Methods for determining the binding of antibodies to complement component 1q (C1q)

Abstract

The present invention relates to a novel method for determining the binding of antibodies to complement component 1q (C1q). The process according to the invention makes it possible to measure the binding of antibodies to C1q in a homogeneous proximity assay (HAS).

Description

[Technical field]

[0001] The present invention relates to a novel method for determining the binding of antibodies to complement component 1q (C1q). The process according to the invention makes it possible to measure the binding of antibodies to C1q in a homogeneous proximity assay (HAS). [Background technology]

[0002] Antibodies are immune proteins that bind to specific antigens. In most mammals, such as humans and mice, antibodies are constructed from paired heavy and light polypeptide chains. Each chain is composed of two distinct regions called the variable (Fv) and constant (Fc) regions. The light chain Fv region and the heavy chain Fv region contain the antigen-binding determinants of the molecule and are responsible for binding the target antigen. The Fc region defines the class (or isotype) of the antibody (e.g., IgG).

[0003] The Fc region interacts with several natural proteins, such as the Fc gamma receptor or complement component 1q (C1q), to trigger important biochemical events such as antibody-dependent cell-mediated cytotoxicity (ADCC) and / or complement-dependent cytotoxicity (CDC). Fc binding to C1q activates the complement cascade of the immune system, which reflects the ability of antibodies to mediate CDC.

[0004] The C1q binding site on IgG is the CH2 domain of the Fc region. Three charged residues on the CH2 domain, Glu-318, Lys-320, and Lys-322, are crucial for C1q binding to IgG. The C1q binding site on IgM is the CH3 domain of the Fc region.

[0005] C1q has six globular heads, each capable of binding (through its Fc region) to a single IgG molecule. Thus, C1q can bind six antibodies, but binding to two antibodies (e.g., two IgGs) is sufficient to activate the complement pathway. C1q forms a complex with C1r and C1s serine proteases to form the C1 complex of the complement pathway.

[0006] To activate the complement pathway, it is necessary for C1q to bind at least two antibodies (e.g., two IgGs). Therefore, in the prior art, solid phases are always used to assemble IgGs into suitable clusters that can bind C1q with high affinity, and then measure the ability of the antibodies to bind C1q. Solid phase methods commonly used in the prior art to evaluate the binding of antibodies to C1q include enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (SPR). There are three methods for clustering IgGs on a solid surface, either an ELISA plate or an SPR surface: (i) randomly immobilizing the antibodies on the surface (passive coating) [Reference 1], (ii) indirectly immobilizing the antibodies on the surface using protein L [Reference 2], or (iii) coating the antigen on the surface and then adding the antibodies [Reference 3] (Figures 26, 27, and 28).

[0007] However, existing solid-phase methods are not fully satisfactory because the hexameric IgG / C1q binding model has geometric constraints and it is difficult to predict the correct IgG clustering for high affinity binding to C1q [4]. Therefore, the results are not always reproducible and each data set often needs to be normalized.

[0008] Therefore, there is a need to develop an efficient, easily performed and reproducible method for determining the binding of an antibody (test antibody) to complement component 1q (C1q). Summary of the Invention [Problem to be solved by the invention]

[0009] The present invention aims to propose a novel in vitro method for determining the binding of an antibody (test antibody) to complement component 1q (C1q), i.e. for evaluating the ability of an antibody (test antibody) to mediate CDC. [Means for solving the problem]

[0010] According to a first aspect, the present invention provides an in vitro method for determining the binding of an antibody (test antibody) to complement component 1q (C1q), comprising the steps of: a) The measurement medium, A test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of the test antibody; Streptavidin, directly or indirectly labeled with a first member of a pair of HPA (Homogeneous Proximity Assay) partners, and C1q, directly or indirectly labeled with the second member of the HPA partner pair; and b) measuring an HPA signal in said measurement medium, the presence of which indicates binding of said test antibody to said C1q; The present invention relates to a method comprising the steps of:

[0011] According to a second aspect, the present invention provides a reagent kit for carrying out the method of the present invention, comprising: (i) a biotinylated anti-Fab ligand; (ii) streptavidin or streptavidin directly labeled with a first member of a pair of HPA partners; (iii) C1q or C1q directly labeled with a second member of the HPA partner pair; (iv) if the streptavidin is not directly labeled with the first member of the HPA partner pair, an anti-streptavidin ligand directly labeled with the first member of the HPA partner pair; (v) if the C1q is not directly labeled with the second member of the HPA partner pair, an anti-C1q ligand directly labeled with the second member of the HPA partner pair; The present invention relates to a kit comprising: [Brief description of the drawings]

