NOVEL HUMAN THERAPEUTIC MONOCLONAL ANTIBODIES AND USES THEREOF

Abstract

New human therapeutic monoclonal antibodies, which are directed against neutrophil proteinase 3. The monoclonal antibodies are specifically directed against a conformational epitope of the neutrophil proteinase 3 and are capable of inhibiting by at least 30% the production for reactive oxygen derivatives by neutrophils. Also, a pharmaceutical composition including as an active substance at least the monoclonal antibody, and a method for early treatment and/or prevention of relapse of granulomatosis with polyangiitis, which includes the administration of the pharmaceutical composition.

Claims

1.-13. (canceled)

14. A monoclonal antibody directed against neutrophil proteinase 3 represented by the sequence SEQ ID NO: 1, said monoclonal antibody: being specifically directed against a conformational epitope of said neutrophil proteinase 3; and being capable of inhibiting by at least 30% the production of reactive oxygen derivatives by neutrophils, said production of reactive oxygen derivatives being induced by the presence of autoantibodies directed against said neutrophil proteinase 3.

15. The monoclonal antibody according to claim 14, comprising: a heavy chain comprising from its N-terminal to its end C-terminal: a CDR1 having at least 80% identity with the sequence SEQ ID NO: 15; a CDR2 having at least 80% identity with the sequence SEQ ID NO: 17; and a CDR3 having at least 80% identity with the sequence SEQ ID NO: 19; and a light chain comprising from its N-terminal to its end C-terminal: a CDR1 having at least 80% identity with the sequence SEQ ID NO: 31; a CDR2 having at least 80% identity with the sequence SEQ ID NO: 33; and a CDR3 having at least 80% identity with the sequence SEQ ID NO: 35.

16. The monoclonal antibody according to claim 14, comprising: a heavy chain comprising a variable region having at least 80% identity with the sequence SEQ ID NO: 7, with the proviso that said variable region of the heavy chain comprises from its N-terminal to its end C-terminal: the CDR1 of sequence SEQ ID NO: 15; the CDR2 of sequence SEQ ID NO: 17; and the CDR3 of sequence SEQ ID NO: 19; and a light chain comprising a variable region having at least 80% identity with the sequence SEQ ID NO: 25, with the proviso that said variable region of the light chain comprises from its N-terminal to its end C-terminal: the CDR1 of sequence SEQ ID NO: 31; the CDR2 of sequence SEQ ID NO: 33; and the CDR3 of sequence SEQ ID NO: 35.

17. The monoclonal antibody according to claim 14, comprising: a heavy chain comprising or consisting of a sequence having at least 80% identity with the sequence SEQ ID NO: 3, with the proviso that said heavy chain comprises the variable region of sequence SEQ ID NO: 7; and a light chain comprising or consisting of a sequence having at least 80% identity with the sequence SEQ ID NO: 21, with the proviso that said heavy chain comprises the variable region of sequence SEQ ID NO: 25.

18. A pharmaceutical composition comprising as active substance at least the monoclonal antibody according to claim 14 in combination with a pharmaceutically acceptable vehicle, said monoclonal antibody being at a dose of 5 mg to 1,000 mg.

19. A method for the early treatment and/or prevention of relapse of granulomatosis with polyangiitis comprising the administration of the pharmaceutical composition according to claim 18.

20. The method according to claim 19, wherein said pharmaceutical composition is formulated to be administered by one of the following routes: oral, parenteral, injectable, topical, inhalation, subcutaneous, nasal or pulmonary.

Description

LIST OF FIGURES

[0232] FIG. 1: Identification and characterisation of the human monoclonal anti-proteinase 3 (PR3) 4C3 antibody.

[0233] A. Recognition of PR3 by 3 antibodies derived from immunization n° IM3-16 (4C3, 4C5 and 5D11) in ELISA. B. Verification of the specificity of 4C3 and 5D11 in response to different antigens indicated on the graph by ELISA (n=5). C. Determination of the IgG subclass of 4C3 by ELISA in comparison with a GPA (ANCA) patient serum containing different anti-PR3 IgG subclasses (n=3). D. Determination of the kappa (black) or lambda (white) light chain of 4C3 in comparison with a GPA patient serum (n=3). E. Determination of the equilibrium dissociation constant (K.sub.D) of the 4C3 by BIACORE™ with respect to the PR3.

[0234] FIG. 2: Production and purification of 4C3.

[0235] A. Confirmation of the monoclonality of 4C3 by Polymerase Chain Reaction (PCR) by studying the rearrangements of immunoglobulin heavy chain (IgH) genes. B. Production of the 4C3 clone in the supernatants of high-density flasks. (PRO 01-17 and PRO 02-18) and evolution of the number of cells harvested during PRO 01-17 (left graph) and PRO 02-18 (right graph). C. Evaluation of the level of anti-PR3 IgG production by ELISA during PRO 01-17 (left graph) and PRO 02-18 (right graph). D. Purification of 4C3 by affinity chromatography (HiTrap™ Protein A). The purity of the different fractions was controlled by SDS-PAGE and Coomassie Blue staining. NR: Not retained; E: Elutions; Mq: Size marker; Avt: Sample before purification; CTL: Purified IgG1 Ac. E. Recognition of PR3 by purified 4C3 from PRO 02-18 production by anti-PR3 ELISA (coating with 1 μg/mL of PR3).

[0236] FIG. 3: Specificity of the recognition of PR3 by the recombinant form of 4C3 (r4C3).

[0237] A. Comparison of the binding of 4C3 (black histograms) and recombinant 4C3 (r4C3) (dark grey histograms) to PR3 compared to BSA and ovalbumin (Ova). A representative experiment of the 3 performed is shown. B. Determination of the equilibrium dissociation constant (K.sub.D) of r4C3 by BIACORE™ with respect to PR3.

[0238] FIG. 4: Modelling and validation of the PR3 epitope recognised by 4C3.

