Anti human annexin A1 antibody

Abstract

The present invention relates to an isolated specific binding molecule which binds human Anx-A1 and comprises the complementarity-determining regions (CDRs) VLCDR1, VLCDR2, VLCDR3, VHCDR1, VHCDR2 and VHCDR3, wherein each of said CDRs has an amino acid sequence as follows: VLCDR1 has the sequence set forth in SEQ ID NO: 1, 36 or 37; VLCDR2 has the sequence set forth in SEQ ID NO: 2; VLCDR3 has the sequence set forth in SEQ ID NO: 3; VHCDR1 has the sequence set forth in SEQ ID NO: 4; VHCDR2 has the sequence set forth in SEQ ID NO: 5; and VHCDR3 has the sequence set forth in SEQ ID NO: 6; or, for each sequence, an amino acid sequence with at least 85% sequence identity thereto. The specific binding molecule disclosed is therapeutically useful and in particular may be used in therapy for T-cell mediated diseases, including autoimmune diseases such as rheumatoid arthritis and systemic lupus erythematosus, obsessive compulsive disorder (OCD), and OCD-related diseases, such as anxiety disorders.

Claims

1. An isolated specific binding molecule which binds human Anx-A1, said specific binding molecule comprising the complementarity-determining regions (CDRs) VLCDR1, VLCDR2, VLCDR3, VHCDR1, VHCDR2 and VHCDR3, wherein each of said CDRs has an amino acid sequence as follows: VLCDR1 has the sequence set forth in SEQ ID NO: 36 or 37; VLCDR2 has the sequence set forth in SEQ ID NO: 2; VLCDR3 has the sequence set forth in SEQ ID NO: 3; VHCDR1 has the sequence set forth in SEQ ID NO: 4; VHCDR2 has the sequence set forth in SEQ ID NO: 5; and VHCDR3 has the sequence set forth in SEQ ID NO: 6 and wherein the specific binding molecule is an antibody or antigen-binding fragment thereof.

2. The specific binding molecule of claim 1, wherein: VLCDR1 has the sequence set forth in SEQ ID NO: 37; VLCDR2 has the sequence set forth in SEQ ID NO: 2; VLCDR3 has the sequence set forth in SEQ ID NO: 3; VHCDR1 has the sequence set forth in SEQ ID NO: 4; VHCDR2 has the sequence set forth in SEQ ID NO: 5; and VHCDR3 has the sequence set forth in SEQ ID NO: 6.

3. The specific binding molecule of claim 1, wherein the antibody or fragment thereof is humanized.

4. The specific binding molecule of claim 1, wherein when said specific binding molecule is an antibody, the antibody is a monoclonal antibody, or when said specific binding molecule is a fragment of an antibody, said fragment is an Fab or F(ab′).sub.2 antibody fragment, or an scFv molecule.

5. The specific binding molecule of claim 1, wherein said specific binding molecule binds human Anx-A1 with a K.sub.d of less than 20 nM.

6. The specific binding molecule of claim 4, wherein the antibody or fragment thereof comprises: (i) a light chain variable region comprising the amino acid sequence set forth in any one of SEQ ID NOs: 48-51, or an amino acid sequence having at least 70% sequence identity thereto; and (ii) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 33 or 35, or an amino acid sequence having at least 70% sequence identity thereto.

7. The specific binding molecule of claim 6, wherein the antibody or fragment thereof comprises: (i) a light chain comprising the amino acid sequence set forth in any one of SEQ ID NOs: 44-47, 54, 76, 78 or 79 or an amino acid sequence having at least 70% sequence identity thereto; and (ii) a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 41, 43, 55, or 80 or an amino acid sequence having at least 70% sequence identity thereto.

8. The specific binding molecule of claim 7, wherein the antibody comprises: (i) a light chain comprising the amino acid sequence set forth in SEQ ID NO: 44; and a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 41; or (ii) a light chain comprising the amino acid sequence set forth in SEQ ID NO: 54; and a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 55.

9. A preparation containing the specific binding molecule of claim 1, wherein at least 90% of the specific binding molecules in the preparation that bind to human Anx-A1 bind with a K.sub.d of less than 20 nM.