[0012] [Figure 1] Figure 1 shows a TR-FRET sandwich assay performed to determine the binding of IgG to human C1q protein using a biotinylated anti-human Fab antibody and C1q indirectly labeled with a donor. The assay is based on HTRF® technology and uses an anti-C1q antibody conjugated to Eu3+ cryptate (donor) and streptavidin labeled with d2 (acceptor). When the test antibody binds to human C1q, it brings the donor into close proximity with the acceptor. Excitation of the donor by a light source (UV region) causes energy transfer towards the acceptor, followed by the emission of a specific fluorescence (TR-FRET signal). [Diagram 2] Figure 2 shows the results of binding of IgG antibodies of various isotypes to human C1q based on the TR-FRET method described in Figure 1. Sigmoidal dose-response curves represent log[antibody] (M) versus specific HTRF signal (delta ratio). A) Results obtained with recombinant human IgG isotype controls (IgG1, IgG2, and IgG4). B) Results obtained with the therapeutic antibody Rituximab (IgG1) and its isotype variants IgG2 and IgG4. C) Results obtained with the therapeutic antibody Cetuximab (IgG1) and its isotype variant IgG2. D) Results obtained with the therapeutic antibody Ipilimumab (IgG1) and its isotype variant IgG2. [Diagram 3]Figure 3 shows the results of binding of therapeutic antibodies and their non-glycosylated or non-fucosylated variants to human C1q based on the TR-FRET method described in Figure 1. Sigmoidal dose-response curves represent log[antibody](M) versus specific HTRF signal (delta ratio). A) Results obtained with rituximab (glycosylated antibody) and its non-fucosylated and non-glycosylated variants. B) Results obtained with cetuximab (glycosylated antibody) and its non-glycosylated variants. [Figure 4] Figure 4 shows the binding profiles of the therapeutic type I and type II anti-CD20 antibodies Rituximab and Obinutuzumab (GA101) to human C1q. A) Results obtained using the TR-FRET method described in Figure 1. Sigmoidal dose-response curves represent log [antibody] (M) versus specific HTRF signal (delta ratio). B) Data from the literature (adapted from [ref. 5]) obtained using an ELISA assay. Binding curves represent [antibody] (μg / mL) versus ELISA signal (OD at 405 nm). [Diagram 5] Figure 5 shows the results of the binding of therapeutic antibodies atezolizumab (IgG1) and spartalizumab (IgG4) to human C1q based on the TR-FRET method described in Figure 1. The sigmoidal dose-response curve represents log[antibody] (M) versus specific HTRF signal (delta ratio). [Figure 6] Figure 6 shows the results of binding of the therapeutic anti-TNF-α antibody adalimumab (pre-complexed with 150 nM human TNF-α or uncomplexed) to human C1q based on the TR-FRET method described in Figure 1. The sigmoidal dose-response curve represents log[antibody] (M) versus specific HTRF signal (delta ratio). [Figure 7]Figure 7 shows a TR-FRET sandwich assay performed to determine the binding of IgG to human C1q protein using a biotinylated anti-human Fab antibody and C1q directly labeled with the donor. The assay is based on HTRF® technology and uses human C1q conjugated to Tb3+ cryptate (donor) and streptavidin labeled with d2 (acceptor). When the test antibody binds to human C1q, it brings the donor into close proximity with the acceptor. Excitation of the donor by a light source (UV region) causes energy transfer towards the acceptor, followed by the emission of a specific fluorescence (TR-FRET signal). [Figure 8] Figure 8 shows the results of binding of IgG antibodies of various isotypes to human C1q based on the TR-FRET method described in Figure 7. The sigmoidal dose-response curves represent log[antibody] (M) versus specific HTRF signal (delta ratio). The antibodies tested are the therapeutic antibody Rituximab (IgG1) and the recombinant human IgG isotype controls IgG1, IgG2, and IgG4. [Figure 9] Figure 9 shows an ALPHA sandwich assay performed to determine the binding of IgG to human C1q protein using a biotinylated anti-human Fab antibody and C1q indirectly labeled with an acceptor. The assay is based on AlphaLISA® technology and uses an anti-C1q antibody conjugated to Alpha streptavidin-coated donor beads (donor) and AlphaLISA acceptor beads (acceptor). When the test antibody binds to human C1q, it brings the donor into close proximity with the acceptor. Excitation of the donor with a 680 nm laser causes the generation and diffusion of singlet oxygen to the acceptor, which then produces a specific fluorescence (ALPHA signal). [Figure 10]Figure 10 shows the results of binding of IgG antibodies of various isotypes to human C1q based on the ALPHA method described in Figure 9. The sigmoidal dose-response curves represent log[antibody](M) vs. ALPHA signal. The tested antibodies are the therapeutic antibody Rituximab (IgG1), and recombinant human IgG isotype controls IgG1, IgG2, and IgG4. [Figure 11] Figure 11 shows a TR-FRET sandwich assay performed to determine the binding of IgG to human C1q protein using biotinylated antigen and C1q indirectly labeled with donor. The assay is based on HTRF® technology and uses an anti-C1q antibody conjugated to Eu3+ cryptate (donor) and streptavidin labeled with d2 (acceptor). When the test antibody binds to human C1q, it brings the donor into close proximity with the acceptor. Excitation of the donor by a light source (UV region) causes energy transfer towards the acceptor, followed by the emission of a specific fluorescence (TR-FRET signal). [Figure 12] Figure 12 shows the results of the binding of the therapeutic anti-TNF-α antibody adalimumab to human C1q using biotinylated recombinant human TNF-α and based on the TR-FRET method described in Figure 11. The sigmoidal dose-response curve represents log[antibody] (M) versus specific HTRF signal (delta ratio). [Figure 13]FIG. 13 shows TR-FRET sandwich assays performed to determine the binding of IgG to human C1q protein using C1q indirectly labeled with donor and anti-human Fab antibody that is biotinylated and conjugated to acceptor-labeled streptavidin or directly conjugated to acceptor. These assays are based on HTRF® technology and use anti-C1q antibody conjugated to Eu3+ cryptate (donor) and A) biotinylated anti-human Fab antibody conjugated to streptavidin-d2 (acceptor) or B) the same anti-human Fab antibody directly labeled with d2. When the test antibody binds to human C1q, it brings the donor into close proximity to the acceptor. Excitation of the donor by a light source (UV region) causes energy transfer towards the acceptor, followed by the emission of specific fluorescence (TR-FRET signal). [Figure 14] Figure 14 shows the results of the binding of the therapeutic antibody rituximab to human C1q using the two different TR-FRET assay formats described in Figure 13. The sigmoidal dose-response curve represents log[antibody] (M) versus normalized HTRF signal (delta F%). [Figure 15] Figure 15 shows the TR-FRET sandwich assays used to perform saturation binding experiments and determine the affinity (Kd) of anti-human Fab antibodies for donor-labeled human IgG of various isotypes (IgG1, IgG2, and IgG4). These assays are based on HTRF® technology and use human IgG (IgG1, IgG2, or IgG4) labeled with Eu3+ cryptate (donor) and A) an anti-human Fab antibody directly labeled with d2 (acceptor), or B) the same anti-human Fab conjugated to biotin and complexed to streptavidin-d2. Binding of the anti-human Fab antibody to human IgG triggers the emission of a TR-FRET signal, which increases proportionally to the concentration of the anti-human Fab antibody tested until a plateau (saturation) is reached. [Figure 16]Figure 16 shows the results of binding of three different anti-human Fab antibodies (1, 2 and 3) to Eu3+cryptate-human IgG1 using two different TR-FRET sandwich assays described in Figure 15. The saturation binding curves represent [anti-human Fab antibody] (nM) versus HTRF signal (HTRF ratio). The total signal corresponds to the signal obtained in the absence of unlabeled human IgG1. The non-specific signal corresponds to the signal obtained in the presence of excess unlabeled IgG1. The specific signal was calculated by subtracting the non-specific signal from the total signal. A) Results obtained with anti-human Fab1 directly labeled with d2. B) Results obtained with biotinylated anti-human Fab1 complexed with streptavidin-d2. C) Results obtained with anti-human Fab2 directly labeled with d2. D) Results obtained with biotinylated anti-human Fab2 complexed with streptavidin-d2. E) Results obtained with anti-human Fab3 directly labeled with d2. F) Results obtained with biotinylated anti-human Fab3 complexed with streptavidin-d2. [Figure 17] Figure 17 shows the results of binding of three different anti-human Fab antibodies (1, 2 and 3) to Eu3+cryptate-human IgG2 using two different TR-FRET sandwich assays described in Figure 15. The saturation binding curves represent [anti-human Fab antibody] (nM) versus HTRF signal (HTRF ratio). The total signal corresponds to the signal obtained in the absence of unlabeled human IgG2. The non-specific signal corresponds to the signal obtained in the presence of excess unlabeled IgG2. The specific signal was calculated by subtracting the non-specific signal from the total signal. A) Results obtained with biotinylated anti-human Fab1 complexed with streptavidin-d2. B) Results obtained with biotinylated anti-human Fab2 complexed with streptavidin-d2. C) Results obtained with biotinylated anti-human Fab3 complexed with streptavidin-d2. [Figure 18]Figure 18 shows the results of binding of three different anti-human Fab antibodies (1, 2 and 3) to Eu3+cryptate-human IgG4 using two different TR-FRET sandwich assays described in Figure 15. The saturation binding curves represent [anti-human Fab antibody] (nM) versus HTRF signal (HTRF ratio). The total signal corresponds to the signal obtained in the absence of unlabeled human IgG4. The non-specific signal corresponds to the signal obtained in the presence of excess unlabeled IgG4. The specific signal was calculated by subtracting the non-specific signal from the total signal. A) Results obtained with anti-human Fab1 directly labeled with d2. B) Results obtained with biotinylated anti-human Fab1 complexed with streptavidin-d2. C) Results obtained with anti-human Fab2 directly labeled with d2. D) Results obtained with biotinylated anti-human Fab2 complexed with streptavidin-d2. E) Results obtained with anti-human Fab3 directly labeled with d2. F) Results obtained with biotinylated anti-human Fab3 complexed with streptavidin-d2. [Figure 19] Figure 19 shows the TR-FRET competition assays used to perform competitive binding experiments and determine the affinity (Ki) of anti-human Fab antibodies for fully human IgG, humanized IgG, or chimeric IgG of various isotypes (IgG1, IgG2, and IgG4). These assays are based on two different HTRF® sandwich assay formats shown in Figure 15. In the absence of competitor (unlabeled IgG), Eu3+cryptate-human IgG binds to the acceptor-anti-human Fab, causing the emission of a TR-FRET signal. In the presence of unlabeled IgG that can interact with the anti-human Fab antibody and compete with Eu3+cryptate-human IgG, the TR-FRET signal decreases proportionally to the concentration of the IgG tested. [Figure 20]Figure 20 shows the results of competition between Eu3+ cryptate-human IgG1 and various unlabeled IgG1 antibodies using three different anti-human Fab antibodies (1, 2, and 3) using one of the two TR-FRET competition assays described in Figure 19. Sigmoidal dose-response inhibition curves represent log[unlabeled IgG1](M) versus % of normalized HTRF signal maximum (= delta F % unlabeled IgG1 antibody / delta F % maximum x 100). A) Results obtained with biotinylated anti-human Fab1 complexed with streptavidin-d2. B) Results obtained with anti-human Fab2 directly labeled with d2. C) Results obtained with anti-human Fab3 directly labeled with d2. [Figure 21] Figure 21 shows the results of competition between Eu3+cryptate-human IgG2 and various unlabeled IgG2 antibodies with biotinylated anti-human Fab1 complexed with streptavidin-d2 as described in Figure 19. Sigmoidal dose-response inhibition curves represent log[unlabeled IgG2](M) versus % of normalized HTRF signal maximum (= delta F % unlabeled IgG1 antibody / delta F % max x 100). [Figure 22] Figure 22 shows the results of competition between Eu3+ cryptate-human IgG4 and various unlabeled IgG4 antibodies using three different anti-human Fab antibodies (1, 2, and 3) using one of the two TR-FRET competition assays described in Figure 19. Sigmoidal dose-response inhibition curves represent log[unlabeled IgG4](M) versus % of normalized HTRF signal maximum (= delta F % unlabeled IgG1 antibody / delta F % maximum x 100). A) Results obtained with biotinylated anti-human Fab1 complexed with streptavidin-d2. B) Results obtained with anti-human Fab2 directly labeled with d2. C) Results obtained with anti-human Fab3 directly labeled with d2. [Diagram 23]Figure 23 shows the results of binding of IgG1 and IgG2 antibodies to human C1q using various concentrations of biotinylated anti-human Fab antibody complexed with streptavidin-d2 (ratio 1 / 1) based on the TR-FRET method described in Figure 1. The sigmoidal dose-response curves represent log[antibody] (M) versus normalized HTRF signal (delta F%). A) Results obtained with the therapeutic antibody Atezolizumab (IgG1). B) Results obtained with the recombinant human IgG2 isotype control. [Figure 24] Figure 24 shows the results of the binding of the therapeutic type I and type II anti-CD20 antibodies Rituximab and Obinutuzumab to human C1q in the presence of various concentrations of NaCl based on the TR-FRET method described in Figure 1. The histograms represent the specific HTRF signal (delta ratio) obtained with 50 nM of each tested antibody in the presence of various concentrations of NaCl. [Diagram 25] Figure 25 shows the results of binding of different IgG antibodies to human C1q in the presence of different concentrations of NaCl based on the TR-FRET method described in Figure 1. The sigmoidal dose-response curves represent log[antibody] (M) versus normalized HTRF signal (delta F%). A) Results obtained in the presence of 155 mM NaCl. B) Results obtained in the presence of 135 mM NaCl. [Figure 26] FIG. 26 shows a method for determining the binding of antibodies to C1q by the prior art (ELISA) in which the antibodies are randomly immobilized on a surface (passive coating) [Reference 1]. [Figure 27] FIG. 27 shows a method for determining the binding of antibodies to C1q by the prior art (SPR), in which the antibodies are indirectly immobilized on a surface using protein L [Reference 2]. [Figure 28] FIG. 28 shows a method for determining the binding of antibodies to C1q by the prior art (ELISA) in which the antigen is coated on a surface and then the antibody is added [Reference 3]. [Figure 29]Figure 29 shows a TR-FRET sandwich assay performed to determine the binding of IgG to human C1q protein using a biotinylated anti-human Fab antibody conjugated to acceptor-labeled streptavidin (Figure 29A) or biotinylated protein L conjugated to acceptor-labeled streptavidin (Figure 29B) and C1q indirectly labeled with a donor. The assay is based on HTRF® technology and uses an anti-C1q antibody conjugated to Eu3+ cryptate (donor) and streptavidin labeled with d2 (acceptor). When the test antibody binds to human C1q, it brings the donor into close proximity with the acceptor. Excitation of the donor by a light source (UV region) causes energy transfer towards the acceptor, which then emits a specific fluorescence (TR-FRET signal). [Diagram 30] Figure 30 shows the results of trastuzumab or adalimumab binding to human C1q based on the TR-FRET method described in Figure 29. The sigmoidal dose-response curves represent log[antibody] (M) versus normalized HTRF signal (delta F%). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] definition The term "C1q" refers to complement component 1q. C1q is a protein involved in the complement system, which is part of the innate immune system. C1q forms the C1 complex together with C1r and C1s. The Fc portion of an antibody can bind to C1q to activate the complement pathway of the complement system. According to the present invention, C1q is preferably derived from the species of the Fc portion of the subject antibody. For example, if the subject antibody is a chimeric (i.e., having a human Fc portion), humanized, or human subject antibody, C1q is preferably human C1q.