[0239] A. Recognition of PR3 by 4C3 by western blot: PR3 in different concentrations and in native or denatured form was deposited on 4-12% Bis-Tris polyacrylamide gel and then transferred to a nitrocellulose membrane and saturated with 5% BSA. Incubation of the membrane with the 4C3 antibody diluted to 1/10,000th one night at 4° C. under agitation. Revelation of the proteins with a secondary anti-human Fc antibody coupled to HRP. B. In silico modelling of the epitope of PR3 recognised by 4C3 by the MabTope method. C. Amino acid sequence of PR3 (SEQ ID NO: 42) with the tested validation peptides underlined. The probability of different residues of the epitope is indicated. D. HTRF measurement of the binding between an anti-Fab d2-coupled antibody that binds to 4C3 and a terbium-coupled streptavidin that binds to the different biotinylated peptides of PR3 (shown on the abscissa). Results expressed as a ratio. ns: not significant. E Amino acid sequence of PR3 whose signal sequence, di-propeptide, pro-peptide, hydrophobic patch, catalytic triad and peptides 3.2 and 4.1 are indicated: [0240] Bold: signal sequence; [0241] AE underlined: di-propeptide cleaved by cathepsin C; [0242] From IVGGH . . . IRSTLR: N-terminal sequence of the active native PR3; [0243] In italics: pro-peptide in C-terminal; [0244] In bold italics: hydrophobic patch; [0245] Bold underlined: catalytic triad;

TABLE-US-00002 (SEQ ID NO: 40) -T CRPHNICTFVPR: peptide 4C3_3.2; and (SEQ ID NO: 41) -GTQCLAMGWGRVGAH: peptide 4C3_4.1.

[0246] F. Effect of alpha 1 antitrypsin (α1AT) in the binding of 4C3 antibody to PR3 in ELISA. PR3 (2 μg/mL) was incubated or not with alpha 1 anti-trypsin at a ratio of 1 PR3 to 5 α1AT for one hour at 37° C. before being coated in wells overnight at 4° C. The results are expressed as an average±standard deviation of the optical density obtained. *p<0.05. The average of the percentage decrease obtained over the 8 experiments is indicated.

[0247] FIG. 5: 4C3 does not inhibit the enzymatic activity of PR3.

[0248] PR3 (10 nM) has been incubated or not with 4C3, 6H4 or α1AT (50 nM) during 30 minutes, i.e. a ratio of 5:1, then the PR3 fluorescent substrate was added. Enzyme activity was measured during kinetics by spectrofluorimetry. The negative control (ctl-) corresponds to the substrate alone. A. Representation of the fluorescence obtained at the ratio of 5 Ac 4C3 for 1 PR3. Similar results were obtained with the ratios 10:1 and 2:1. B. Representation of the difference in activity of PR3 expressed in ΔRFU. A representative experiment of 5 is shown.

[0249] FIG. 6: Marking of 4C3 on the surface of immune cells in flow cytometry.

[0250] A. Analysis strategy for human neutrophils purified by flow cytometry (FACS) after elimination of duplicates (“single cells”) then labelling of the membrane PR3 with the 4C3 antibody coupled with AF488 (graph at bottom right). B. Human neutrophils were purified from the blood of a healthy donor and then incubated or not (histogram “0 μg/ml”) with different concentrations of 4C3 antibodies coupled to AF488 (1, 20 or 100 μg/mL) for 20 minutes at 4° C. in order to study the membrane expression of 4C3. C. Human neutrophils from healthy donors were pre-activated (TNFα+) or not (TNFα−) with TNFα at 2 ng/mL for 15 minutes before being labelled with 4C3-AF488 at 20 μg/ml for 20 minutes. A representative experiment of the 10 carried out is shown. D. Blood cells from a healthy donor were labelled with a CD45-APC mix. H7/CD3-BV786/CD14-VioBlue/CD15-PE/4C3-AF488 before switching to FACS. Representation of PR3 marking with 4C3-AF488 on the surface of T lymphocytes (CD3.sup.+, light grey histogram), monocytes (CD14.sup.+, dark grey histogram) and neutrophils (CD15+, hatched histogram). A representative experiment of the 5 performed is shown.

[0251] FIG. 7: Cytoplasmic marking of 4C3-AF488 cANCA.

[0252] Purified neutrophils from healthy donors were fixed with ethanol (top row) or formalin (bottom row) before being labelled with the nuclear marker DAPI (1.sup.st column) and 4C3-AF488 ( 1/100th) (2.sup.nd column). Reading the slides with a fluorescence microscope using the 60× objective. Superimpose the fluorescence (“Merged”, 3.sup.rd column) using the ImageJ software. The scale (10 μm) is indicated on the phase contrast images (“Brightfield”, 4th column) A representative experiment of the 3 experiments is shown.

[0253] FIG. 8: 4C3 specifically recognises purified native PR3 and PR3 contained in neutrophils.

[0254] A. Neutrophils from a GPA patient were purified and then pre-activated (TNFα+) or not (TNFα−) with TNFα at 2 ng/mL for 15 minutes at 37° C. 10 μg of neutrophils (wells 1 and 2) were deposited in a reduced condition. HeLa cells without PR3 were used as a negative control (3.sup.rd well) and purified native human PR3 as a positive control (4.sup.th well). Incubation of the membranes with Hsc70 antibody ( 1/10,000.sup.th) or 4C3 antibody ( 1/10,000.sup.th) overnight under agitation at 4° C. B. The 4C3 does not bind the proteases Elastase and Cathepsin G (Cath G) but only PR3 at 5 μg and 2 μg. Incubation of 4C3 overnight at 4° C.

[0255] FIG. 9: 4C3 induces the expression of PR3 on the surface of pre-activated neutrophils against a non-relevant antibody.

[0256] Neutrophils from healthy donors were pre-activated by TNFα at 2 ng/mL at 37° C. for 15 minutes (triangle) and then incubated for 90 minutes at room temperature with monoclonal antibody (MAb) 4C3 (circle) and MAb 6H4 (square) at 2 μg/mL (n=3), at 20 μg/mL (n=4) and at 100 μg/mL (n=2) or with PMA-ICa (inverted triangle). The membrane expression of PR3 (mbPR3) was analysed by flow cytometry after labelling with the anti-PR3 antibody WGM2-FITC. The mean as well as standard deviations for healthy donors are shown. The results have been normalised to the TNFα condition and are expressed as a ratio of Mean Fluorescence Intensity (MFI). *p<0.05.

[0257] FIG. 10: 4C3 does not induce the production of reactive oxygen species (Reactive Oxygen Species; ROS) by human neutrophils.

[0258] Purified neutrophils from eight independent healthy donors were pre-activated with TNFα (2 ng/mL) for 15 minutes at 37° C. (white columns) before being incubated for 45 minutes with 4C3 (grey columns), IgG preparations (unmixed) from two healthy donors (hatched columns) or from four GPA patients active at the time of diagnosis (gridded columns) or IgG from patient P2 (vertical rows). The production of ROS was evaluated by measuring the fluorescence (MFI) of DHR 123 by flow cytometry. The results are expressed as an average±SEM obtained in eight independent experiments, each circle representing one experiment. NS: not significant; *p<0.05; **p<0.005; ***p<0.0005.

