Monoclonal antibodies binding to the CD160 transmembrane isoform

Abstract

The present invention relates to monoclonal antibodies that bind to the CD160-TM isoform. The inventors developed new monoclonal antibodies which bind to the CD160-TM isoform but dot not bind to the CD160 GPI-anchored isoform not to the CD160 soluble isoform. In particular, the antibodies of the present invention are suitable for amplifying NK cell activation and therefore cytotoxic functions NK cells.

Claims

1. A monoclonal antibody or an antigen binding fragment thereof, wherein said antibody or antigen binding fragment thereof comprises a light chain and a heavy chain, wherein the light chain comprises the following CDRs: i) VL-CDR1 as set forth in SEQ ID NO: 6 wherein X.sub.11 is Y or S and X.sub.12 is G or Y; ii) VL-CDR2 as set forth in SEQ ID NO: 7; and iii) VL-CDR3 as set forth in SEQ ID NO: 8 wherein X.sub.3 is S or Y, and wherein the heavy chain comprises the following CDRs: i) VH-CDR1 as set forth in SEQ ID NO: 9 wherein X.sub.3 is S or Y; ii) VH-CDR2 as set forth in SEQ ID NO: 10 wherein X.sub.1 is Y or G and X.sub.10 is N or S; and iii) the VH-CDR3 as set forth in SEQ ID NO: 11.

2. The monoclonal antibody or the antigen binding fragment thereof of claim 1, wherein said antibody or antigen binding fragment thereof comprises a light chain and a heavy chain, wherein the light chain comprises the following CDRs: i) VL-CDR1: AGTSSDVGGYYGVS (SEQ ID NO: 20); ii) VL-CDR2: YDSYRPS (SEQ ID NO: 7); and iii) VL-CDR3: SSSTYYSTRV (SEQ ID NO: 24), and wherein the heavy chain comprises the following CDRs: i) VH-CDR1: NYSMN (SEQ ID NO: 26); ii) VH-CDR2: YIYGSSRYISYADFVKG (SEQ ID NO: 29); and iii) VH-CDR3: GMDV (SEQ ID NO: 11).

3. The monoclonal antibody or the antigen binding fragment thereof of claim 1, wherein said antibody or antigen binding fragment thereof comprises a light chain and a heavy chain, wherein the light chain comprises the following CDRs: i) VL-CDR1: AGTSSDVGGYSYVS (SEQ ID NO: 23); ii) VL-CDR2: YDSYRPS (SEQ ID NO: 7); and iii) VL-CDR3: SSYTYYSTRV (SEQ ID NO: 25), and wherein the heavy chain comprises_the following CDRs: i) VH-CDR1: NYYMN (SEQ ID NO: 27); ii) VH-CDR2: GIYGSSRYINYADFVKG (SEQ ID NO: 30); and iii) VH-CDR3: GMDV (SEQ ID NO: 11).

4. The monoclonal antibody or the antigen binding fragment thereof of claim 1 comprising a heavy chain and a light chain, wherein the heavy chain has at least 95% of identity with SEQ ID NO: 12 or SEQ ID NO: 14 and wherein the light chain has at least 95% of identity with SEQ ID NO: 13 or SEQ ID NO: 15.

5. The monoclonal antibody or the antigen binding fragment of claim 1 comprising a heavy chain and a light chain wherein the heavy chain is identical to SEQ ID NO: 12 or SEQ ID NO: 14 and wherein the light chain is identical to SEQ ID NO: 13 or SEQ ID NO: 15.

6. A conjugated antibody comprising the monoclonal antibody or the antigen binding fragment thereof of claim 1 and a cytotoxic moiety.

7. A nucleic acid molecule which encodes a heavy chain or a light chain of the antibody or the antigen binding fragment thereof of claim 1.

8. The nucleic acid molecule of claim 7 which comprises a nucleic acid sequence identical to SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18 or SEQ ID NO: 19.

9. A method of activating natural killer (NK) cells in a human subject, comprising administering to the subject the monoclonal antibody or the antigen binding fragment thereof of claim 1, effective to activate NK cells.