10. A pharmaceutical composition comprising a specific binding molecule as defined in claim 1 and one or more pharmaceutically acceptable diluents, carriers or excipients.

11. The pharmaceutical composition of claim 10, wherein the specific binding molecule is an antibody or fragment thereof comprising: (i) a light chain variable region comprising the amino acid sequence set forth in any one of SEQ ID NOs: 48-51, or an amino acid sequence having at least 70% sequence identity thereto; and (ii) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 33 or 35, or an amino acid sequence having at least 70% sequence identity thereto.

12. The pharmaceutical composition of claim 10, further comprising at least one second therapeutically active agent.

13. The specific binding molecule of claim 1, wherein: VLCDR1 has the sequence set forth in SEQ ID NO: 36; VLCDR2 has the sequence set forth in SEQ ID NO: 2; VLCDR3 has the sequence set forth in SEQ ID NO: 3; VHCDR1 has the sequence set forth in SEQ ID NO: 4; VHCDR2 has the sequence set forth in SEQ ID NO: 5; and VHCDR3 has the sequence set forth in SEQ ID NO: 6.

14. The specific binding molecule of claim 7, wherein the antibody comprises: (i) a light chain comprising the amino acid sequence set forth in SEQ ID NO: 45; and a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 43; or (ii) a light chain comprising the amino acid sequence set forth in SEQ ID NO: 78; and a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 80.

15. An isolated specific binding molecule which binds human Anx-A1, wherein the specific binding molecule is an antibody or a fragment thereof and is humanized and comprises: (i) a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 32 or 34; and (ii) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 33 or 35.

16. The specific binding molecule of claim 15, wherein the specific binding molecule is an antibody comprising: (i) a light chain comprising the amino acid sequence set forth in SEQ ID NO: 40, 42, 75 or 77, or an amino acid sequence with at least 80% sequence identity thereto; and (ii) a heavy chain comprising the amino acid sequence set forth in SEQ ID NO: 41, 43, 55 or 80, or an amino acid sequence with at least 80% sequence identity thereto.

17. A method of treatment for obsessive compulsive disorder (OCD) or an anxiety disorder, comprising administering to a subject in need thereof a specific binding molecule as defined in claim 1.

18. The method of claim 17, wherein the specific binding molecule is an antibody or fragment thereof comprising: (i) a light chain variable region comprising the amino acid sequence set forth in any one of SEQ ID NOs: 48-51, or an amino acid sequence having at least 70% sequence identity thereto; and (ii) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 33 or 35, or an amino acid sequence having at least 70% sequence identity thereto.

Description

FIGURE LEGENDS

(1) FIG. 1 shows the results of an ELISA assay, demonstrating the binding of Mdx001 to Anx-A1. A492 values are proportionate to OPD degradation by the HRP conjugated to the secondary antibody, and thus represent Mdx001 binding.

(2) FIG. 2 shows the results of Biacore analysis of the binding of Mdx001 to Anx-A1. Parts A, B and C each present the results of a separate assay. Assay 1 (part A) shows a K.sub.D of 9.43 nM; Assay 2 (part B) shows a K.sub.D of 9.58 nM; Assay 3 (part C) shows a K.sub.D of 6.46 nM.

(3) FIG. 3 shows the light and heavy chain variable regions of MDX-L1H4 and MDX-L2H2, and the variants that were generated from MDX-L1H4 and MDX-L2H2. Sequences are presented in single letter amino acid code. CDR sequences are in bold. Amino acid substitutions in the variant sequences (relative to MDX-L1H4 and MDX-L2H2) are highlighted.

(4) FIG. 4 shows the results of an ELISA assay, demonstrating the binding of antibodies MDX-L1H4 and its variants (A) and MDX-L2H2 and its variants (B) to Anx-A1. As in FIG. 1, the A492 values are proportionate to OPD degradation by the HRP conjugated to the secondary antibody, and thus represent antibody binding to Anx-A1.