[0014] In the sense of the present invention, the term "ligand" refers to a molecule capable of specifically and reversibly binding to a target molecule. In the context of the present invention, the target molecule is Fab, streptavidin, or C1q. The description therefore refers to "anti-Fab ligand", "anti-streptavidin ligand", or "anti-C1q ligand", respectively. The ligand may be proteinaceous (e.g., protein or peptide) or nucleotide (e.g., DNA or RNA). In the context of the present invention, the ligand is advantageously selected from an antibody, an antibody fragment, a protein, a peptide, or an aptamer, preferably an antibody or an antibody fragment. The ligands used in the methods of the present invention are capable of binding to their target molecule with sufficient affinity so that the ligand is useful as a diagnostic agent in targeting C1q.

[0015] An "antibody", also commonly referred to as an "immunoglobulin", comprises a heterotetramer composed of two heavy chains (heavy chains called H chains) of about 50-70 kDa each and two light chains (light chains called L chains) of about 25 kDa each, linked together by intra- and inter-chain disulfide bridges. At the N-terminal position, each chain is composed of a variable region or domain called VL for the light chain and VH for the heavy chain, and at the C-terminal position, the light chain is composed of a single domain called CL, and the heavy chain is composed of a constant region composed of three or four domains called CH1, CH2, CH3, CH4. Antibodies according to the invention may be of mammalian origin (e.g., human or murine, or camelid), humanized, chimeric, recombinant. Preferably, they are monoclonal antibodies recombinantly produced by genetically modified cells using techniques well known to those skilled in the art. The antibody can be of any isotype, for example, IgG, IgM, IgA, IgD, or IgE, and any subtype, for example, IgG1, IgG2, IgG3, or IgG4.

[0016] The term "antibody fragment" refers to any part of an immunoglobulin obtained by enzymatic digestion or by biotechnology that contains at least one disulfide bridge, including Fv, Fab, Fab', Fab'-SH, F(ab')2 The term refers to any portion of an immunoglobulin capable of binding to an antigen recognized by a whole antibody, such as a diabody, a single domain antibody (also known as an sdAb or nanobody), or a single chain antibody (e.g., scFv). Enzymatic digestion of immunoglobulins with pepsin results in F(ab') fragments split into several peptides. 2 F(ab') fragment and Fc fragment. 2 The Fab fragment is formed from two Fab' fragments linked by an interchain disulfide bridge. The Fab fragment contains the variable domains (V L and V H ), CH1 domain, and C L The Fab' fragment is formed by the Fab region and the hinge region. Fab'-SH refers to a Fab' fragment in which the cysteine ​​residue in the hinge region bears a free thiol group.

[0017] The term "antigen" refers to a molecule or molecular structure, e.g. a protein, that is bound by the antigen-binding site of an antibody or antibody fragment. Thus, according to the present invention, a "biotinylated antigen" that may be used in the method of the present invention is a molecule or molecular structure, e.g. a protein, that is bound by the antigen-binding site of a test antibody.

[0018] The term "biotinylated" according to the present invention means directly conjugated with a biotin molecule. According to the present invention, an anti-Fab ligand can be conjugated with a biotin molecule that reacts with the lateral NH2 group of a lysine or with the N-terminal NH2 group of an anti-Fab antibody, antibody fragment, or antigen. Biotinylation reagents and kits are commonly used in the art.

[0019] Streptavidin is a 52.8 kDa protein (tetramer) purified from the bacterium Streptomyces avidinii. The streptavidin homotetramer has a very high affinity for biotin. Approximately 10 -14With a dissociation constant (Kd) on the order of mol / L, the binding of biotin to streptavidin is one of the strongest non-covalent interactions known in nature. Streptavidin is widely used in molecular biology for the streptavidin-biotin complex.

[0020] The term "homogeneous assay" refers to an assay format that, unlike a heterogeneous assay, does not require processing the sample by separation, e.g., allows the assay-measurement to be performed by a mix-and-measure procedure without washing steps and / or without centrifugation steps.

[0021] The term "HPA" refers to "homogeneous proximity assay." HPA is well known in the art and can be defined as a homogeneous assay that measures a signal resulting from the proximity of a donor (hereinafter also referred to as a "donor compound") and an acceptor (hereinafter also referred to as an "acceptor compound").

[0022] The term "HPA partner pair" refers to a pair consisting of a donor (hereinafter also referred to as a "donor compound") and an acceptor (hereinafter also referred to as an "acceptor compound"), which emit an HPA signal when they are in close proximity to each other and the donor is excited.

[0023] The term "HPA signal" refers to any measurable signal indicative of the HPA between a donor compound and an acceptor compound.

[0024] The term "RET" refers to "resonance energy transfer." RET can be FRET or BRET.

[0025] The term "FRET" refers to "fluorescence resonance energy transfer". FRET is defined as a non-radiative energy transfer resulting from dipole-dipole interactions between a donor and an energy acceptor. This physical phenomenon requires energetic compatibility between these molecules. This means that the emission spectrum of the donor must at least partially cover the absorption spectrum of the acceptor. According to Förster's theory, FRET is a process that depends on the distance between two molecules, i.e., the donor and the acceptor, and when these molecules are in close proximity to each other, a FRET signal is emitted. FRET can be TR-FRET (time-resolved FRET).