10. The method of claim 9, wherein the subject has a cancer, and wherein the cancer is a gastrointestinal cancer, a skin cancer, a lung cancer, a stomach cancer, a colon cancer, a head cancer, a neck cancer, a kidney cancer, a liver cancer, or an esophagus cancer.

11. A fusion protein comprising the monoclonal antibody or the antigen binding fragment thereof of claim 1.

12. The monoclonal antibody or antigen binding fragment thereof of claim 1, which binds to the extracellular domain of the CD160-TM isoform, wherein said antibody or antigen binding fragment thereof does not bind to the GPI-anchored isoform nor to the CD160 soluble isoform, wherein the epitope of said monoclonal antibody or antigen binding fragment thereof comprises at least one amino acid residue from amino acid residues 175 to 189 of SEQ ID NO: 1, and wherein said epitope further comprises at least one amino acid residue from amino acid residues 62 to 85 of SEQ ID NO: 1.

13. The monoclonal antibody or antigen binding fragment thereof of claim 1, wherein the antibody or antigen binding fragment is chimeric, humanized or human.

14. The method of claim 9, wherein the subject has a disease or disorder selected from the group consisting of a cancer wherein the cancer cells do not express CD160-TM, an infectious disease, and an inflammatory disease.

Description

FIGURES

(1) FIG. 1: Binding specificity of A12 and B6 antibodies on CD160-GPI vs CD160TM expressing cells. CHO or HEK cells forced to express CD160-GPI or CD160TM isoform, respectively, were labelled with the CL1-R2 monoclonal antibody or BY55 antibody (both specific for CD160-GPI) or A12 or B6 antibodies (white histograms). Mouse or human isotype control Igs were used as negative controls (black histograms). Bound antibodies were revealed with the appropriate PE-conjugated secondary reagents.

(2) FIG. 2: Binding specificity of A12 on IL2-treated human PBMC. PBMC were either left untreated (Day 0) or incubated with IL2. Immuno-labelling were performed at the indicated time points with either an isotypic contral IgG (black histogram) or A12 antibody (grey histogram) plus PE-coupled goat anti-human IgG antibodies. Lymphocytes subsets were further identified by addition of a mix of CD8-FITC, CD56-PCS, CD3-APC and CD4-PC7 mAbs. Shown are the labellings obtained on each gated lymphocyte population.

(3) FIG. 3: Assessment of A12 specificity by immunoprecipitation. Post-nuclear lysates were prepared from HEK-CD160TM cells and subjected to immunoprecipitation with either mouse (mu IgG) or human (hu Ig) isotypic control IgG, chimeric murine A12 (mu A12) or fully human A12 (hu A12) antibody. Immuno-precipitated proteins were separated by SDS-10% PAGE under non-reducing conditions, transferred on nitrocellulose and revealed by Western blot using an anti-Flag mAb. Arrows indicate CD160TM-related signals.

(4) FIG. 4: sCD160 is recognized by the antibodies CL1-R2, R&D 6700 and RB312 but not A12. 1 ug per well of antibody was coated overnight on a 96 well maxisorb plate. After saturation with PBS-5% BSA, 10ng of recombinant soluble CD160-His Tag were added to each well and incubated 2h at room temperature. After washing, revelation was done with anti-His-HRP then TMB substrate. Experimental conditions were performed in triplicate and results displayed for each antibody were obtained after subtraction of the respective OD Ig control (OD=OD capture antibody−OD respective Ig control).

(5) FIG. 5 shows the alignments of VH and VL sequences of A12 and B6 antibodies.

(6) FIG. 6 shows the conformation epitope recognized A12 and B6 antibodies composes of 2 peptides. The peptides are indicated in BOLD and UNDERLINED. The different domain of the CD160-TM isoform are also represented.

