human IgG, 250 ng/ml, respectively, in PBS supplemented with 1% (w/v) BSA II (100 l per well, 60 min. incubation at ambient temperature, with shaking), and subsequently washed three times with 0.9% (w/v) NaCl/0.05% Tween 20. In the next step, per well 100 l of the antibody solution to be assayed (culture supernatant) were added, and incubated for 60 min. at ambient temperature, with shaking. After three wash steps with 0.9% (w/v) NaCl/0.05% Tween 20 per well, 100 l of a horseradish peroxidase-labeled F(ab).sub.2 fragment of a polyclonal sheep anti-mouse Fc antibody were added for the detection of bound sample antibody, and incubated for 60 min. at ambient temperature, with shaking. Subsequently, washing was performed as above. Finally, 100 l per well of ABTS (Roche Diagnostics GmbH, Mannheim, Germany; catalog no. 1684302) were added. After 30 min. incubation at ambient temperature, the extinction (OD) was measured at 405 and 492 nm [405/492] in a commercial microtiter plate ELISA Reader. This screening led to a selection of antibodies binding well to humanized IgG as well as to human IgG. This selection of antibodies was further subjected to assay b).

(20) b) Selection of Antibodies with Minimal Cross-Reactivity to IgG of Other Species

(21) Biotinylated human IgG was bound to the wells of a streptavidin-coated microtiterplate (SA-MTP) in the first step. The excess of unbound antibody was removed by washing. Afterwards the samples and the reference standards (e.g. anti-human IgG antibody as obtained with Example 2) were diluted in buffer and 10% cynomolgus serum. Diluted samples were added to the plate and incubated for 60 min. at ambient temperature, with shaking. After having washed away unbound substances, the human IgG of the first step in digoxygenylated form was added to the wells of the plate and incubated for another 60 min. After washing, the bound digoxygenylated antibody was detected with an anti-digoxygenin antibody-HRP conjugate. The HRP (horseradish peroxidase) of the antibody-enzyme conjugates catalyzes the color reaction of ABTS substrate. The signal is measured by ELISA reader at 405 nm wavelength (reference wavelength: 490 nm). Absorbance values of each serum sample were determined in triplicates.

(22) Antibodies with high assay response in cynomolgus serum as well as in buffer were selected (see FIG. 7). This second screening led to a selection of antibodies binding well to human IgG with minimal cross-reactivity to IgG of other species.

EXAMPLE 4

(23) Assessment of Antibody Binding/Specificity by Surface Plasmon Resonance

(24) All measurements were performed with the BIAcore T100 instrument using a CMS-chip. Coating of this chip with an antibody was achieved by standard amine coupling. Unless otherwise indicated, all incubations were performed in HBS-buffer (HEPES, NaCl, pH 7.4) at 25 C. A saturating amount of a polyclonal goat anti-mouse Fc-gamma antibody was immobilized by amine coupling on one flow cell of the CMS-chip. Subsequently, the different monoclonal mouse antibodies directed against human IgG were injected for 60 seconds at a flow rate of 30 l/min and were bound by the anti mouse Fc antibody. All animal sera were diluted in HBS buffer. Binding was analyzed by injection of the 1 in 100 diluted sera and incubation for 60 sec. at a flow rate of 30 l/min. Dissociation was measured by washing the chip surface with HBS buffer for 180 sec. Using BIAevaluation Software from BIAcore the dissociation constant values (=K.sub.D) were calculated with a 1:1 Langmuir fitting model. For all animal sera this calculation was based on the assumption that the IgG level is 15 mg/ml. The signal values 80 sec. after start of the injection of the test antibody were chosen for the comparison of the amount of IgG bound (see Table 1).