[0026] The term "BRET" refers to "Bioluminescence Resonance Energy Transfer."

[0027] In the sense of the present invention, streptavidin and C1q are labeled with a member of the RET partner pair. Streptavidin and C1q can be directly or indirectly labeled by methods well known to those skilled in the art, for example as described below.

[0028] In certain embodiments, the streptavidin is directly labeled by covalent attachment to a member of the RET partner pair.

[0029] In another specific embodiment, C1q is indirectly labeled with an anti-C1q antibody labeled with a member of a RET partner pair. According to this embodiment of the present invention, the skilled artisan will understand that the biotinylated anti-Fab ligand in step a) should not bind to the Fab of the anti-C1q antibody. For example, the species of the Fab of the test antibody is different from the species of the Fab of the anti-C1q antibody, for example, the Fab of the test antibody is human and the Fab of the anti-C1q antibody is mouse.

[0030] The term "RET partner pair" refers to a pair consisting of an energy donor compound (hereinafter referred to as "donor compound") and an energy acceptor compound (hereinafter referred to as "acceptor compound"), which emit a RET signal when they are in close proximity to each other and excited with the excitation wavelength of the donor compound. It is known that for two compounds to be RET partners, the emission spectrum of the donor compound must partially cover the excitation spectrum of the acceptor compound. For example, when a fluorescent donor compound and an acceptor compound are used, a "FRET partner pair" is used, and when a donor bioluminescent compound and an acceptor compound are used, a "BRET partner pair" is used.

[0031] The term "RET signal" refers to any measurable signal that represents RET between a donor compound and an acceptor compound. Thus, for example, a FRET signal can be a variation in the intensity or lifetime of the fluorescence of a fluorescent donor compound or an acceptor compound (if the latter is fluorescent).

[0032] The term "LOCI" refers to "luminescent oxygen channeling assay." LOCI is an induced luminescence immunoassay described in U.S. Pat. No. 5,340,716, the entire contents of which are expressly incorporated herein by reference. LOCI technology involves a homogeneous assay (i.e., no separation step) with high sensitivity, which uses several reagents and requires the proximity of two of these reagents (called "donor bead" and "acceptor bead"), which are held by other immunoassay reagents, to obtain a signal. When exposed to light of a certain wavelength, the donor bead releases singlet oxygen, which is transferred to the acceptor bead when the two beads are in close proximity, which causes a chemical reaction that causes the acceptor bead to emit light that can be measured at a different wavelength.

[0033] The term "LOCI partner pair" refers to a pair of singlet oxygen donor compounds (hereinafter referred to as "donor beads") and singlet oxygen acceptor compounds (hereinafter referred to as "acceptor beads") that emit a LOCI signal when they are in close proximity to each other and excited at the excitation wavelength of the donor beads. The term "LOCI signal" refers to any measurable signal that represents singlet oxygen transfer between the donor bead and the acceptor bead. Thus, for example, the LOCI signal can be the amount of variation in the intensity of light emitted by the acceptor bead.

[0034] The term "vessel" refers to a well of a plate, a test tube, or other suitable container for mixing the membrane preparation with the reagents necessary for carrying out the method according to the invention.

[0035] Methods for determining binding of antibodies to C1q According to a first aspect, the present invention relates to an in vitro method for determining the binding of an antibody (test antibody) to complement component 1q (C1q), i.e. for assessing the ability of an antibody (test antibody) to mediate CDC.

[0036] According to a first aspect, the present invention provides an in vitro method for determining the binding of an antibody (test antibody) to complement component 1q (C1q), comprising the steps of: a) The measurement medium, a test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of a test antibody; streptavidin, directly or indirectly labeled with a first member of a pair of HPA (Homogeneous Proximity Assay) partners, and C1q directly or indirectly labeled with the second member of the HPA partner pair and b) measuring an HPA signal in the measurement medium, the presence of which indicates binding of the test antibody to C1q; The present invention relates to a method comprising the steps of:

[0037] Process a) According to the invention, step a) comprises the steps of: a test antibody, preferably an IgG or IgM test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of a test antibody; streptavidin, directly or indirectly labeled with a first member of a pair of HPA (Homogeneous Proximity Assay) partners, and C1q directly or indirectly labeled with the second member of the HPA partner pair The method comprises the step of contacting the

[0038] The measurement medium may be contained in a container. The various elements may be introduced into the measurement medium sequentially in any order, or simultaneously or nearly simultaneously. For example, the various elements are introduced into the measurement medium in the following order: a biotinylated anti-Fab ligand, streptavidin directly or indirectly labeled with a first member of the HPA partner pair, a test antibody, and C1q directly or indirectly labeled with a second member of the HPA partner pair.

[0039] The mixing of the elements makes it possible to obtain a reaction solution suitable for carrying out the HPA. Therefore, other elements can be added to the measurement medium to make the solution suitable for carrying out the HPA. For example, the measurement medium can also contain a buffer, an antigen of the antibody to be tested (assuming that the antigen should not be biotinylated), and / or any compound suitable for carrying out the method of the invention, such as albumin, surfactants, preservatives.

[0040] In some embodiments, the measurement medium has an osmolality adapted to detect HPA signals.Preferably, the measurement medium has an osmolality of 250mOsm / L to 500mOsm / L, preferably 250mOsm / L to 350mOsm / L, such as 275mOsm / L to 325mOsm / L, such as 275mOsm / L to 310mOsm / L, such as 280mOsm / L to 300mOsm / L, such as about 282mOsm / L.

[0041] In some embodiments, the measurement medium comprises a sodium buffer, such as NaCl, and / or a phosphate buffer, such as Na2HPO4 and / or KH2PO4, in an amount sufficient to detect the HPA signal. Advantageously, the measurement medium may comprise a sodium buffer (e.g. NaCl) of 50 mM to 250 mM, such as 100 mM to 200 mM, such as 125 mM to 150 mM, such as 135 mM. A suitable NaCl concentration for carrying out the method of the invention can be easily determined by the skilled person, for example as described in Example 9.

[0042] In a preferred embodiment, the pH of the measurement medium is suitable for detecting the HPA signal. Advantageously, the pH is between pH 6.0 and pH 8.0, such as between pH 7.0 and pH 8.0, such as between pH 7.2 and pH 7.6, such as pH 7.4. A person skilled in the art can easily determine the pH suitable for carrying out the method of the present invention.

[0043] In a preferred embodiment, the measurement medium comprises (ii) a phosphate buffer, such as Na2HPO4 and / or KH2PO4, (ii) a sodium buffer, such as NaCl, and finally (iii) albumin.

[0044] The measurement medium may also contain a surfactant, such as Tween-20, and / or a preservative, such as ProClin-300.

[0045] In a very specific embodiment, the measurement medium comprises: Na2HPO4 3 mM, KH2PO4 1 mM, NaCl 135 mM, BSA (preferably protease-free and IgG-free) 0.1%, Tween-20 0.05%, ProClin-300 0.01%, pH 7.4.

[0046] According to the present invention, the test antibody can be IgG or IgM. IgG and IgM are known to bind to C1q. Preferably, the test antibody is IgG, e.g., IgG1, IgG2, or IgG4. In some embodiments, the test antibody is IgG1 or IgG2.

[0047] The subject antibody can be a chimeric antibody, a humanized antibody, or a human antibody.

[0048] According to the invention, the biotinylated anti-Fab ligand is capable of binding to the Fab region of the test antibody. As explained above, the anti-Fab ligand is advantageously selected from an anti-Fab antibody, an anti-Fab antibody fragment, an anti-Fab peptide, an anti-Fab aptamer, or an antigen. Even though the anti-Fab antibody can be an anti-Fab polyclonal antibody, it is preferred that the anti-Fab antibody is an anti-Fab monoclonal antibody.

[0049] In one embodiment, the biotinylated anti-Fab ligand is a C L Preferably, the anti-Fab ligand binds to the CH1 region and / or the CH2 region. In this embodiment, the anti-Fab ligand is an anti-Fab antibody or an anti-Fab antibody fragment.

[0050] In another embodiment, the biotinylated anti-Fab ligand is a Fab ligand that binds to the variable region of the test antibody, e.g., the V H and / or V L In this embodiment, the anti-Fab ligand is preferably an antigen.