[0259] FIG. 11: Dose effect of 4C3 on the intracellular production of reactive oxygen species (Reactive Oxygen Species; ROS) by neutrophils.

[0260] Neutrophils from healthy donors were incubated with cytochelin B (5 μg/mL). 5 minutes at 37° C. then sodium azide (2 mM) and DHR (2.5 μM) were added. 5 minutes at 37° C. before pre-activating neutrophils with TNFα at 2 ng/mL at 37° C. for 15 minutes (triangle). At the end of the pre-activation, the neutrophils were incubated 45 minutes at 37° C. with 4C3 AcMo (circle) or 6H4 AcMo (square) at 2 μg/ml (n=9), at 20 μg/mL (n=20) and at 100 μg/mL (n=4) or with PMA-ICa (inverted triangle). The production of ROS was analysed by flow cytometry by measuring the MFI after marking with DHR 123. The mean as well as standard deviations for healthy donors are shown. The results have been normalised to the condition TNFα and are expressed as a ratio of MFI. ns: not significant; *p<0.05; **p<0.005.

[0261] FIG. 12: 4C3 and r4C3 do not induce an increase in cathepsin G activity by pre-activated neutrophils.

[0262] A. Purified neutrophils from eight independent healthy donors were pre-activated with TNFα (2 ng/mL) for 15 minutes at 37° C. (white columns) before being incubated for 45 minutes with 4C3 (grey columns), IgG preparations (unmixed) from two healthy donors (hatched columns) or from four GPA patients active at the time of diagnosis (gridded columns) or IgG from patient P2 (vertical rows). Neutrophil degranulation was assessed by CD63 expression as a percentage of positive cells. The results are expressed as an average±SEM obtained in eight independent experiments, each circle representing one experiment. NS: not significant; *p<0.05; **p<0.005; ***p<0.0005. B. Neutrophils from healthy donors were pre-activated by TNFα at 2 ng/mL at 37° C. for 15 minutes (left histogram) and then incubated for 90 minutes at room temperature with 2 and 20 μg/mL r4C3 and r4C3 MAc or with PMA-ICa. Cathepsin G activity was measured in neutrophil activation supernatants after addition of its substrate and reading by spectrofluorimetry. A representative experiment of 3 is shown. The results are expressed in ΔRFU (Relative Fluorescence Units).

[0263] FIG. 13: 4C3 does not increase the adhesion phenotype of human neutrophils.

[0264] A and B. Purified neutrophils from eight independent healthy donors were primed with TNFα (2 ng/mL) for 15 minutes at 37° C. (white columns) before being incubated for 45 minutes with 4C3 (grey columns), separate IgG preparations (unmixed) from two healthy donors (hatched columns) or from four GPA patients active at the time of diagnosis (gridded columns) or IgG from patient P2 (vertical rows). The neutrophil adhesion criterion was evaluated by measuring the surface expressions of CD11b (A) and CD18 (B) by flow cytometry. The results are expressed as an average±SEM obtained in eight independent experiments, each circle representing one experiment. NS: not significant; *p<0.05; **p<0.005; ***p<0.0005.

[0265] FIG. 14: Analysis by SDS-PAGE of the different shapes of 4C3 obtained.

[0266] 4C3 native (well 2), 4C3 recombinant (well 3), 4C3 deglycosylated (well 5) and 4C3 Fab (well 6) compared to 6H4 (well 4; anti-ovalbumin IgG1). The left well (No. 1) corresponds to the molecular weight (MW) marker in kilo Dalton (kDa).

[0267] FIG. 15: Derivatives of 4C3 do not increase PR3 expression and do not induce the production of SAR by pre-activated neutrophils.

[0268] A. Neutrophils from healthy donors were pre-activated by TNFα at 2 ng/mL at 37° C. for 15 minutes and then incubated for 90 minutes at room temperature with MoA 4C3 and MoA 6H4 at 40 μg/mL or with PMA-ICa or in mol equivalent for Fab and F(ab′).sub.2. The membrane expression of PR3 (mbPR3) was analysed by flow cytometry after labelling with 4C3-AF488 antibody. An experiment representative of 3 realised is shown. B. The production of ROS was measured in kinetics in the presence of the Fab, F(ab′).sub.2 and deglycosylated Fc (dFc) derivatives of the 4C3 antibody. n=1.

[0269] FIG. 16: r4C3 does not increase the intracellular production of SAR by neutrophils. Neutrophils from healthy donors were pre-activated with TNFα (triangle) and then incubated for 90 minutes with r4C3 (circle, n=11), 4C3 (diamond, n=20) or with PMA-ICa (inverted triangle). The production of ROS was analysed by flow cytometry by measuring the MFI after marking with DHR 123. The mean as well as standard deviations for healthy donors are shown. The results have been normalised to the condition TNFα and are expressed as a ratio of MFI. NS: not significant.

[0270] FIG. 17: Analysis in SDS electrophoresis—PAGE of the different fractions obtained after purification of a GPA patient serum on G protein.

[0271] Serum (well 2), unbound fraction (NR, well 3), washes (L, wells 4 to 9), elutions (E, wells 10 to 14). The left (No. 1) and right (No. 15) wells correspond to the molecular weight (MW) marker in kDa. Representative image of the 6 experiments carried out.

[0272] FIG. 18: Purified IgG from GPA patient serum contains anti-PR3 IgG in contrast to purified IgG from healthy donor serum.

[0273] The IgG contained in the serum of healthy donors and GPA patients was purified and then the anti-PR3 IgG assay was performed by ELISA using the Eurolmmun™ kit. The kit's internal controls (negative control/positive control/20 UR/mL calibrator) were used in comparison with the 50 μg/mL AcMo 4C3 and unpurified sera. A representative experience of the 6 performed is shown.

[0274] FIG. 19: Purified IgG induces an increase in the intracellular production of ROS by neutrophils from healthy donors.

[0275] Neutrophils from healthy donors were pre-activated with TNFα at 2 ng/mL at 37° C. for 15 minutes (triangle) and then incubated with purified IgG from GPA patients (circle, n=32) or from healthy donors (diamond, n=16) at 200 μg/mL or with 4C3 (inverted triangle, n=20). The production of ROS was analysed by flow cytometry by measuring the MFI after marking with DHR 123. The mean as well as standard deviations are shown. The results have been normalised to the condition TNFα and are expressed as a ratio of MFI. ns: not significant; *p<0.05; **p<0.005; ***p<0.0005.