(5) FIG. 5 shows the results of Biacore analysis of the binding of MDX-L1M2H4 and MDX-L2M2H2 to Anx-A1. Parts A-C each present the results of a separate assay for MDX-L1M2H4 binding to Anx-A1; Parts D-F each present the results of a separate assay for MDX-L2M2H2 binding to Anx-A1. For MDX-L1M2H4, Assay 1 (part A) shows a K.sub.D of 3.96 nM; Assay 2 (part B) shows a K.sub.D of 3.94 nM; Assay 3 (part C) shows a K.sub.d of 4.04 nM. For MDX-L2M2H2 Assay 1 (part D) shows a K.sub.D of 4.44 nM; Assay 2 (part E) shows a K.sub.D of 4.37 nM; Assay 3 (part F) shows a K.sub.d of 5.17 nM.

EXAMPLES

Example 1: Sequencing of Anx-A1-Binding Antibody Produced by Hybridoma ECACC 10060301

(6) mRNA was extracted from hybridoma ECACC 10060301. The extracted mRNA was transcribed into cDNA using a reverse transcription protocol. The cDNA was sequenced by standard dye-terminator capillary sequencing by Aldevron (USA), using proprietary primers.

(7) Cycle sequencing was performed using BigDye® Terminator v3.1 Cycle Sequencing kits under a standard protocol provided by Life Technologies®. All data was collected using a 3730xl DNA Analyser system and the Unified Data Collection software provided by Life Technologies® for operation of the 3730xl DNA Analyser and to collect data produced by the 3730xl DNA Analyser.

(8) Sequence assembly was performed using CodonCode Aligner (CodonCode Corporation, USA). Mixed base calls are resolved by automatically assigning the most prevalent base call to the mixed base calls. Prevalence is determined by both frequency of a base call and the individual quality of the base calls.

(9) The sequences of the light and heavy chain variable regions obtained by cDNA sequencing are presented in SEQ ID NOs: 26 and 27, respectively.

(10) The sequences presented in SEQ ID NOs: 26 and 27 were run against a database of known germ lines and a germline for the antibody was identified. This showed that the sequence obtained for the light chain was truncated and missing 5 amino acids at its N-terminus. The complete sequences were reconstructed by Fusion Antibodies (UK) based upon the identified germline sequences and codon-optimised for expression in CHO cells. The codon-optimised variable domains have the sequences presented in SEQ ID NOs: 28 (light chain) and 29 (heavy chain).

Example 2: Production of the Anx-A1-Binding Mdx001 Antibody

(11) The codon-optimised sequences were cloned into the vector pD2610-v13 (ATUM, USA) using standard recombinant techniques and transfected into ExpiCHO cells (Thermo Fisher Scientific, USA). 200 ml of culture was generated. Antibody (Mdx001) was recovered from the cell supernatant using a protein A affinity column and eluted into a phosphate buffer medium.

Example 3: Mdx001 Binds Anx-A1

(12) Mdx001 binding to Anx-A1 was confirmed by ELISA, performed the The Antibody Company (UK) using standard ELISA techniques. ELISA plates were coated with 25 μg/ml Anx-A1 and coating buffer (45 mM Na.sub.2CO.sub.3, pH 9.6 supplemented with 1 mM CaCl.sub.2) overnight at 4° C. (Ca.sup.2+ was found to be required for Mdx001 binding to Anx-A1, and so all binding experiments were carried out in the presence of 1 mM CaCl.sub.2.)

(13) Plates were then blocked for 1 hr at room temperature with blocking buffer (1 mM CaCl.sub.2, 10 mM HEPES, 2% w/v BSA). Primary antibody (Mdx001) was then applied to the plates. The antibody was applied in duplicate in four-fold dilutions made across the plate, starting at a concentration of 1 μg/ml and ending at a concentration of 2.38×10.sup.−7 μg/ml. The antibody was diluted in wash buffer (10 mM HEPES, 150 mM NaCl, 0.05% (v/v) TWEEN-20 and 1 mM CaCl.sub.2) supplemented with 0.1 BSA. The primary antibody was applied to the plate for 1 hr at room temperature, and the plate then washed with wash buffer.

(14) The detection antibody was then applied. For detection a horseradish peroxidase (HRP)-conjugated goat anti-mouse antibody (Sigma-Aldrich, A2554) was used at a dilution of 1:1000. This was applied to the ELISA plate for 1 hour at room temperature. The ELISA plate was then washed again with wash buffer.