[0051] It would not be difficult for a person skilled in the art to select a suitable biotinylated anti-Fab ligand depending on the test antibody. For example, when the test antibody is a human antibody, a humanized antibody, or a chimeric antibody, the biotinylated anti-Fab ligand is a C of the test antibody. L The biotinylated anti-Fab ligand may bind to the CH1 region and / or the CH2 region. Thus, the biotinylated anti-human Fab ligand may be a biotinylated anti-human Fab ligand, such as a biotinylated anti-human Fab antibody, a biotinylated anti-human Fab antibody fragment, a biotinylated anti-human Fab peptide, a biotinylated anti-human Fab aptamer, or a biotinylated antigen.

[0052] C of the test antibody L Anti-Fab ligands that bind to the C1 and / or CH1 regions of a test antibody, e.g. L Standard methods for obtaining anti-Fab antibodies that bind to the domain and / or CH1 are widely disclosed in the prior art, for example [Reference 8].

[0053] Both biotinylated and non-biotinylated anti-Fab ligands, for example biotinylated and non-biotinylated anti-Fab antibodies, are commercially available. Examples of biotinylated and non-biotinylated anti-Fab ligands are biotinylated recombinant human TNF-α protein from Abcam (#ab167747), recombinant human TNF-α protein from R&D Systems (#10291-TA), anti-human IgG Fab (mouse IgG2b, monoclonal antibody, clone 4A11) from ThermoFisher Scientific (#SA1-19255), anti-human IgG Fab (mouse IgG2b, monoclonal antibody, clone 2A11) from GeneTex (#GTX27497), anti-human IgG Fab (goat IgG, polyclonal antibody) from Sigma-Aldrich (#I5260), biotinylated anti-human Ig (IgG, IgM, and IgA) Fab (goat IgG, polyclonal antibody) from SouthernBiotech (#2085-08), biotinylated anti-human IgG from Sigma-Aldrich. Fab (goat IgG, polyclonal antibody) (#SAB3701251), biotinylated anti-human IgG Fab (goat IgG, polyclonal antibody) (#109-065-006) from Jackson ImmunoResearch Inc., and biotinylated anti-human IgG Fab (chicken IgY, polyclonal antibody) (#SA1-72044) from ThermoFisher Scientific. If the anti-Fab ligand is not biotinylated, it will be easy for the skilled artisan to biotinylate it using standard procedures known in the art, for example, anti-human IgG Fab (mouse IgG2b, monoclonal antibody, clone 4A11) (#SA1-19255) from ThermoFisher Scientific, anti-human IgG Fab (mouse IgG2b, monoclonal antibody, clone 2A11) (#GTX27497) from GeneTex, or anti-human IgG Fab (goat IgG, polyclonal antibody) (#I5260) from Sigma-Aldrich. Thus, it will be easy for the skilled artisan to obtain a biotinylated anti-Fab ligand for each and every antibody to be tested.

[0054] In a particular embodiment, the biotinylated anti-Fab ligand is a mouse biotinylated anti-Fab antibody or antibody fragment, such as a mouse biotinylated anti-Fab antibody or antibody fragment. The applicant has shown that the method of the present invention can be carried out using a mouse biotinylated anti-Fab antibody or antibody fragment. In fact, the applicant has shown that a mouse biotinylated anti-Fab antibody or antibody fragment does not interfere with the binding of the test antibody to C1q and therefore does not change the RET signal in step b).

[0055] Preferably, the biotinylated anti-Fab ligand is in excess compared to the test antibody. For example, as described in Example 8, it will be easy for a person skilled in the art to identify a suitable concentration of biotinylated anti-Fab ligand for carrying out the method of the present invention. In some embodiments, the concentration of biotinylated anti-Fab ligand can be 10-200 times, such as 50-150 times, such as 100 times, the dissociation constant (Kd) in nM of the biotinylated anti-Fab ligand for the test antibody. For example, if the Kd is 0.5 nM, the concentration of biotinylated anti-Fab ligand is 50 nM. For example, as described in Example 7, it will be easy for a person skilled in the art to determine the Kd of the biotinylated anti-Fab ligand for the test antibody.

[0056] The streptavidin and C1q can be directly or indirectly labeled.

[0057] Direct labeling of streptavidin and C1q with members of the HPA partner pair, e.g., fluorescent compounds in the case of HPA FRET, can be carried out by conventional methods known to those skilled in the art, based on the presence of reactive groups on streptavidin and C1q. For example, the following reactive groups can be used: terminal amino group, carboxylate groups of aspartic acid and glutamic acid, amino group of lysine, guanidine group of arginine, thiol group of cysteine, phenol group of tyrosine, indole ring of tryptophan, thioether group of methionine, imidazole group of histidine. Streptavidin and C1q can also be indirectly labeled, e.g., by introducing into the measurement medium an antibody or antibody fragment that is itself covalently bound to an acceptor / donor compound, this second antibody or antibody fragment specifically recognizing streptavidin or C1q. Clearly, it is important that the indirect labelling of streptavidin or C1q with a member of the HPA partner pair does not involve biotinylation of streptavidin, C1q, or a member of the HPA partner pair.

[0058] Advantageously, the streptavidin is directly labelled with a first member of the HPA partner pair and the C1q is indirectly labelled with an anti-C1q ligand, such as an anti-C1q antibody or antibody fragment, labelled with a second member of the HPA partner pair.

[0059] In some embodiments, (i) a first member of the HPA partner pair is an acceptor and a second member of the HPA partner pair is a donor, or (ii) a first member of the HPA partner pair is a donor and a second member of the HPA partner pair is an acceptor.

[0060] It will not be difficult for a person skilled in the art to identify a suitable concentration ratio [streptavidin]:[biotinylated anti-Fab ligand] for carrying out the method of the present invention. For example, the concentration ratio [streptavidin]:[biotinylated anti-Fab ligand] may be about 1:1.

[0061] Step b) Step b) consists of measuring an HPA signal in a measurement medium, the presence of which indicates binding of the test antibody to C1q.

[0062] Herein, the term "presence" in step b) can be replaced with the term "intensity". Those skilled in the art understand that an increase in the binding of the test antibody to C1q increases the intensity of the HPA signal.

[0063] The measured signal corresponds to the signal obtained in the measurement medium in the presence of the test antibody. Measurement can be performed by conventional methods well known to those skilled in the art, without any particular problems. To detect and measure the HPA signal, devices such as the PHERAstar FS microplate reader (BMG Labtech) using TR-FRET or the VICTOR Nivo (PerkinElmer) using ALPHA are usually used.

[0064] In a particular embodiment, the HPA signal of the test antibody is compared with the HPA signal obtained with another antibody, such as a standard antibody or a reference antibody. In this particular embodiment, the HPA signal of the other antibody is obtained by carrying out the method of the invention using another antibody, i.e. by replacing the test antibody with the other antibody. The other antibody may be a standard antibody whose level of HPA signal is already known, or it may be a reference antibody to which the test antibody is compared. The comparison is particularly interesting when the test antibody is a biosimilar antibody and the reference antibody is a princeps antibody. In this particular embodiment, the HPA signal in the measurement medium containing the test antibody can be carried out in a first container, and the HPA signal in the measurement medium containing the other antibody can be carried out in a second container, and then the HPA signals obtained in each of the two containers are compared.

[0065] In some embodiments, steps (a) and (b) are repeated with various concentrations of the test antibody, preferably the method further comprising: the dissociation constant (Kd) of the binding of the test antibody (or other antibody) to C1q, and / or EC20 binding of the test antibody (or other antibody) to C1q 50 and (c) determining

[0066] According to the present invention, the HPA can be selected from (i) chemically amplified luminescence oxygen channeling immunoassays (LOCI), such as chemically amplified luminescence proximity homogeneous assays (ALPHA), (ii) resonance energy transfer (RET), such as fluorescence resonance energy transfer (FRET), and (iii) spatial proximity analytical reagent capture luminescence (SPARCL). Thus, in this specification, the term "HAS" can be replaced with "RET", "LOCI", or "SPARCL". Specific embodiments of "RET", "LOCI", and "SPARCL" are detailed below.

[0067] SPARCL method SPARCL technology is a proximity-dependent, non-separation chemiluminescent detection method. In the SPARCL assay, a chemiluminescent substrate (acridan) is brought into proximity with an oxidase (horseradish peroxidase, HRP) by a specific antigen / antibody interaction. Addition of a trigger solution containing H2O2 and parahydroxycinnamic acid (pHCA) produces a flash of light that is proportional to the amount of analyte present in the sample.