[0276] FIG. 20: 4C3 is capable of inhibiting the intracellular ROS production of human neutrophils induced by the presence of IgG GPA+.

[0277] Neutrophils from healthy donors were incubated with cytochelin B (5 μg/mL). 5 minutes at 37° C. then sodium azide (2 mM) and DHR (2.5 μM) were added. 5 minutes at 37° C. before pre-activating the neutrophils with TNFα at 2 ng/mL at 37° C. for 15 minutes. At the end of pre-activation, neutrophils were incubated for 15 minutes at 37° C. with AcMo 4C3 at 20 μg/mL before adding IgG from patients with GPA at 200 μg/mL for 35 minutes at 37° C. ROS production was analysed by flow cytometry by measuring MFI after labelling with DHR 123. The results obtained with 4 IgG from GPA patients are shown. The results are expressed in MFI and the percentage of inhibition induced by 4C3 is shown. n=4.

EXAMPLES/MATERIALS & METHODS/RESULTS

[0278] 1. Patient Selection

[0279] Before the 4C3 antibody was obtained, three immortalisations were performed from blood samples of 3 different patients with Wegener's disease or Granulomatosis With Polyangiitis (GPA). Clones could be obtained but none of them resulted in the production of a relevant antibody. The causes of these failures being multiple: lymphopenia, thrombocytosis, hyperleukocytosis, it was decided to adapt the process by targeting patients more specifically according to the clinic and biology, and by optimising the immortalisation technique.

[0280] The 4C3 clone is derived from a patient suffering from granulomatosis with polyangiitis (GPA) who was followed up in consultation in the pneumology department of the Tours Regional and University Hospital Centre (CHRU). This patient was diagnosed with GPA in 2011 following ear, nose and throat (ENT) and lung problems and an alveolar haemorrhage. This patient received an initial 6-month treatment with Endoxan™ followed by corticosteroids and methotrexate. The last biological examination at the time of sampling indicated a circulating B-cell count of 72/mm.sup.3 and a of autoantibodies against proteinase 3 (cANCA) greater than 177 IU/mL. The patient agreed to participate, after informed consent, in a collection of human biological samples declared to the Ministry of Research (n°DC-2012-1636) in accordance with decree n°2007-1120 of 10 Aug. 2007. After collecting clinical data and signing the informed consent, the patient's blood was collected on ACD (Dextrose Citrate anti-coagulant) and then sent via the immunology laboratory of the Tours CHRU to the BCRessources platform.

[0281] 2. Immortalisation of B-Lymphocytes

[0282] To obtain anti-PR3 IgG-producing clones (experiment n°IM3-16), a kit marketed by the company Dendritics™ (DDXK-HuBB™ kit) was used, which uses the EBV (Epstein Barr Virus) immortalisation technique. Nevertheless, the protocol recommended by this company has been optimised by the inventors and is different from the one described in the kit. Normally, the kit is used on fresh total Peripheral Blood Mononuclear Cells (PBMC) (from a daily blood sample) and requires a first immortalisation step and then sub-cloning steps to obtain monoclonal cells. To obtain the 4C3 antibody, optimisation of the kit led to the use of thawed (and not fresh) PBMCs, which were then enriched by cell sorting with autologous memory B lymphocytes (LB) in the culture wells, i.e. a new stage of the protocol was developed by the inventors. In addition, and as a consequence of this optimisation, the culture conditions that were put in place also made it possible to dispense with the subcloning step, which is described as essential in the kit protocol.

[0283] More precisely, for the enrichment in memory B lymphocytes, two successive steps were carried out: the first step allowed the isolation by negative selection of the total B lymphocytes (“B-cell isolation kit II” from Miltenyi Biotec˜, Ref 130-091-151) and then the specific sorting of the CD19.sup.+ CD27.sup.+ memory B lymphocytes with the Astrios MoFlo™ sorter. Different co-culture conditions were tested in flat-bottom P96: PBMC alone, PBMC enriched in memory B-lymphocytes, PBMC depleted in memory B-lymphocytes and enriched in memory B-lymphocytes with different quantities of cells During the second step, the cells were stimulated and immortalised in the presence of EBV viral particles using the kit marketed by Dendritics™ (kit DDXK-HuBB™). During the first ten days of culture, cluster formation was evaluated under the microscope and the cells were restimulated after 9 days (D9) of culture.

[0284] 3. Selection of Clones by ELISA

[0285] From culture days 20, the identification of IgG producing clones by ELISA has been started. Total IgG positive supernatants were identified and the corresponding clone cultures were continued. Conversely, wells that came out twice negative were eliminated. In a first step, 62 clones were selected on the basis of their detectable total IgG production.

[0286] These positive clones were again screened by ELISA in the presence of PR3 (Athens Research and Technology™, reference 16-14-161820). After coating the wells of the MaxiSorp™ ELISA plate (96 wells) with 2 μg/mL of native PR3 overnight at 4° C., the wells were saturated with 4% PBS-BSA (Phosphate Buffer Saline—Bovine Serum Albumin) buffer to prevent non-specific binding of antibodies to the well plastic. The B-cell culture supernatants were incubated for 2 hours at 37° C. before adding the secondary antibody, i.e. anti-Fc human IgG coupled with HRP (HorseRadish Peroxidase). The revelation was made in the presence of TMB (3,3′,5,5′-TetraMethylbenzidine), peroxidase substrate. The absorbance reading was taken at a wavelength of 620 nm. The threshold of positivity was set at twice the absorbance value of the negative control. Cells whose supernatants came out positive at least three times in ELISA were amplified in order to have enough cells for further testing. Conversely, wells that tested negative twice in a row were discarded. Thus, only 5 clones tested positive in anti-PR3 ELISA. The 5 clones were expanded, re-tested by ELISA and the best 3 were retained (i.e. 4C3, 4C5 and 5D11) (FIG. 1A). Only 4C3 was found to be specific for PR3, 5D11 being non-specific (FIG. 1B) and 4C5 having stopped producing IgG.

[0287] 4. Characterisation of 4C3 Antibody

[0288] After confirmation of the specificity of the 4C3 antibody (FIG. 1B), it was shown by ELISA (commercially available ELISA tests) that this anti-PR3 antibody of the IgG isotype had an IgG1 subclass and a kappa chain (FIGS. 1C and 1D). The affinity of 4C3 for PR3 is high with a dissociation constant K.sub.D of 7.41.10-10 M (value obtained by the BIACORE technique) (FIG. 1E+Table 2 below).

