Human antibody and fragments thereof for use in the treatment of gastric cancer (GC) and other types of tumours expressing the MICA protein (MHC class I chain-related protein A gene)

Abstract

A bivalent, anti-MICA human monoclonal antibody formed by two or more heavy and light chains with a variable immunoglobulin domain neutralises the MICA protein in its soluble state and opsonises tumour cells expressing the antigen, stimulating adaptive immunity in the treatment of gastric cancer or other types of cancer in which the tumour cells express MICA in the soluble form or abundantly on their surface.

Claims

1. An isolated monoclonal antibody or antigen binding fragment thereof that specifically binds to the α1 subunit of MICA, SEQ ID NO: 1, wherein said antibody or antigen binding fragment comprises a heavy chain variable domain of SEQ ID NO: 2 and a light chain variable domain SEQ ID NO: 3.

2. The antibody or antigen binding fragment according to claim 1, wherein said antigen binding fragment comprises a scFv formed by a heavy chain variable domain of SEQ ID NO: 2 and a light chain variable domain of SEQ ID NO:3 contained in a single peptide chain linked via a peptide linker selected from SEQ ID NO: 12 and SEQ ID NO: 13.

3. The antibody or antigen binding fragment according to claim 2, wherein said antigen binding fragment comprises CDR-L1 of SEQ ID NO: 6; CDR-L2 of SEQ ID NO: 7; CDR-L3 of SEQ ID NO: 8; CDR-H1 of SEQ ID NO: 9; CDR-H2 of SEQ ID NO: 10; and CDR-H3 of SEQ ID NO: 11.

4. The antibody or antigen binding fragment according to claim 2, wherein said scFv fragment comprises the sequence SEQ ID NO:14.

5. The antibody or antigen binding fragment according to claim 1, wherein said antigen binding fragment is a scFv-Fc type formed by a heavy chain variable domain and a light chain variable domain contained in a single peptide chain linked through a linker peptide selected from SEQ ID NO: 12 and SEQ ID NO: 13 and is fused with the CH2 and CH3 domain of human immunoglobulin gamma-1 consisting of SEQ ID NO: 15.

6. The antibody or antigen binding fragment according to claim 5, wherein said antigen binding fragment comprises CDR-L1 of SEQ ID NO: 6; CDR-L2 of SEQ ID NO: 7; CDR-L3 of SEQ ID NO: 8; CDR-H1 of SEQ ID NO: 9; CDR-H2 of SEQ ID NO: 10; and CDR-H3 of SEQ ID NO: 11.

7. The antibody or antigen binding fragment according to claim 1, wherein said antigen binding fragment comprises CDR-L1 of SEQ ID NO: 6; CDR-L2 of SEQ ID NO: 7; CDR-L3 of SEQ ID NO: 8; CDR-H1 of SEQ ID NO: 9; CDR-H2 of SEQ ID NO: 10; and CDR-H3 of SEQ ID NO: 11.

8. The antibody or antigen binding fragment according to claim 1, wherein said antigen binding fragment comprises heavy chain domain of human immunoglobulin gamma-1 consisting of SEQ ID NO: 17 and light chain domain of human immunoglobulin kappa consisting of SEQ ID NO: 16.

9. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 1, and a pharmaceutically acceptable carrier.

10. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 2, and a pharmaceutically acceptable carrier.

11. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 3, and a pharmaceutically acceptable carrier.

12. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 5, and a pharmaceutically acceptable carrier.

13. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 7 and a pharmaceutically acceptable carrier.

14. A pharmaceutical composition wherein said composition comprises the antibody or antigen binding fragment described in claim 6, and a pharmaceutically acceptable carrier.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1. Detection of scFv binding by native MICA. AGS (A) and MKN-45 (B) gastric cancer lines were incubated with a commercial anti-MICA antibody or anti-MICA scFv (2 h, 37° C.). After rinsing, the cells were incubated with mouse anti-IgG conjugated to FITC, those treated with commercial antibody, and anti-HA epitope murine antibody and then mouse anti-IgG antibody conjugated to FITC, those treated with commercial antibody, and anti-HA epitope murine antibody, and then mouse anti-IgG antibody conjugated to FITC. The binding of the anti-MICA scFv to native MICA was analysed by flow cytometry, using the AGS gastric cancer lines (upper panel) and MKN gastric cancer lines (lower panel). In the left panels, the population used for the analysis is shown and, on the right side, the FITC signal of unlabeled cells (light grey, solid line), cells labelled with a commercial anti-MICA antibody conjugated to FITC (grey, dotted line), and the scFv signal, detected with an anti-HA antibody conjugated to FITC (black, solid line), are shown. The average fluorescence in the FL1-H channel is shown in the table below.