[0068] RET Method When the HPA is RET, the invention therefore relates to an in vitro method for or for determining the binding of an antibody (test antibody) to complement component 1q (C1q), comprising: a) The measurement medium, a test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of a test antibody; streptavidin, directly or indirectly labeled with a first member of a pair of RET (resonance energy transfer) partners, and C1q directly or indirectly labeled with the second member of the RET partner pair and b) measuring a RET signal in the measurement medium, the presence of which indicates binding of the test antibody to C1q; The present invention relates to a method comprising the steps of:

[0069] Other elements can be added to the measurement medium to make the solution suitable for carrying out RET. For example, coelenterazine h (benzyl-coelenterazine) or bisdeoxycoelenterazine (DeepBlueC TM ) or dihydrocoelenterazine (coelenterazine-400a) or D-luciferin can be added.

[0070] Streptavidin and C1q labeling with members of the RET partner pair The reactive group can form a covalent bond with a reactive group on a member of a RET partner pair. Suitable reactive groups on members of a RET partner pair are well known to those of skill in the art, for example, a donor or acceptor compound functionalized with a maleimide group could be covalently bonded to a thiol group on a cysteine ​​on a protein or peptide, such as streptavidin or C1q. Similarly, a donor / acceptor compound with an N-hydroxysuccinimide ester could be covalently bonded to an amine containing protein or peptide.

[0071] In the context of the present invention, streptavidin and C1q may each be labeled with a member of a RET partner pair, one member of which is a fluorescent donor or luminescent donor compound and the other member of the pair is a fluorescent acceptor compound or a non-fluorescent acceptor compound (quencher).

[0072] Labeling for FRET In certain embodiments, RET is FRET. Thus, streptavidin and C1q are each labeled with a member of a FRET partner pair, i.e. a fluorescent donor compound or a fluorescent energy acceptor compound.

[0073] The choice of FRET partner pairs to obtain FRET signals is within the knowledge of those skilled in the art. For example, donor-acceptor pairs that can be used to study the FRET phenomenon are described in the work by Joseph R. Lakowicz (Principles of fluorescence spectroscopy, 2nd edition 338), which those skilled in the art can refer to.

[0074] Fluorescent Donor Compounds Long-lived energy-donating fluorescent compounds (greater than 0.1 m, preferably in the range of 0.5-6 ms), especially rare earth chelates or cryptates, are advantageous because they allow time-resolved FRET without the need to deal with a large part of the background noise emitted by the measurement medium. For this reason, they are generally preferred for carrying out the process according to the invention. Advantageously, these compounds are lanthanide complexes. These complexes (such as chelates or cryptates) are particularly suitable as members of an energy-donating FRET pair.

[0075] Europium (Eu 3+ ), Terbium (Tb 3+ ), or samarium (Sm 3+ ) complexes are also suitable rare earth complexes for the present invention, and europium (Eu 3+ ) and terbium (Tb 3+ ) complexes are particularly preferred.

[0076] In one embodiment, the lanthanide complex Ln 3+ is selected from one of the following complexes: [ka] [ka]

change

change

change

change

change

change

change

change

change

change

change

change

change

change

change

change

change

change

[0077] Depending on the pH, the -SO3H, -CO2H, and -PO(OH)2 groups are either in the deprotonated form or not.- , -CO2 - , and -PO(OH)O - Also represents.

[0078] The lanthanide complexes C1 to C90 are described in the following publications or patents: These complexes are commercially available or can be obtained by synthetic routes as described in the prior art, such as WO 2020 / 157439 A1.

[0079] Advantageously, the fluorescent donor compound is a FRET partner selected from europium cryptates, europium chelates, terbium chelates, terbium cryptates, ruthenium chelates, quantum dots, allophycocyanins, rhodamines, cyanines, squaraines, coumarins, proflavines, acridines, fluoresceins, boron-dipyrromethene derivatives, and nitrobenzoxadiazoles.

[0080] Particularly advantageously, the fluorescent donor compound is a FRET partner selected from europium cryptates, europium chelates, terbium chelates, terbium cryptates, ruthenium chelates and quantum dots, with europium and terbium chelates and cryptates being particularly preferred.

[0081] Fluorescent Acceptor Compounds The fluorescent acceptor compounds may be selected from the following groups: allophycocyanins, in particular those known under the trade name XL665, luminescent organic molecules, such as rhodamines, cyanines (such as Cy5), squaraines, coumarins, proflavines, acridines, fluoresceins, boron-dipyrromethene derivatives (commercially available as "Bodipy"), the fluorophores known as "Atto", the fluorophores known as "DY", the compounds known as "Alexa", nitrobenzoxadiazoles. Advantageously, the fluorescent acceptor compounds are selected from allophycocyanins, rhodamines, cyanines, squaraines, coumarins, proflavines, acridines, fluoresceins, boron-dipyrromethene derivatives, nitrobenzoxadiazoles.

[0082] The terms "cyanine" and "rhodamine" should be understood as "cyanine derivatives" and "rhodamine derivatives", respectively. The skilled artisan is familiar with the different fluorophores available on the market.

[0083] The "Alexa" compounds are marketed by Invitrogen, the "Atto" compounds are marketed by Attotec, the "DY" compounds are marketed by Dyomics, the "Cy" compounds are marketed by Amersham Biosciences, and other compounds are marketed by various chemical reagent suppliers such as Sigma, Aldrich, or Acros.

[0084] The following fluorescent proteins can also be used as fluorescent acceptor compounds: cyan fluorescent protein (AmCyan1, Midori-Ishi Cyan, mTFP1), green fluorescent protein (EGFP, AcGFP, TurboGFP, Emerald, Azami Green, ZsGreen), yellow fluorescent protein (EYFP, Topaz, Venus, mCitrine, YPet, PhiYFP, ZsYelllow1, mBanana), orange and red fluorescent proteins (Orange kusibari, mOrange, tdtomato, DsRed, DsRed2, DsRed-Express, DsRed-Monomer, mTangerine, AsRed2, mRFP1, JRed, mCherry, mStrawberry, HcRed1, mRaspberry, HcRed-Tandem, mPlim, AQ143), far-red fluorescent proteins (mKate, mKate2, tdKatushka2).

[0085] Advantageously, the fluorescent acceptor compound is an allophycocyanin, a rhodamine, a cyanine, a squaraine, a coumarin, a proflavine, an acridine, a fluorescein, a boron-dipyrromethene derivative, a nitrobenzoxadiazole, and a FRET partner selected from quantum dots, a GFP variant selected from GFP, GFP10, GFP2, and eGFP, a YFP variant selected from YFP, eYFP, YFP topaz, YFP citrine, YFP venus, and YPet, mOrange, DsRed.

[0086] ● Signs for conducting BRET In certain embodiments, streptavidin and C1q are each labeled with a member of a BRET partner pair, i.e., a luminescent donor compound or a fluorescent energy accepting compound.

[0087] The direct labeling of streptavidin and C1q with a luminescent donor compound or a protein-type fluorescent acceptor compound, which are members of a BRET partner pair, can be carried out according to classical methods known to those skilled in the art, in particular the article by Tarik Issad and Ralf Jockers (Bioluminescence Resonance Energy Transfer to Monitor Protein-Protein Interactions, Transmembrane Signaling Protocols pp 195-209, Part of the Methods in Molecular Biology 2010), to which the skilled artisan may refer. TM This can be done by the classical method described in the book series MIMB, Vol. 332.

[0088] Direct labeling of streptavidin and C1q with organic molecule-type fluorescent acceptor compounds, which are members of a BRET partner pair, can be performed by classical methods known to those skilled in the art, based on the presence of reactive groups on the aforementioned ligands. For example, the following reactive groups can be used: terminal amino group, carboxylate groups of aspartic acid and glutamic acid, amino group of lysine, guanidine group of arginine, thiol group of cysteine, phenol group of tyrosine, indole ring of tryptophan, thioether group of methionine, imidazole group of histidine.

[0089] The reactive group can form a covalent bond with a reactive group on a member of a BRET partner pair. Suitable reactive groups on members of a BRET partner pair are well known to those of skill in the art, for example, an acceptor compound functionalized with a maleimide group could be covalently bonded to a thiol group on a cysteine ​​on a protein or peptide, such as C1q. Similarly, an acceptor compound having an N-hydroxysuccinimide ester could be covalently bonded to an amine containing protein or peptide.