TABLE-US-00003 TABLE 2 Affinity of 4C3 antibody 4C3 ka (1/Ms) kd (1/s) K.sub.D (M) Rmax Chi.sup.2 (RU.sup.2) U-value Curve 1.26E+07 0.00936 7.41E−10 (RU) Conc (M) 0.723 3 Cycle: 5 78.35 1.25E−09 Cycle: 6 76.5 2.50E−09 Cycle: 7 80.18 5.00E−09 Cycle: 8 71.98 1.00E−08 Cycle: 9 75.01 2.00E−08

[0289] 5. Production in High Density Flasks

[0290] After an initial characterisation of the antibody secreted in the culture supernatants, clone 4C3 was amplified by successive passages of the cells in P24 wells, then in P6 wells and finally in flasks in order to increase the number of cells. The monoclonality of this clone was confirmed by Polymerase Chain Reaction (PCR) by studying the rearrangements of immunoglobulin heavy chain (IgH) genes (FIG. 2A). The 4C3 clone was then put into production in a high-density flask (CELLine™ Wheaton™) composed of two compartments separated by a semi-permeable membrane that allows the passage of proteins below 10 kDa. The first compartment contains the extracellular nutrient medium composed of DMEM plus penicillin/streptomycin and glutamine but without horse serum while the second compartment contains the cells in which the antibodies produced every three to four days over a period of at least 70 days were collected.

[0291] Two productions were carried out (experiments PRO 01-17 and PRO 02-18) during which the number of cells (FIG. 2B) and the level of anti-PR3 IgG production (FIG. 2C) were checked.

[0292] 6. Purification of 4C3 Antibody

[0293] All the supernatants collected were filtered through a 0.2 μm filter after centrifugation. 10 minutes at 500 g. The antibodies contained in the supernatants were purified by affinity chromatography (HiTrap™ Protein A 1 mL GE Healthcare™) on the AKTA™ (GE Healthcare™) apparatus. The purity of the different fractions was controlled by 4-12% Bis-Tris gel electrophoresis and Coomassie Blue staining (FIG. 2D). The final concentration of 4C3 antibody, after BCA™ assay, is 5,974 mg/mL. The specificity of 4C3 was confirmed by ELISA in the presence of coated native PR3 (1 μg) (FIG. 2E).

[0294] 7. Specificity of 4C3 in its Recombinant Form

[0295] In order to produce the recombinant 4C3 antibody (r4C3), the sequences of the PR3-specific variable regions (SEQ ID NOs: 6 and 24) were subcloned into recombinant IgG expression vectors (commercially available vectors). The expression vectors thus constructed were transfected into HEK-293 mammalian cells (ATCC® CRL-1573™) for the production of monoclonal IgG isotype in the form secreted in the culture supernatant. After harvesting 50 mL of culture supernatant, the r4C3 antibody was purified as previously described and yielded a final r4C3 concentration of 3.5 mg/mL.

[0296] To confirm the specificity of the recombinant 4C3 antibody, an ELISA test was performed in the presence of different amounts of coated PR3 compared to 4C3 (FIG. 3A). It was shown that r4C3 is able to bind to PR3 in a manner comparable to 4C3 (FIG. 3A). This result has been confirmed by BIACORE as r4C3 has an affinity to PR3 comparable to 4C3 (FIG. 3B+Table 3 below).

TABLE-US-00004 TABLE 3 Affinity of the r4C3 antibody r4C3 ka (1/Ms) kd (1/s) K.sub.D (M) Rmax Chi.sup.2 (RU.sup.2) U-value Curve 1.37E+07 0.009358 6.84E−10 (RU) Conc (M) 1.06 4 Cycle: 14 161.8 1.25E−09 Cycle: 15 128.1 2.50E−09 Cycle: 16 110.9 5.00E−09 Cycle: 17 98.36 1.00E−08 Cycle: 18 100.5 2.00E−08

[0297] The two 4C3 and r4C3 antibodies share the same variable region of the heavy chain (SEQ ID NOs: 6 to 19) and the same light chain (SEQ ID NOs: 20 to 35) but differ only in the constant region of the heavy chain by the mutation R>K in position 224 (or AGA>AAA in position 670-672). This mutation confers a different allotype to these 2 antibodies, namely G1m3-1 for 4C3 and G1m17-1 for r4C3.

[0298] 8. Identification of the Epitope Recognised by 4C3

[0299] In order to identify the epitope of PR3 recognised by 4C3, a western blot in a non-denaturing and denaturing condition was first made. Proteinase 3 (native or denatured) was separated by electrophoresis on 4-12% Bis-Tris polyacrylamide gel, then transferred to a nitrocellulose membrane and saturated with 5% BSA in 0.1% Tween™ 20 TBS at room temperature for one hour under agitation. The membrane was then incubated with 4C3 ( 1/10,000.sup.th dilution) overnight at 4° C. under agitation. To reveal the proteins, the membranes were incubated with a secondary anti-human Fc antibody coupled to HRP and revealed with the Pierce™ ECL Plus (ThermoFisher™) kit. It was thus determined that 4C3 preferentially recognised a conformational epitope of PR3 with observation of a less intense labelling of PR3 in denaturing condition (FIG. 4A).

[0300] The identification of the epitope recognised by the 4C3 antibody was carried out by MabSilico. The first step enabled the PR3 epitope to be modelled using the MabTope method, which consists of the computer prediction of an ordered list of PR3 peptides likely to belong to the epitope from the sequence of the variable part (V.sub.H and V.sub.L) of 4C3 (FIGS. 4B and 4C). The second step allowed the experimental measurement of the specific binding of a number of these peptides to the antibody. This measurement was made by HTRF (Homogeneous Time Resolved Fluorescence) between an anti-Fab d2-coupled antibody (CisBio™) which binds to the antibody of interest, and a terbium-coupled streptavidin (CisBio™) which binds to the biotinylated peptide (FIG. 4D). The peptide 4C3_3.2 (sequence SEQ ID NO: 40) induced the highest HTRF ratio (FIG. 4D) thus demonstrating that the epitope recognised by the 4C3 consists of a majority of these amino acids (FIG. 4E).

[0301] In addition, the binding of 4C3 to PR3 was altered in the presence of alpha 1 anti-trypsin (α1AT), the natural PR3 inhibitor, which is able to induce a conformational change in PR3 that prevents anti-PR3 Ac's that recognise epitope 1 from binding to it (FIG. 4F).

[0302] 4C3 therefore targets a conformational epitope of PR3 close to its catalytic site (amino acids in bold underlined, FIG. 4E) and to the hydrophobic patch (amino acids in italic bold, FIG. 4E) and over a region overlapping that of epitope 1 of PR3 (FIG. 4B).