(7) FIGS. 7A-7B: A12 induces NK cells degranulation and activation. The NK92 cell line was pre-incubated with isotype control muIgG or a chimeric Fc murine version of the human A12 antibody (muA12) plus rabbit anti-mouse IgG antibodies. Effector cells were then incubated in the presence of the NK sensible target cells (K562 cell line) at the indicated E/T ratio. The NK cell line NK92 cell degranulation and activation was monitored by detection of membrane associated CD107a (FIG. 7A) and CD137 (FIG. 7B), respectively. Given are the percentages of positive cells pre-treated with mulgG control (circles) or muA12 (squares).

EXAMPLE 1

Characterization of A12 Antibody

(8) Material & Methods

(9) Cells

(10) CHO or HEK cells were transfected with a eukaryotic expression vector encompassing CD160-GPI or Flag-tagged CD160-TM cDNA, respectively. Stable transfectants were obtained by selection with the appropriate antibiotic and named thereafter CHO-CD160-GPI and HEK-CD160TM. Proper expression of CD160 isoforms was assessed by flow cytometry using the anti-CD160-GPI specific mAb CL1R2 or anti-Flag mAb plus PE-coupled goat anti-mouse IgG.

(11) PBMC were obtained by gradient density from healthy volunteers venous blood. Activation was achieved by addition of recombinant human IL2 (100 U/ml).

(12) Selection of A12 fully human antibody

(13) Fully human aglycosylated anti-CD160TM antibodies were selected by phage display on HEK-CD160TM cells. Among the obtained antibodies, A12 was identified as the one giving the best recognition profile on HEK-CD160TM cells by flow cytometry. A murine chimeric version of A12 antibody, where the human IgG1 Fc portion was exchanged for a murine IgG2a Fc fragment, was also generated.

(14) Flow Cytometry

(15) CHO and HEK transfected cells were labelled with the anti-CD160-GPI mAb CL1-R2 or BY55, the fully human A12 or B6 antibody or their corresponding isotypic control IgG. Bound antibodies were further revealed by addition of PE-coupled goat anti-mouse or anti-human IgG. Cell acquisition was performed on a CytoFlex cytometer and results were analysed with FlowJo software.

(16) For PBMC, CD160 labelling was performed as above. Following washes and addition of normal mouse serum, cells were incubated with a mix of CD8-FITC, CD56-PCS, CD3-APC and CD4-PC7 mAbs. After cell acquisition, analyses were performed to distinguish the CD3.sup.+CD4.sup.+ and CD3.sup.+CD8.sup.+ T lymphocytes, and the CD3.sup.−CD56.sup.+ NK cells within the lymphocytes population.

(17) Immuno-Precipitation and Western Blot

(18) HEK-CD160TM cells, that express a Flag-tagged version of CD160TM isoform, were lysed in 1% NP40 lysis buffer. Post-nuclear lysates were prepared and subjected to immuno-precipitation with the fully human A12 antibody or mouse chimeric A12. Human and mouse IgG were used as negative controls, respectively. Immune complexes were further collected with protein G Sepharose beads. Following washes, non-reducing sample buffer (devoid of reducing agent) was added and samples were finally heat-denatured. Proteins were separated by SDS-10% PAGE, electrically transferred on a nitrocellulose membrane and subjected to immuno-blotting with and anti-Flag mAb plus HRP-coupled goat anti-mouse IgGs. Revelation was performed by enhanced chemiluminescence and images acquired with an ImageQuant LAS device.

(19) Results

(20) After selection on HEK-CD160TM cells, A12 and B6 specificity for CD160TM isoform was first verified by flow cytometry on both CD160-GPI and CD160TM expressing transfectants. As shown on FIG. 1, a positive labelling was obtained with A12 and B6 antibodies on HEK-CD160TM cells but not on CHO-CD160-GPI cells. Conversely, the anti-CD160-GPI mAb CL1-R2 or BY55 gave a positive labelling only on CD160-GPI-expressing cells, ruling out the possibility that A12 and B6 negativity was linked to the non-expression of CD160-GPI on CHO transfectants.