[0090] The selection of BRET partner pairs to obtain BRET signal is within the knowledge of those skilled in the art.For example, the donor-acceptor pairs that can be used to investigate the BRET phenomenon are described in particular in the article by Dasiel O. Borroto-Escuela (BIOLUMINESCENCE RESONANCE ENERGY TRANSFER (BRET) METHODS TO STUDY G PROTEIN-COUPLED RECEPTOR-RECEPTOR TYROSINE KINASE HETERORECEPTOR COMPLEXES, Cell Biol.2013;117:141-164), which those skilled in the art can refer to.

[0091] Light-emitting donor compounds In certain embodiments, the light emitting donor compound is a BRET partner selected from luciferase (luc), Renilla luciferase (Rluc), a mutant of Renilla luciferase (Rluc8), and firefly luciferase.

[0092] Fluorescent Acceptor Compounds In certain embodiments, the fluorescent acceptor compound is a BRET partner selected from allophycocyanin, rhodamine, cyanine, squaraine, coumarin, proflavine, acridine, fluorescein, boron-dipyrromethene derivatives, nitrobenzoxadiazole, quantum dot, GFP, GFP variants (GFP10, GFP2, eGFP), YFP, YFP variants (eYFP, YFP topaz, YFP citrine, YFP venus, YPet), mOrange, DsRed.

[0093] As explained above, streptavidin and / or C1q may also be indirectly labeled with a member of the RET partner pair, such as by introducing into the measurement medium an antibody or antibody fragment which is itself covalently bound to an acceptor / donor compound, and which antibody or antibody fragment specifically recognizes streptavidin or C1q.

[0094] LOCI method In the case where HPA is LOCI, the present invention therefore provides an in vitro method for determining the binding of an antibody (test antibody) to complement component 1q (C1q), comprising: a) The measurement medium, a test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of a test antibody; Streptavidin directly or indirectly labeled with a first member of a pair of LOCI (Luminescent Oxygen Channeling Immunoassay) partners, C1q directly or indirectly labeled with the second member of the LOCI partner pair and b) measuring a LOCI signal in the measurement medium, the presence of which indicates binding of the test antibody to C1q; The present invention relates to a method comprising the steps of:

[0095] There are several LOCI technologies used to investigate biomolecular interactions in microplate format, such as chemically amplified luminescence proximity homogeneous assay (ALPHA). For example, ALPHA kits are manufactured by PerkinElmer, Waltham, Massachusetts, and are marketed under the trademarks AlphaScreen® and AlphaLISA®. These technologies are non-radioactive homogeneous proximity assays. Binding of molecules captured on beads results in singlet oxygen diffusion from one bead to the other, ultimately resulting in a detectable luminescence / fluorescence signal, providing qualitative and quantitative information about one or more analytes in the sample.

[0096] A pair of LOCI partners includes two bead types: donor beads and acceptor beads. The donor beads contain a photosensitizer, e.g., phthalocyanine, which upon irradiation at 680 nm converts ambient oxygen into singlet oxygen, an excited reactive form of oxygen. Singlet oxygen is not a radical, but an oxygen molecule with a single excited electron. Like other excited molecules, singlet oxygen has a limited lifetime before returning to the ground state. Within its 4 μs half-life, singlet oxygen can diffuse about 200 nm in solution, in comparison, TR-FRET has a maximum migration distance of about 10 nm. If an acceptor bead is in its vicinity, energy is transferred from the singlet oxygen to a thioxene derivative in the acceptor bead, which then leads to light generation within a wavelength range, e.g., 520-620 nm (AlphaScreen®) or at a specific wavelength, e.g., 615 nm (AlphaLISA®). If no acceptor bead is present, the singlet oxygen drops to the ground state and no signal is generated. This proximity-dependent chemical energy transfer is the basis for the homogeneity of LOCI; unlike ELISA, electrochemiluminescence, and flow cytometry assays, no washing steps are required, thereby providing a significant advantage.

[0097] The acceptor beads have three dyes embedded in them: thioxene, anthracene, and rubrene. The final fluorescent dye, rubrene, emits light detectable between 520 and 620 nm (e.g., AlphaScreen®).

[0098] Anthracene and rubrene may be substituted with europium chelates (e.g. AlphaLISA®). The europium (Eu) chelate is directly excited by 340 nm light resulting from the conversion of thioxene to a diketone derivative after reaction with singlet oxygen. The excited europium chelate produces intense light detectable within a much narrower wavelength band centered at about 615 nm. Thus, in contrast to acceptor beads not substituted with europium chelates, acceptor beads substituted with europium chelate emission is less susceptible to interference by artificial or natural compounds (such as hemoglobin) that absorb light between 500 and 600 nm.

[0099] Streptavidin and C1q labeling by members of the LOCI partner pair Direct labeling of streptavidin and C1q with members of a pair of LOCI partners can be performed by conventional methods known to those skilled in the art, based on the presence of reactive groups on streptavidin and C1q, for example the following reactive groups can be used: the N-terminal amino group, the carboxylate groups of aspartic acid and glutamic acid, the amino group of lysine, the thiol group of cysteine.

[0100] The reactive group can form a covalent bond with a reactive group on a member of a LOCI partner pair. Suitable reactive groups on members of a LOCI partner pair are well known to those of skill in the art, for example, a donor or acceptor compound functionalized with a maleimide group could be covalently bonded to a thiol group on a cysteine ​​on a protein or peptide, such as streptavidin or C1q. Similarly, a donor / acceptor compound with an N-hydroxysuccinimide ester could be covalently bonded to an amine containing protein or peptide.

[0101] ALPHA beads (donor or acceptor) can be directly conjugated to peptides, proteins (e.g., streptavidin or C1q), or antibodies (e.g., anti-C1q antibodies). This conjugation is based on a reductive amination reaction between reactive aldehyde groups present on the surface of the ALPHA beads and free amine groups (lysine and N-terminal) of the molecule of interest. This reaction is carried out in the presence of sodium cyanoborohydride (NaBH3CN), which stabilizes the bond formed, and is then stopped by the addition of carboxymethylamine (CMO), which blocks the remaining free aldehyde groups on the ALPHA beads.

[0102] As explained above, streptavidin and / or C1q may also be indirectly labeled with a member of the LOCI partner pair, such as by introducing into the measurement medium an antibody or antibody fragment which is itself covalently bound to an acceptor / donor compound, and which antibody or antibody fragment specifically recognizes streptavidin or C1q.

[0103] Reagent kit for carrying out the method of the present invention According to a second aspect, the present invention provides a reagent kit for carrying out the method of the present invention, comprising: (i) a biotinylated anti-Fab ligand; (ii) streptavidin or streptavidin directly labeled with a first member of a pair of HPA partners; (iii) C1q or C1q directly labeled with a second member of the HPA partner pair; (iv) an anti-streptavidin ligand directly labeled with a first member of the HPA partner pair, if the streptavidin is not directly labeled with the first member of the HPA partner pair; (v) if C1q is not directly labeled with the second member of the HPA partner pair, an anti-C1q ligand directly labeled with the second member of the HPA partner pair; The present invention relates to a kit comprising:

[0104] The particular embodiments of the various components of the kit of the invention, as detailed in the preceding description, also apply to the kit of the invention.

[0105] In a particularly preferred embodiment, the reagent kit for carrying out the method of the present invention comprises: (i) a biotinylated anti-Fab antibody; (ii) a first member of an HPA partner pair, preferably a first member of a TR-FRET partner pair, e.g., streptavidin directly labeled with d2; (iii) C1q; (iv) a second member of the HPA partner pair, preferably a second member of the TR-FRET partner pair, e.g., an anti-C1q antibody directly labeled with europium cryptate; Includes.

[0106] In a preferred embodiment, the kit also contains a buffer, an antigen of the test antibody (assuming the antigen is not to be biotinylated), and / or any compound suitable for carrying out the method of the invention, e.g., albumin, a detergent, and / or a preservative, as disclosed herein above.

[0107] The reagents of the kit are contained in one or more containers, preferably multiple containers. EXAMPLES

[0108] Example 1: Method for determining binding of test antibodies to C1q using biotinylated anti-human Fab antibody and C1q indirectly labeled with donor (TR-FRET)

[0109] To determine the binding of the test antibodies to C1q, a TR-FRET sandwich assay was performed as shown in FIG.