[0303] 9. Enzymatic Activity of PR3 in the Presence of 4C3

[0304] To better characterise the 4C3 antibody, the enzymatic activity of PR3 was measured after incubation with 4C3 for 30 minutes at different ratios (10 Ac: 1 PR3/5 Ac: 1 PR3 and 2 Ac: 1 PR3) and after addition of the commercial PR3 fluorescent substrate. (Abz-VADnVADYQ-YNO2). It has thus been shown that 4C3 does not inhibit the enzymatic activity of PR3 under these experimental conditions, just like 6H4, which is a non-relevant antibody, whatever the ratio used (FIG. 5). Indeed, in the presence of these antibodies, the substrate was degraded (increased fluorescence) whereas the addition of alpha 1 anti-trypsin (1AT) which is a natural PR3 inhibitor did not induce any degradation of the PR3 substrate (FIG. 5).

[0305] 10. Labeling of the Membrane PR3 by Flow Cytometry

[0306] The ability of the 4C3 antibody to bind to PR3 was also validated by flow cytometry. For this purpose, neutrophils were purified from the blood of healthy donors at the Etablissement Français du Sang (EFS) using a commercially available kit (Stem Cell kit, EasySep™ Direct Human Neutrophil Isolation; Kit reference 19666) (FIG. 6A). Unstimulated neutrophils were incubated for 30 minutes at 4° C. with 4C3 antibody previously coupled with fluorochrome AF488 (kit marketed ThermoFisher™). Several concentrations of 4C3 were tested (0, 1, 20 or 100 μg/mL) and revealed a more intense labelling of PR3 on the surface of the neutrophils depending on the concentration of coupled antibody used (FIG. 6B). In addition, it has previously been described in the literature that priming neutrophils with TNF alpha (TNF) induced an increase in membrane PR3 expression. This was confirmed by an increase in labelling with 4C3-AF488 after stimulation of neutrophils for 15 minutes at 37° C. with TNF alpha (10 ng/mL) demonstrating that Ac 4C3-AF488 recognises PR3 well (FIG. 6C). Finally, this result was reinforced by the absence of 4C3-AF488 labelling on the surface of lymphocytes (lymphocytes not expressing membrane PR3), intermediate labelling on monocytes (PR3 weakly expressed) and very strong labelling on neutrophils (PR3 constitutively expressed) (FIG. 6D).

[0307] 11. Labeling of Cytoplastic 4C3 cANCA Type

[0308] Purified neutrophils from healthy donors were first fixed with formalin or ethanol and then labelled with Ac 4C3 coupled with AF488 and DAPI (nuclear labelling). Analysis of the slides under a fluorescence microscope revealed a diffuse cytoplasmic cANCA-type labelling of 4C3 (FIG. 7).

[0309] 12. Recognition of PR3 by the 4C3 in Western Blot

[0310] To confirm anti-PR3 specificity, 4C3 was used as a primary antibody in western blot (FIGS. 8A and 8B) on protein lysates from neutrophils, HeLa cells (ATCC® CCL-2™) and human PR3. According to FIG. 8, 4C3 did recognise human PR3 with a size-labelling at the expected size, which was also present with neutrophils and no labelling was detected in HeLa cells, which do not express PR3 (FIG. 8A). In addition, 4C3 did not recognise other proteases such as elastase and cathepsin G which are the same size as PR3 (FIG. 8B). This result therefore confirmed the fact that 4C3 antibody is specific for PR3.

[0311] 13. Depletion of 4C3 Fc Fragments

[0312] In order to produce anti-PR3 antibodies that neutralise the ANCA-PR3 interaction and thus decrease neutrophil activation during GPA, different fragments of 4C3 were tested after depletion of the Fc fragments of 4C3, which are capable of activating neutrophils via the Fc/FcγR interaction. To do this, two enzymes were used: pepsin and papain. Pepsin is a proteolytic enzyme that allows the antibody to be cleaved just below the disulphide bridges that link the two heavy chains to obtain an F(ab′).sub.2 fragment (Table 4 below). Papain is a cysteine protease capable of cleaving the antibody above the hinge region to give two distinct Fab fragments on the one hand and Fc fragments on the other (Table 5 below).

TABLE-US-00005 TABLE 4 List of samples deposited by track DIGESTION OF 4C3 BY PEPSIN Track Samples mg/mL 1 F(ab′).sub.2 (Flowthrough Protein A) 0.259 2 F(ab′).sub.2 (Flowthrough Protein A) 0.257 3 F(ab′).sub.2 (Protein A wash) 0.068 4 F(ab′).sub.2 (Protein A wash) 0.034 5 E1 Protein A 0.051 6 E2 Protein A 0.044 7 E2 Protein A 0.051 8 4C3 purified 5.9

TABLE-US-00006 TABLE 5 List of samples deposited by track DIGESTION OF 4C3 BY PAPAIN Track Samples mg/mL 1 Fc (E1) 0.294 2 Fc (E2) 0.231 3 Fc (E3) 0.012 4 Fab (Flowthrough Protein A) 0.542 5 Fab (Protein A wash) 0.166 6 E1 2.sup.nd pass of Fab on Protein A 0.002 7 E2 2.sup.nd pass on Fab on Protein A 0.0027 8 E3 2.sup.nd pass on Fab on Protein A 0.044 9 4C3 purified 5.9

[0313] 14. Functional Tests on Neutrophils

[0314] Incubation of ANCA with neutrophils pre-activated by TNFα leads to degranulation, production of ROS and the formation of NET (Neutrophil Extracellular Traps) and therefore to neutrophil activation. In order to test the neutralising action of 4C3 and its derivatives (r4C3, F(ab′).sub.2, Fab, deglycosylated 4C3 . . . ) on human neutrophils, various in vitro functional tests have been set up.

[0315] Material & Methods

14.1 Purification of IgG from Serum

[0316] For functional tests, we have selected sera from patients with active disease according to clinical examinations, treatment and PR3-ANCA level. Patients with an active disease were diagnosed at the University Hospital Centre of Tours. We purified IgG from five sera of independent GPA patients during the active phase of the disease after diagnosis (IgG GPA) (see Table 6 below) or during remission for patient P2 (IgG P2) and two independent healthy donors with a rapid flow Sepharose™ 4 protein G kit (GE HealthCare®, USA). Briefly, unmixed sera were incubated with protein G for one hour at room temperature before elution. The samples were filtered using a Spin-X® UF Concentrator ultrafiltration system. The purified IgG from the unmixed sera was subjected to electrophoresis on a Bis-tris 4-12% before being coloured with Coomassie blue. The presence of PR3-ANCA in separate IgG preparations from active GPA patients and patient P2 was confirmed by ELISA using a EuroImmun anti-PR3 kit, whereas separate purified IgG preparations from healthy donors did not contain PR3-ANCA.