(15) The colourimetric substrate OPD (o-phenylenediamine dihydrochloride, Sigma-Aldrich P4664) was then applied to the plate. OPD solution was made up according to the manufacturer's instructions to yield a 0.4 mg/ml OPD solution in phosphate-citrate buffer, pH 5. 40 μl of 30% H.sub.2O.sub.2 was added per 100 ml OPD solution immediately prior to use. 100 μl of the resultant OPD solution was then added to each well of the plate

(16) The plate was incubated for 20 mins in the dark at room temperature, after which 50 μl of 3M H.sub.2SO.sub.4 was added to stop the reaction. Immediately after addition of H.sub.2SO.sub.4 the absorbances of the plate were read at 492 nm (absorbance at 492 nm is abbreviated A492). Mdx001 was found to bind well to Anx-A1. The results are shown in Table 1 and FIG. 1. Mdx001 was found to bind across the plate except at the very highest dilutions. Even at the highest dilution of the anti-Anx-A1 antibody some binding is evident with a difference in absorbance seen between the blank values and the highest antibody dilution values. (Blank wells were coated with 25 μg/ml Anx-A1 in coating buffer and treated identically to experiment wells except no primary antibody was added.) The binding falls rapidly below 0.0625 μg/ml primary antibody (Mdx001). Binding appears to plateau at primary antibody concentrations below 3.9×10.sup.−3 μg/ml.

(17) TABLE-US-00002 Mdx001 Concentration (μg/ml) A492 1 1.772 0.25 1.66765 0.0625 1.59375 0.015625 0.8651 0.003906 0.31225 0.000977 0.20415 0.000244 0.17125  6.1 × 10.sup.−5 0.14415 1.53 × 10.sup.−5 0.16205 3.81 × 10.sup.−6 0.1608 9.54 × 10.sup.−7 0.1316 2.38 × 10.sup.−7 0.1345

Example 4: Biacore Analysis of Mdx001 Binding to Annexin-A1

(18) Biacore analysis was performed at the NMI, University of Tübingen, Germany. Standard Biacore procedures were used to analyse purified Mdx001 expressed in CHO cells as described above.

(19) The running buffer used was follows: HEPES 10 mM, NaCl 150 mM, CaCl.sub.2 1 mM, Tween 20 0.05% v/v, pH 7.4. The buffer was filtered using a 0.22 μM filter and de-gassed by sonication for 15 mins.

(20) The Mdx001 antibody was immobilised on a chip via a goat anti-mouse IgG. Ligand (Anx-A1) was passed over the immobilised antibody in running buffer. Anx-A1 was used at a concentration of 5, 10, 20, 40 or 80 nM. In each experiment, a total of 150 μl Anx-A1-containing running buffer was passed over the antibody, at a flow rate of 30 μl/min.

(21) Regeneration was performed using a regeneration buffer, 10 mM glycine-HCl, pH 2. To regenerate the chip, 70 μl regeneration buffer was passed over the chip at a rate of 10 μl/min.

(22) Experiments were performed in triplicate. Results of each of the 3 experiments are shown in FIG. 2. The three experiments gave K.sub.D values for the binding of Mdx001 to Anx-A1 of 9.43 nM, 9.58 nM and 6.46 nM, an average of 8.49 nM.

Example 5—Humanisation of Mdx001

(23) Mdx001 was humanised using standard CDR grafting techniques coupled with antibody structure and database analysis of known human framework region sequences. All framework region sequences used were derived from mature IgG isolated from humans and so are expected to be non-immunogenic and retain the canonical structure of the CDR loops.

(24) The humanisation process yielded two antibodies, MDX-L1H4 and MDX-L2H2, with sequences as set forth hereinbefore.

Example 6—Binding of Mdx002 to Anx-A1

(25) The humanised antibodies were analysed to identify sites of possible post-translational modification, using standard bioinformatic tools. De-amidation is a major degradation pathway in antibodies so the humanised sequences were checked for the deamidation motifs Ser-Asn-Gly, Glu-Asn-Asn, Leu-Asn-Gly and Leu-Asn-Asn. A deamidation motif with the sequence Ser-Asn-Gly was identified in VLCDR1 of MDX-L1H4/MDX-L2H2. Modifications were made to MDX-L1H4 and MDX-L2H2 to remove this sequence motif. Humanised antibodies comprising a modified VLCDR1 sequence were generated in order to identify a functional modified VLCDR1 sequence.