(25) TABLE-US-00002 TABLE 1 Binding signals [RU] and K.sub.D-values for binding of animal sera to different monoclonal anti-human IgG antibodies. Antibody M-R10Z8E9 M-1.3.2 Sample (serum) Bound RU K.sub.D mol/l Bound RU K.sub.D mol/l Chimpanzee 159 2.21 10.sup.10 95.7 1.12 10.sup.09 Human 151.3 1.77 10.sup.10 80.1 1.43 10.sup.09 Dog 35.5 3.17 10.sup.8.sup. 1.9 no binding Rhesus-monkey 1.9 no binding 2.3 no binding Marmoset 18.9 2.04 10.sup.7.sup. 2 no binding Baboon 1.5 no binding 2.2 no binding Cynomolgus 1.4 no binding 2 no binding Antibody M-1.5.8 M-1.7.10 Sample (serum) Bound RU K.sub.D mol/l Bound RU K.sub.D mol/l Chimpanzee 109.4 1.29 10.sup.09 109.4 1.94 10.sup.09 Human 77 1.43 10.sup.09 77 7.55 10.sup.09 Dog 2.4 no binding 2.4 no binding Rhesus-monkey 2.7 no binding 2.7 no binding Marmoset 2.1 no binding 2.1 no binding Baboon 2.1 no binding 2.1 no binding Cynomolgus 2.1 no binding 2.1 no binding

(26) Table 1 shows that the three anti-human IgG antibodies do not cross-react with serum from other species except Chimpanzee. In contrast, for M-R10Z8E9 an additional interaction with serum from dog and marmoset was detected.

EXAMPLE 5

(27) a) Purification of Mouse Monoclonal Anti-Human IgG Antibody

(28) The fermentation supernatant of the cell lines obtained in Example 2 was concentrated about tenfold and transferred to a buffer with 20 mM TRIS, 1 M ammonium sulfate, pH 9.0, and applied to a protein A-Sepharose chromatography column. The eluate obtained with 0.2 M sodium citrate, 0.2 M ammonium sulfate at pH 5.0 was dialyzed against phosphate buffer, pH 7.5. Contaminants of bovine IgG (from FCS in the fermentation broth) were separated by immunoadsorption with immobilized antibodies against bovine IgG.

(29) b) Preparation of Biotinylated Anti-Human IgG Antibody

(30) The anti-human IgG antibody obtained in a) in phosphate buffer, pH 8.5, was adjusted to a protein concentration of about 5 mg/ml. D-biotinoyl-aminocaproic acid-N-hydroxysuccinimide was dissolved in DMSO and added to the antibody solution in a molar ratio of 1:5. The reaction was stopped after 60 min. by adding L-lysine, and the surplus of the labeling reagent was removed by dialysis against 50 mM potassium phosphate buffer, with 150 mM NaCl, pH 7.5.

(31) c) Preparation of Digoxigenylated Anti-Human IgG Antibody

(32) The anti-human IgG antibody obtained in a) in phosphate buffer, pH 8.5, was adjusted to a protein concentration of about 5 mg/ml. Digoxigenin 3-O-methylcarbonyl--aminocaproic acid-N-hydroxysuccinimide was dissolved in DMSO and added to the antibody solution in a molar ratio of 1:4. The reaction was stopped after 60 min. by adding L-lysine, and the surplus of the labeling reagent was removed by dialysis against 50 mM potassium phosphate buffer, with 150 mM NaCl, pH 7.5.

EXAMPLE 6

(33) Fully Generic Assay for Quantification of Human Antibodies (Human IgG) in a Sample from an Experimental Animal

(34) Biotinylated antibody M-R10Z8E9 (plate 1) or antibody M-1.7.10 (plate 2) was bound to streptavidin-coated microtiter plates (SA-MTP) in the first step. The excess of unbound antibody was removed by washing. Samples/standards, e.g. anti-IL1R antibody, anti-IL13R1 antibody, anti-Abeta antibody and anti-IL6R antibody, spiked in cynomolgus serum were added in a concentration series to the plate and incubated for 60 min. at ambient temperature, with shaking. After having washed away unbound antibodies, 100 l digoxygenylated antibody M-R10Z8E9 (plate 1) or antibody M-1.7.10 (plate 2) was added to the plate. After washing, the bound digoxygenylated antibodies were detected with an anti-digoxygenin-antibody-HRP conjugate. Absorbance values of each serum sample were determined in triplicates (see FIG. 1b) for plate 1 and FIG. 1c) for plate 2).