TABLE-US-00007 TABLE 6 Main characteristics of active GPA patients whose IgG has been purified to induce autoimmune neutrophil activation GPA 1 GPA 2 GPA 3 GPA 4 GPA 5 Gender Male Male Male Male Male Age at 68 years 71 years 79 years 77 years 53 years diagnosis old old old old old Reach Renal, Joint Renal Lung Renal neurological and and and and and ENT ENT joint ENT pulmonary Type of cANCA cANCA cANCA cANCA cANCA ANCA Level of 83 36 43 112 58 PR3-ANCA IU/mL IU/mL IU/mL IU/mL IU/mL

[0317] 14.2 Neutrophil Purification and Pre-Activation by TNFα

[0318] Human neutrophils from healthy independent donors have been purified by negative magnetic selection with the commercial “EasySep® Direct Human Neutrophil Isolation Kit” (StemCells®, Canada) following the manufacturer's instructions. At the end of the isolation, the neutrophils were suspended in a calcium- and magnesium-free HBSS solution. The purity of the isolated neutrophils was >95%, after evaluation by flow cytometry (CD15-PE and Live dead, Miltenyi Biotech®, Germany). The cells were pre-activated with TNFα (Sigma-Aldrich®, USA) at a final concentration of 2 ng/mL for 15 minutes at 37° C. in a water bath.

[0319] 14.3 Assessment of SAR Production by Neutrophils

[0320] Neutrophil activation from healthy independent donors was assessed by the production of ROS using a dihydrorhodamine 123 (DHR 123) test. Purified neutrophils were suspended in HBSS with 1 mM Ca2+ and 1 mM Mg2+ and incubated with 5 μg/mL cytochalasin B (Cayman Chemical®, USA) to increase the production of oxygen radicals, for 5 minutes at 37° C. The cells were then loaded with 2 μM DHR 123 and 2 mM sodium azide (NaN3) for 5 minutes at 37° C. with agitation. Primed neutrophils were incubated with 4C3 (2-100 μg/mL), r4C3 or separate IgG preparations from two healthy donors and four active GPA patients (200 μg/mL) for 45 minutes at 37° C. An irrelevant antibody (6H4, IgG1K, anti-ovalbumin) was used as a negative control. A combination of phorbol myristate acetate (PMA, 50 ng/mL) and calcium ionophore (ICa, 10 μM), both potent neutrophil activators, was used as a positive control for neutrophil activation. The reaction was stopped with iced EDTA PBS (1 mM) prior to measuring the fluorescence of DHR 123 by flow cytometry. For the neutralisation experiments, pre-treated primate neutrophils were first incubated with 4C3 (20 μg/mL) for 15 minutes at 37° C. and then separate IgG preparations of five patients active at the time of GPA diagnosis (IgG GPA, 200 μg/mL) were added for a further 45 minutes.

[0321] 14.4 Degranulation and Adhesion of Neutrophils

[0322] Neutrophils were pre-activated with TNFα and stimulated with 4C3 (2 and 20 μg/mL), r4C3 (2 and 20 μg/mL), IgG from GPA patients (200 μg/mL) or PMA-ICa for 45 minutes at 37° C. After incubation, the cells were washed and labelled with CD63 FITC-coupled (degranulation) or labelled with CD11b VioBlue/CD18 FITC (BD Biosciences®, USA) antibodies (adhesion phenotype) for 20 minutes at 4° C. before being analysed by flow cytometry. The percentage of positive cells and the mean fluorescence intensity (MFI) were determined using FlowJo® software. Neutrophil degranulation was also evaluated by CatG release. The supernatants were collected and incubated with a cathepsin G (ABZ-TPFSGQ-YNO2 from GeneCust®) fluorescent substrate (Attucci S, et al. Measurement of free and membrane-bound cathepsin G in human neutrophils using new sensitive fluorogenic substrates. Biochem J. 2002 Sep. 15; 366(Pt 3):965-70. doi: 10.1042/BJ20020321. PMID: 12088507; PMCID: PMC1222843) for 30 minutes prior to reading the fluorescence spectrofluorimetry reading at 420 nm. The results are expressed as a report from ΔRFU compared to the normalized state from TNFα (basal degranulation).

[0323] Results

[0324] The first test consisted of evaluating the expression of membrane PR3 after incubation of neutrophils with 4C3 (FIG. 9). The 6H4 antibody, which is a non-relevant antibody, was used in comparison in the different tests and the PMA-Ionomycin calcium (PMA-ICa) solution was used as a positive control for neutrophil activation. Purified human neutrophils were pre-activated with TNFα for 15 minutes at 37° C. before being incubated for 1.5 hours with different concentrations of 6H4 and 4C3 antibodies (2, 20 and 100 μg/mL). At the end of incubation, membrane PR3 expression was analysed by flow cytometry with WGM2 antibody (murine anti-human PR3 antibody) (FIG. 9). The results showed that incubation of neutrophils with increasing doses of 4C3 leads to an increase in PR3 expression on the surface of neutrophils, whereas incubation with 6H4, regardless of the dose, has little or no effect (FIG. 9). The result was identical whether PR3 was revealed with a commercially available antibody or with 4C3-AF488.

[0325] The second functional test that was studied was the production of VSR, which was evaluated by DHR123-FITC marking by flow cytometry (FIGS. 10 and 11). For these experiments, total IgG from healthy donors or patients with GPA was also purified. Here, 4C3 was not able to induce ROS production, whereas separate IgG preparations from patients with active GPA (IgG GPA) led to a significant increase in ROS production. On the contrary, separate IgG preparations from healthy donors (IgG HD), obtained and used under the same conditions, did not induce a high production of ROS by pre-activated neutrophils (FIG. 10). It is interesting to note that the stimulation of neutrophils by purified IgG from patient P2 did not induce significant ROS production compared to purified IgG GPA from GPA patients at the time of diagnosis (FIG. 10). Furthermore, at a dose of 2 μg/mL, both 4C3 and 6H4 did not induce a significant increase in ROS production by neutrophils. By increasing the dose to 20 and 100 μg/mL, there was also no significant increase in the production of ROS by neutrophils, thus demonstrating that 4C3 in its whole form is not an activating antibody (FIG. 11).