(2) FIG. 2. Fluorescence emission in NOD-SCID-IL-null (or NSG) mice with MKN-45 cell tumour. Mice inoculated subcutaneously with 1×100 MKN-45 gastric adenocarcinoma cells, once the tumour was established on day 60, were intravenously (lateral tail vein) injected with 50 pg of scFv-MICA conjugated to DyLight 650. As a negative control of tumour binding, 50 pg bovine serum albumin (BSA) conjugated to DyLight 650 was used. The mice were intramuscularly anesthetised with a ketamine:xylazine mixture in a 100:10 v:v ratio, once anesthetised, the fluorescence emission in the tumour area was analysed using LUMINA II IN/IS (In Vivo Imaging Systems). (A) scFv-MICA-DyLight 650 fluorescence emission, 0, 5 and 20 minutes approximately post-inoculation. (B) BSA-DyLight 650 fluorescence emission, approximately 20 minutes post-inoculation. (C) fluorescence emission difference between conjugated BSA and conjugated scFv-MICA approximately 20 minutes post-inoculation. In Figure A, the thoracic region shows two signals: an upper one, which corresponds to the heart and a lower one, which corresponds to the tumour. Both signals are seen at 5 and 20 min post-inoculation; however, it is seen that the signal from the heart is lower at 20 minutes, while the signal from the tumour is higher, indicating that there is accumulation of scFv-MICA-DyLight 650 in the latter.

DETAILED DESCRIPTION OF THE INVENTION

(3) The invention corresponds to fully human bivalent monoclonal antibodies which specifically bind to the MICA a1 subunit, SEQ ID NO: 1, which comprise a heavy-chain variable domain SEQ ID NO: 2 and comprise a light-chain variable domain SEQ ID NO: 3.

(4) In a specific preferred embodiment, the antibodies are bivalent of a scFv-Fc-type chain (wherein scFv is an antibody fragment consisting of a heavy-chain variable domain and a light-chain variable domain contained in a heavy chain, and a light-chain variable domain contained in a single peptide chain linked via a chorus peptide linker (SEQ ID NO: 12, Linker 1 and SEQ ID NO: 13 Linker 2) formed by the fusion of anti-MICA scFv (SEQ ID NO: 14) with the CH2 and CH3 domain of human gamma 1 immunoglobulin (SEQ ID NO: 15) and/or bivalent two-Fab-type-chain antibodies, i.e. the separate peptide chains linked by non-covalent interactions, wherein the light chain contains the human Kappa immunoglobulin CH1 domain (SEQ ID NO: 16) and the heavy chain contains the CH1, CH2 and CH3 domains of human gamma 1 immunoglobulin (SEQ ID NO: 17).

(5) In a preferred embodiment of the invention it comprises nucleic acid sequences, which code for the heavy-chain variable domain (SEQ ID NO: 4) and the light-chain variable domain (SEQ ID NO: 5).

(6) In a more specific embodiment, the antibody or fragments thereof according to the present invention comprises CDR complementarity determining regions in accordance with the following definitions: CDR-L1: described in SEQ ID NO: 6; CDR-L2: described in SEQ ID NO: 7; CDR-L3: described in SEQ ID NO: 8; CDR-H1: described in SEQ ID NO: 9; CDR-H2: described in SEQ ID NO: 10; CDR-H3: described in SEQ ID NO: 11.

(7) In a preferred embodiment, the invention relates to pharmaceutical compositions comprising monoclonal antibodies and a pharmaceutical{circumflex over ( )} acceptable carrier.

(8) In a preferred embodiment, the invention relates to a method for detecting the soluble MICA factor in a sample, wherein the method comprises the following steps: (a) taking the sample from a patient, from blood or tumour cell culture supernatant; (b) placing the sample in contact with 0.3 micrograms of developed monoclonal antibody, fixed to a microtitration plate; (c) incubating the sample with a secondary commercial anti-MICA antibody conjugated to peroxidase; and (d) quantifying the concentration of the soluble factor by means of a colorimetric signal, after adding the enzyme substrate, in a spectrophotometer.