(35) TABLE-US-00003 TABLE 2 OD data for capture and detection reagent antibody M-R10Z8E9. anti- anti- anti- anti- IL13R1 Abeta IL1R IL6R ng/ml antibody antibody antibody antibody 0.00 0.022 0.023 0.024 0.024 1.56 0.119 0.139 0.085 0.105 3.13 0.226 0.264 0.153 0.190 6.25 0.408 0.482 0.276 0.348 12.50 0.772 0.881 0.546 0.664 25.00 1.229 1.310 0.980 1.084 50.00 1.672 1.707 1.521 1.565 100.00 1.967 1.927 1.877 1.819

(36) TABLE-US-00004 TABLE 3 OD data for capture and detection reagent antibody M-1.7.10. anti- anti- anti- anti- IL13R1 Abeta IL1R IL6R ng/ml antibody antibody antibody antibody 0.00 0.038 0.036 0.035 0.037 1.56 0.178 0.149 0.187 0.181 3.13 0.325 0.264 0.326 0.312 6.25 0.570 0.472 0.568 0.540 12.50 1.004 0.853 1.013 0.955 25.00 1.592 1.407 1.588 1.498 50.00 1.995 1.923 2.013 1.947 100.00 2.197 2.213 2.209 2.185

EXAMPLE 7

(37) Assay for Quantification of Human Antibody Derivates (e.g. Fab-Fragments) in a Sample from an Experimental Animal

(38) Biotinylated antibody M-1.7.10 was bound to streptavidin-coated microtiter plates (SA-MTP) in the first step. The excess of unbound antibody was removed by washing. Samples/standards, e.g. anti-IGFIR antibody Fab fragment, spiked in cynomolgus serum were added to the wells and incubated for 60 min. at ambient temperature, with shaking. After having washed away unbound antibodies, 100 l digoxigenylated antibody M-1.3.2 was added to each well of the plate. After washing, the bound digoxygenylated antibodies were detected with an anti-digoxygenin antibody-HRP conjugate. Absorbance values of each serum sample were determined in triplicates (see FIG. 5b)).

(39) TABLE-US-00005 TABLE 4 OD data. ng/ml OD 405 nm SD 0.00 0.042 0.000 1.56 0.047 0.000 3.13 0.057 0.002 6.25 0.103 0.001 12.50 0.247 0.016 25.00 0.694 0.007 50.00 1.535 0.043 100.00 1.882 0.013

EXAMPLE 8

(40) Assay to Prove Structural Integrity of Human IgG in a Sample from an Experimental Animal

(41) Biotinylated antibody M-R10Z8E9 directed against human Fc was bound to streptavidin-coated microtiter plates (SA-MTP) in the first step. The excess of unbound antibody was removed by washing. Samples/standards, e.g. anti-IL13R1 antibody, spiked in cynomolgus serum were added to the plate and incubated for 60 min. at ambient temperature, with shaking. After having washed away unbound antibodies, 100 l digoxygenylated antibody M-1.3.2 was added to the plate. After washing, the bound digoxygenylated antibodies were detected with an anti-digoxygenin antibody-HRP conjugate. Absorbance values of each serum sample were determined in triplicates (see FIG. 3a) for M-1.3.2, FIG. 3b) for M-1.5.8, and FIG. 3c) for M-1.7.10).