[0326] The third test consisted of analysing neutrophil degranulation by measuring CD63 expression at the neutrophil membrane and cathepsin G activity in the neutrophil supernatant after incubation with antibodies (FIG. 12). The previously obtained result was confirmed since it was shown that there was a reciprocal absence of CD63 expression (FIG. 12A) and an absence of cathepsin G release in the presence of 4C3 or r4C3 at a dose of 2 and 20 μg/mL (FIG. 12B).

[0327] The fourth test consisted of analysing the expression of CD11b and CD18 (Mac1 complex) in order to explore the neutrophil adhesion phenotype after stimulation with 4C3 (FIG. 13). The presence of 4C3 did not induce an increase in CD11b/CD18 surface expression, whereas unmixed IgG preparations from healthy donors or patients with GPA at the time of diagnosis induced a significant upregulation of these two adhesion markers (FIGS. 13A and 13B). It should be noted that IgG from patient P2 induced an intermediate adhesion phenotype (FIGS. 13A and 13B).

[0328] All these results, obtained by functional tests on neutrophils, have shown that 4C3 and r4C3 antibodies do not activate human neutrophils.

[0329] Alongside these results, the effect of 4C3 derivatives was also studied. Before testing their functionality on neutrophils, the quality of these derived forms was verified by SDS-PAGE gel and Coomassie Blue staining (FIG. 14). For native 4C3, 6H4 and r4C3, a single migration band was observed around the molecular weight of 150 kDa, usually corresponding to the molecular weight of IgG1 (wells 2 to 4, FIG. 14). Quality and purity appear to be identical between 4C3, r4C3 and 6H4. Deposition of the Fab fragment revealed several bands: two bands around 50 kDa, one corresponding to the Fab fragment and the other to the Fc fragment which was not completely removed after purification on protein A-coupled resin and a third band around 25 kDa corresponding to the reduced Fab (6th well, FIG. 14). Deposition of the deglycosylated form revealed two bands: one slightly below 150 kDa corresponding to an IgG1 which had lost its glycosylation site and the other around 50 kDa corresponding to the Fc fragment (5th well, FIG. 14).

[0330] The expression of PR3 (FIG. 15A) and the neutrophil activation profile (FIGS. 15B and 16) in the presence of these derived forms were studied. No increase in membrane PR3 expression was observed after incubation of neutrophils with the Fab and F(ab′).sub.2 fragments of 4C3, unlike 4C3 in its whole form (FIG. 15A). In addition, the F(ab′).sub.2, Fab and degycosylated Fc (4C3 dFc) fragments did not induce SAR production by neutrophils at the 20 μg/mL dose (FIG. 15B).

[0331] Interestingly enough, it has been shown that r4C3 in its whole form is not capable of inducing a significant increase in SAR production and that this production is at the same level as that induced by TNFα: average MFI ration of the r4C3 condition=0.96±0.09 versus average ratio of the 4C3 condition=1.48±0.26 (p=0.0016) (FIG. 16), which corresponds to 0% increase in activation with r4C3.

[0332] In view of these results and with a view to obtaining a neutralising effect of the cANCA/PR3 interaction, the native and/or recombinant 4C3 in their whole form are the most interesting.

[0333] 15. Neutralisation Tests with 4C3

[0334] In order to evaluate the blocking potential of this antibody (i.e. 4C3) on the activation of neutrophils induced by ANCA (anti-PR3 auto-antibodies), different strategies have been put in place: [0335] functional tests were performed on neutrophils before and after incubation with the whole antibody or with its derivatives (Fab, F(ab′).sub.2, deglycosylated Fc, etc.); and [0336] a dose effect was evaluated and different incubation kinetics were tested.

[0337] To develop these experiments to neutralise cANCA, it was decided to use sera from healthy donors (blood samples obtained from the EFS) and sera from GPA patients (sera obtained after anonymisation in the immunology laboratory of the CHRU de Tours) in order to purify the IgG and in particular the anti-PR3 IgG (cANCA) of GPA patients. Following the purification step on human serum G protein, the different fractions obtained (not retained, washes and elutions) were deposited on a gel. SDS-PAGE in order to check their quality and purity in comparison with the original serum (FIG. 17). This confirmed that the purification step of the sera on protein G allowed IgG to be obtained with the presence of a majority band at 150 kDa (FIG. 17) while the wash fractions contain other proteins such as albumin (band at 50 kDa).

[0338] After successfully purifying total IgG from sera, it was verified that among the purified IgG from GPA patients, anti-PR3 cANCA type IgG was present. By ELISA, it was found that the purified IgG batches from GPA patient sera contained at least 20 RU (Relative Unit)/mL of anti-PR3 IgG (light grey histograms, before and after purification, FIG. 18) in comparison with the calibrator (cal), 4C3 and the positive control (dark grey histograms, FIG. 18). In addition, no anti-PR3 IgG was detected with this technique in the purified IgG batches from healthy donor sera (black histograms, before and after purification, FIG. 18). Although this semi-quantitative assay confirmed the presence of anti-PR3 IgG in purified IgG from GPA patients (a representative experiment of 6 batches is shown), it did not reveal the true proportion.

[0339] Purified IgG was used at 200 μg/mL as described in the literature. The functionality of purified IgG was evaluated on their ability to induce ROS production in human neutrophils from healthy donors pre-activated by TNFα (as previously) and after 45 minutes incubation. Purified IgG from GPA patients significantly (p=0.0002; n=32) induced greater ROS production (mean MFI ratio=4.54±4.73) than purified IgG from healthy donors (mean MFI ratio=1.79±1.49) (FIG. 19). In addition, purified IgG from GPA patients and purified IgG from healthy donors induced significantly higher ROS production than for TNFα and condition 4C3 (FIG. 19). This result therefore demonstrated that purified IgG from GPA patients can be used as a neutrophil activator to test the neutralising potential of 4C3 and other derivatives.

[0340] In order to test the neutralising action of 4C3 on the production of ROS, it was chosen to first incubate neutrophils with 4C3 for 15 minutes and then add purified IgG from GPA patients for a further 45 minutes. The ROS production was then analysed as described above. It was found that 4C3 antibody at a concentration of 20 μg/mL does not activate neutrophils when used alone and quite unexpectedly, 4C3 antibody alone is also able to inhibit ROS activation induced by the presence of IgG from GPA patients (FIG. 20). Indeed, percentages of inhibition ranging from 36% to 71% were observed (average of 51% neutralisation over 4 separate experiments), which could be improved by increasing the amounts of 4C3 and/or modifying the incubation kinetics.