(21) To further confirm A12 specificity for CD160TM isoform, immuno-labelling were performed on human PBMC. Because CD160TM main feature is its unique expression on NK cells when activated, flow cytometry analyses were performed on untreated or IL2-activated cells. The corresponding results showed no recognition of the CD4.sup.+ and CD8.sup.+ T cells by A12 even at the latest activation points (FIG. 2). In contrast a positive signal was detected on part of the NK cell population 2 days after the beginning of the activation that remained visible up to 15 days. Thus A12 antibody fulfilled the characteristics for being a specific CD160TM antibody.

(22) The ability of A12 antibody to recognize CD160TM was additionally tested by performing immuno-precipitation experiments. To this aim HEK-CD160TM cells, that expressed a Flag-tagged CD160TM receptor, were lysed and subjected to immuno-precipitation with either the fully human A12 antibody or its chimeric murine counterpart. Human or mouse IgG were used as negative controls. Immune complexes were separated by gel electrophoresis under non-reducing conditions to allow detection of CD160TM according to its multimerization state. Proteins revelation by Western blot with an anti-Flag mAb showed no specific signal in the immuno-precipitate performed with the fully human A12 when compared to control human IgG, suggesting that the antibody was unable to recognize CD160TM when partially denatured (FIG. 3). In contrast, the use of the murine chimeric A12 antibody led to the detection of proteins bands with an apparent molecular weight of 34-38, 56 and 100 kDa, that most likely correspond to the mono-, di-, and quadrimeric form of the receptor, respectively. Finally we show that A12 does not recognize the CD160 soluble isoform contrary to the antibody of the closest prior art, namely the RB312 described in Giustiniani J. et al. (Curr Mol Med. 2012 February; 12(2):188-98, see FIG. 4). Same results were obtained with the B6 antibody (data not shown).

EXAMPLE 2

Characterization Of B6 Antibody

(23) B6 antibody also results from the phage display selection as described in EXAMPLE 1. B6 was identified as giving a very good recognition profile on HEK-CD160TM cells by flow cytometry. A murine chimeric version of B6 antibody, where the human IgG1 Fc portion was exchanged for a murine IgG2a Fc fragment, was also generated. FIG. 5 shows the alignment of the VH and VL sequences of A12 and B6 antibodies and we can conclude that B6 is very similar to A12.

EXAMPLE 3

Characterization of the Epitope Recognized by A12 AND B6

(24) Epitope mapping of A12 and B6 was performed according to published protocols (Sloostra et al, Mol. Divers. (1996), Timmerman et al, J. Mol. Recognit. 20 5J:283-299 (2007)). Briefly, the binding of antibody to each peptide was tested in a PEPSCAN-based ELISA. Surprisingly, we found that the epitope recognized by A12 an B6 antibodies is a conformational epitope composed of 2 peptides: SEQ ID NO: 4 and SEQ ID NO: 5. Despite the fact that the first peptide is commonly shared by CD160-GPI and CD160TM, the second peptide is specific to CD160TM explaining the specificity of both B6 and A12 antibodies for the CD160-TM isoform.

EXAMPLE 4

NK Cell Activation and CD107a Analysis Methods

(25) The blood derived human chronic myelogenous leukemia cell line K562 (target cells) and the NK cell lymphoma derived NK92 cell line (Effector cells) growth in complete RPMI 1640 (10% FCS, 2% glutamine, 1% antibiotics) and for NK92 cell line supplemented with IL-2 (200 UI/m1).

(26) Effector cells were incubated 30 min with isotype control muIgG or chimeric A12 (muA12) diluted at 20 μg/m1 and rabbit anti-mouse IgG (3 μ/test) before co-culture with target cells at different ratio (E/T: 10/1, 5/1, 2.5/1, 1/1). After 5 h of co-culture, cells were washed with PBS then stained with CD3-FITC, CD137-PE, CD107-APC and CD56-PC7. CD137 and CD107a expressions were analyzed on gated CD3.sup.− CD56.sup.+ cells

(27) Results:

(28) Engagement of CD160-TM with an antibody of the present invention (muA12 antibody) enhances the expression of CD137 and the cell cytotoxicity (expression of CD107a) against K652 cells (FIGS. 7A-7B). Best results are obtained at low E/T ratio (1/1 and 2.5/1). Similar results were obtained with muB6 antibody (data not shown).