(26) Three variants were generated for each humanised antibody. FIG. 3 shows the light and heavy chain variable regions for the variants that were generated from MDX-L1H4 and MDX-L2H2.

(27) The first of the modified antibodies, variant 1 (i.e. MDX-L1M2H4) comprised a VLCDR1 with the amino acid sequence set forth in SEQ ID NO: 36, i.e. a substitution of the glycine residue at position 11 for an alanine residue. The second of the modified antibodies, variant 2 (i.e. MDX-L1M3H4) comprised a VLCDR1 with the amino acid sequence set forth in SEQ ID NO: 37, i.e. a substitution of the serine residue at position 9 for a threonine residue. The third of the modified antibodies, variant 3 (annotated as LC1(mod 1)HC4) comprised a modified Mdx001 VLCDR1 sequence, in which the asparagine residue at position 10 was substituted for an aspartic acid residue. The LC1(mod 1)HC4 VLCDR1 sequence is set forth in SEQ ID NO: 56.

(28) The variants from MDX-L2H2 are related to those from MDX-L1H4 insofar as they contain the same modifications in VLCDR1. They are referred to as variant 1 (MDX-L2M2H2), variant 2 (MDX-L2M3H2) and variant 3 (LC2(mod 1)HC2).

(29) Both antibodies contained humanised variable and constant domains as described in Example 5. The CDR sequences were otherwise unaltered relative to MDX-L1H4 and MDX-L2H2, and with the exception of the above-described VLCDR1 modifications were identical to their parent sequences.

(30) Binding of the modified humanised antibodies to Anx-A1 was initially tested by ELISA, using the same method as described in Example 3. The results of this ELISA are presented in FIG. 4. As controls, MDX-L1H4 and MDX-L2H2 were used. As can be seen in FIG. 4A, MDX-L1M2H4 (LC1(mod 2)HC4) and MDX-L1M3H4 (LC1(mod 3)HC4) (and MDX-L2M2H2 (LC2(mod 2)HC2) and MDX-L2M3H2 (LC2(mod 3)HC2), FIG. 4B) bound Anx-A1 comparably to MDX-L1H4 (or MDX-L2H2), but binding of LC1(mod 1)HC4 (and LC2(mod 1)HC2) to Anx-A1 was significantly weaker than for the control. This demonstrated that substitution of asparagine 10 in Mdx001 VLCDR1 for aspartic acid negatively impacted on binding of the antibody to Anx-A1. However, substitution of glycine 11 for alanine or threonine 9 for serine in the same CDR sequence did not negatively impact binding.

(31) In light of the ELISA results, the LC1(mod 1)HC4 and LC2(mod 1)HC2 antibodies were discarded. The LC1(mod 2)HC4 and LC2(mod 2)HC2 antibodies, which demonstrated the best binding of the antibodies with modified VLCDR1 sequences, were taken forward for further analysis. Binding of LC1(mod 2)HC4 and LC2(mod 2)HC2 to Anx-A1 was quantified by Biacore, using the same method as described in Example 4. As in Example 4, the Biacore experiments were performed in triplicate. The results of each of the three experiments are presented in FIG. 5. As shown, for LC1(mod 2)HC4 the three experiments gave K.sub.D values of 3.96 nM, 3.94 nM and 4.04 nM, an average of 3.98 nM. This demonstrated that LC1(mod 2)HC4, which has a K.sub.D of 3.98 nM for Anx-A1 binding, binds Anx-A1 with significantly higher affinity than does Mdx001, which has a K.sub.D of 8.49 nM for Anx-A1 binding. LC1(mod 2)HC4 was given the name MDX-L1M2H4. For LC2(mod 2)HC2 the three experiments gave K.sub.D values of 4.44 nM, 4.37 nM and 5.17 nM, an average of 4.66 nM. This demonstrated that LC2(mod 2)HC2, which has a K.sub.D of 4.66 nM for Anx-A1 binding, also binds Anx-A1 with significantly higher affinity than does Mdx001, which has a K.sub.D of 8.49 nM for Anx-A1 binding. LC2(mod 2)HC2 was given the name MDX-L2M2H2.