(9) In another preferred embodiment, the invention relates to a kit which comprises a pharmaceutical composition made with the antibody of the invention, which is stored in a pharmaceutical{circumflex over ( )} acceptable container.

(10) In another preferred embodiment, the invention relates to a method for therapeutic treatment against cancers selected from the following: hepatocellular carcinoma, melanoma, kidney.

(11) In a more particular preferred embodiment, the cancer is gastric cancer.

EXAMPLES

Example 1: Choice of the Epitope or Segment of the MICA Molecule to which the Recombinant Antibody Will Bind

(12) In order to select an epitope, and not leave to chance the site of the molecule against which the antibody would bind, a Multiple Antigen Peptide System (MAPS) was designed, which consists of a nucleus of lysines and 8 arms of the same peptide. For the design of MAPS, the polymorphisms of the MICA protein, its tertiary structure and its binding site with the NKG2D receptor, obtained from the structure of the NKG2D-MICA complex (PDB: 1HYR) were analysed. The least polymorphic MICA alpha helix segment was chosen and the RDLTGNGKDLRMTLAHIKDQ (MICA Alphal) peptide was generated (SEQ ID NO: 1). Subsequently, a massive sequencing analysis of 50 samples from patients with gastric cancer was performed in the laboratory and no variants were found in this segment. Therefore, this segment was selected for the later stages due to its low polymorphic variability.

Example 2: Selection and Characterisation of a Viral Particle that Expresses Anti-MICA scFv (scFv-aMICA-Phage)

(13) Selection or panning of the viral particle carrying the gene that codes for scFv-aMICA (scFv-aMICA-phage) was performed with the MAPS peptide from a library presenting scFvs on the surface of M13 phage, previously constructed in the laboratory from healthy donors as described in Sotelo et al. (An efficient method for variable region assembly in the construction of scFv phage display libraries using independent strand amplification. mAbs 4, 542-550, doi:10.4161/mabs.20653 (2012)). After three cycles of panning with the peptide, 70 clones were randomly selected and an ELISA assay was performed, with recombinant MICA-sensitised plates, wherein the peptide and two different proteins blocked the remaining sites of the plate. The clones with the highest signal were selected and their reactivity against recombinant MICA (rMICA) and native MICA was analysed by ELISA and flow cytometry, respectively. The five clones with the highest reactivity against MICA were selected, amplified, and the segments coding for scFv were sequenced. Of these, three had the same sequence and a fourth one had a minor modification, for which reason one of these clones was selected to continue.

Example 3: Generation, Production and Characterisation of the Anti-MICA scFv Protein (scFv-aMICA)

(14) To produce the scFv-aMICA protein, the gene coding for it was subcloned into a variant of the pUCH1 phagemid, called pUCH1 Amber, both constructed in the laboratory. This variant allows independent production of M13 phage capsid components and therefore the purification thereof. It was observed that the molecule maintains its ability to bind to a rMICA and native protein, respectively, by ELISA assay and in vitro culture cytometry of commercial gastric adenocarcinoma cell lines (FIG. 1). Likewise, the anti-MICA scFv, a monomeric molecule, has an affinity constant (KD) of 135±45 nM, Kaff of 7.4±2.5×10® M″.sup.1. FIG. 3 shows a diagram of the scFv-aMICA model with MICA proposed and subsequently validated in vitro using an animal model of immunosuppressed mice NOD-SCID IL-2Rgnull or NSG 45 (NOD.Cg-Prkdcscid Il2rgtml Wjl/SzJ) (Jackson Laboratories, USA) xenografted with human adenocarcinoma cells (FIG. 2). MICA is a molecule that is absent in mice, thus ensuring the absence of a cross-reaction with molecules from the animal. In this model, it was verified by immunohistochemistry that the human tumour cells, in addition to growing in this murine model, express MICA. Subsequently, fluorescent scFv-aMICA was inoculated (FIG. 2) and it was observed significantly that the scFv-aMICA antibody fragment was located in the tumour unlike fluorescent BSA, its negative control.