(42) TABLE-US-00006 TABLE 5 OD data. ng/ml OD 405 nm SD 0.00 0.023 0.018 0.78 0.094 0.008 1.56 0.172 0.007 3.13 0.304 0.011 6.25 0.588 0.015 12.50 1.051 0.007 25.00 1.604 0.004 50.00 2.019 0.001

EXAMPLE 9

(43) Assay for Quantification of Human Antibodies (Human IgG) in a Sample from an Experimental Animal Using a Fc-Fusion Protein (Antigen) in Combination with Anti-Human IgG Antibodies as Reported Herein

(44) Soluble extracellular domain of a human receptor X is fused to the Fc-fragment of human IgG1 class. The biotinylated fusion protein (Bi-X-Fc) was bound to streptavidin-coated microtiter plates (SA-MTP) in the first step. The excess of unbound receptor was removed by washing. Afterwards anti-X antibody spiked in cynomolgus serum was bound to the immobilized human receptor X. After washing away unbound substances, the bound anti-X antibody was detected with a) digoxygenylated monoclonal antibody against human Fc fragment (antibody M-R10Z8E9) or with b) digoxygenylated monoclonal antibody against human Fab fragment (antibody M-1.7.10) followed by incubation with a horse-radish peroxidase labeled anti-digoxygenin antibody. Absorbance values of each serum sample are determined in triplicates.

EXAMPLE 10

(45) Dot BlotConformation Vs. Linear Epitope

(46) To determine, whether the anti-human IgG antibodies detect a conformation epitope or a linear epitope, a dot-blot analytic was performed.

(47) During this test, the antigen-protein (human IgG) was dotted to a nitrocellulose membrane in a native and a denaturized form. To receive the denaturized form, the antigen-protein was incubated with SDS on a shaker at 37 C. overnight. Both forms were dotted in a concentration series to the membrane. After complete drying of the membrane, the surface was blocked with a blocking buffer (Roti-Block, Roth, Germany) for 60 min. at ambient temperature with shaking. After washing of the membrane, it was incubated with a solution containing digoxygenylated antibody M-R10Z8E9 or one of the three different antibodies M-1.3.2, M-1.5.8, or M-1.7.10. After washing, the bound digoxygenylated antibodies were detected with an anti-digoxygenin antibody-HRP conjugate. The HRP of the antibody-enzyme conjugates catalyzes the color reaction of BM-Blue substrate. The signal can directly be controlled visually and captured with a scanner.

EXAMPLE 11

(48) Assessment of Antibody Binding/Specificity by a Bridging ELISA Assay

(49) To determine which kind of Human IgG subclass is bound by the researched anti human antibodies, a bridging ELISA analytic was performed.

(50) Biotinylated antibodies M-R10Z8E9, M-1.3.2, M-1.5.8 and M-1.7.10 were bound to the streptavidin microtiterplate in the first step. In a second step, human IgG antibodies of different subclasses were incubated. Human IgG1 kappa; human IgG1 lambda; human IgG4; chimeric human IgG1; human IgG2 (polyclonal purified human IgG2) and human IgG3 (polyclonal purified human IgG3) were prepared in a dilution series and incubated to the streptavidin microtiter plate, coated with biotinylated anti human antibody. After a washing step, the same antibodies as used for coating were used as detection antibodies in digoxygenylated form. This means that the same anti human antibody clone was used for coating and detection. For example, one plate was coated with M-1.7.10 Bi and M-1.7.10-Dig was used for detection. After incubation and a washing step, this step was followed by incubation with a horse-radish peroxidase labeled anti-digoxygenin antibody. Absorbance values of each serum sample have been determined in triplicates.

(51) TABLE-US-00007 TABLE 6 Resume of bridging ELISA analytics Antibody used for coating/detection mAb mAb mAb mAb Sample M-R10Z8E9 M-1.3.2 M-1.5.8 M-1.7.10 IgG1-kappa ++ ++ ++ ++ IgG1-Lambda ++ IgG4 ++ + + ++ Chimeric IgG1 ++ + + ++ IgG2 + + + ++ IgG3 + IgG1-kappa Fab ++ ++ ++ IgG1-Lambda Fab ++ strong binding + binding + weak binding no binding