[0110] TR-FRET detection was based on HTRF® technology (PerkinElmer / Cisbio Bioassays). HTRF® technology uses the fluorescent donor dye Eu3+ Cryptate or Tb 3+ It corresponds to fluorescence resonance energy transfer between a cryptate and a fluorescent acceptor dye (acceptor), d2, each of which is covalently labeled to a molecule (antibody or protein).

[0111] The method described here was based on the use of a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a biotinylated mouse monoclonal anti-human IgG Fab antibody (ThermoFisher Scientific, #SA1-19255) that was used to capture and aggregate the test antibody in solution (Figure 1). Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was detected by Eu 3+ Detection was with mouse monoclonal anti-C1q antibody conjugated to cryptate (HycultBiotech, #HM2382).

[0112] Biotinylation of anti-human IgG Fab antibody A 5 mM stock solution of biotin N-hydroxysuccinimide ester (Sigma-Aldrich, #B2643) was prepared in anhydrous DMSO. Antibodies were placed at a concentration of 1 mg / mL in 0.1 M, pH 9 carbonate buffer, and biotin was added at a biotin / antibody molar ratio of 6. After 30 min of incubation at room temperature (RT), antibodies were purified to remove excess unconjugated biotin in 0.1 M, pH 7 phosphate buffer using a pre-packed gel filtration gravity flow column (Amersham NAP-10 column, Cytiva, #17085401) according to the manufacturer's instructions. Biotinylated antibody stock solutions were supplemented with 0.1% BSA and stored at -20 °C until use.

[0113] EU 3+ Conjugation of anti-C1q antibodies to cryptates EU 3+A 5 mM stock solution of cryptate N-hydroxysuccinimide ester (Cisbio Bioassays) was prepared in anhydrous DMSO. Antibodies were placed at a concentration of 1 mg / mL in 50 mM, pH 8 phosphate buffer. 3+ Cryptate to Eu 3+ Cryptate / antibody was added at a molar ratio of 15. After incubation for 30 min at room temperature (RT), the antibody was purified to remove excess unconjugated Eu in 0.1 M, pH 7 phosphate buffer using a pre-packed gel filtration gravity flow column (Amersham NAP-5 column, Cytiva, #17085301) according to the manufacturer's instructions. 3+ Cryptates were removed. Eu 3+ Cryptate-antibody stock solutions were supplemented with 0.1% BSA and stored at −20° C. until use.

[0114] Description of the test antibody The characteristics of the antibodies tested in the assay are detailed in Table 1. They are all purified antibodies of known concentrations.

[0115] [Table 1]

[0116] Preparation of TR-FRET assay components Immediately prior to use, all assay components were diluted in either Buffer 1 (Na2HPO4 3 mM, KH2PO4 1 mM, NaCl 135 mM, BSA 0.1%, Tween-20 0.05%, ProClin-300 0.01%, pH 7.4) (Figures 2, 3, 4A, and 5) or Buffer 2 (Na2HPO4 4 2.13 mM, KH2PO4 7.86 mM, BSA 0.1%, Tween-20 0.05%, pH 7.4) (Figure 6), depending on the experiment.

[0117] Serial dilutions of the test antibodies were prepared at 4x, aiming for final concentrations in the assay comprised between 1 and 200 nM. Adalimumab was also tested in the presence of its antigen, human TNF-α (R&D Systems, #10291-TA). In that case, serial dilutions were prepared in buffer supplemented with 150 nM TNF-α.

[0118] A premix [biotinylated anti-human IgG Fab antibody / streptavidin-d2] was prepared at 4x in a ratio of 1 / 1 to obtain a final concentration of 50 nM / 50 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu was added at 100 mM for a final concentration of 0.6 nM (Figure 6) or 2.4 nM (Figures 2, 3, 4A, and 5), depending on the experiment. 3+ Cryptate-anti-C1q antibody was prepared at 4×.

[0119] TR-FRET Assay Protocol 4× solutions of the assay components were sequentially dispensed into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) · 5 μL of premix [biotinylated anti-human IgG Fab antibody / streptavidin-d2] 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0120] Plates were covered with a sealer and incubated at room temperature for 3 hours (Figures 2, 3, 4A, and 5) or overnight (Figure 6), depending on the experiment. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording of signal emission at 665 nm and 620 nm).

[0121] TR-FRET signal analysis and data processing The "HTRF ratio" is calculated using the following formula: Signal 665nm / signal 620nm ×10,000. The specific signal "delta ratio" was then calculated as follows: Delta ratio = HTRF ratio 抗体 -HTRF ratio 陰性対照 Binding curves were represented by plotting log[antibody](M) versus delta ratio and fitted in GraphPad Prism 9 using the "Sigmoidal dose-response curve-variable slope (4 parameters)" model as shown in Figures 2, 3, 4A, 5, and 6. For each binding curve, signal intensity was determined by calculating signal to background (S / B) as follows: S / B=HTRF ratio maximum 抗体 / HTRF ratio 陰性対照 . E.C. 50 The results obtained for each antibody tested are shown in Table 2.

[0122] [Table 2]

[0123] The TR-FRET assay was able to distinguish the C1q binding ability of various IgG antibodies of isotypes 1, 2, and 4 (Figure 2).

[0124] As shown in Figure 2A, the best S / B and EC were obtained with the human IgG1 isotype control. 50 This isotype was shown to bind human C1q most efficiently, with a lower S / B and increased EC2 than was obtained with the human IgG2 isotype control. 50indicates that human IgG2 interacts weakly with human C1q. No significant signal was measured in the human IgG4 isotype control, indicating that this isotype does not bind to human C1q. These results are consistent with [Reference 6]. Similar results were obtained when the therapeutic IgG1 antibody rituximab was compared with its IgG2 and IgG4 isotype variants (Figure 2B), and when the therapeutic IgG1 antibodies cetuximab and ipilimumab were compared with their corresponding IgG2 isotype variants (Figures 2C and 2D).

[0125] The TR-FRET assay allowed us to distinguish the interaction capabilities of glycosylated versus non-glycosylated antibodies against human C1q (FIG. 3).

[0126] The therapeutic IgG1 antibody rituximab was tested in parallel with a non-fucosylated (here used as an irrelevant control) and a non-glycosylated variant (Figure 3A). It is known that the absence of fucose residues on the Fc domain only modulates ADCC activity and not C1q binding. As expected, assays with the non-fucosylated antibody did not result in a detectable change in C1q binding compared to rituximab.

[0127] Conversely, the interaction ability of the non-glycosylated mutant was altered (reduced S / B, EC 50 (increased IgG1 binding), consistent with the fact that antibody Fc glycosylation is important for C1q binding. Similarly, this assay showed a significant decrease in the ability of the non-glycosylated version of cetuximab to bind human C1q compared to the therapeutic IgG1 antibody cetuximab (Figure 3B).

[0128] The TR-FRET assay allowed us to distinguish the C1q binding ability of therapeutic type I and type II anti-CD20 antibodies (Figure 4).

[0129] The reduced signal intensity obtained with obinutuzumab (type II anti-CD20) compared to rituximab (type I anti-CD20) indicates that obinutuzumab binds human C1q less efficiently (Figure 4A). These results are in accordance with data from the literature obtained using an ELISA assay (Figure 4B adapted from [reference 5]).

[0130] The therapeutic antibodies atezolizumab and spartalizumab were also tested in the TR-FRET assay (Figure 5).

[0131] As expected, anti-PD-L1 IgG1 atezolizumab efficiently interacted with human C1q, whereas anti-PD-1 IgG4 spartalizumab showed no ability to bind to the protein.

[0132] The TR-FRET assay allowed us to detect the interaction between human C1q and the therapeutic anti-TNF-α IgG1 antibody adalimumab, either precomplexed or uncomplexed with the antigen human TNF-α (Figure 6).

[0133] Furthermore, this assay allows the ability of an antibody to bind C1q in the absence or presence of its antigen, as well as the EC 50 Based on the improvement in values, it was possible to distinguish. These results are consistent with [Reference 7].

[0134] Example 2: Method for determining binding of antibodies to C1q using biotinylated anti-human Fab antibodies and donor directly labeled C1q (TR-FRET)

[0135] Another method consists of performing a TR-FRET assay with human C1q protein directly conjugated to the donor, as shown in FIG.

[0136] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0137] The method described herein is based on the use of a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a biotinylated mouse monoclonal anti-human IgG Fab antibody (ThermoFisher Scientific, #SA1-19255) that was used to capture and aggregate the test antibody in solution (Figure 7). Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was detected by Tb 3+ It was directly conjugated to the cryptate.

[0138] Anti-human IgG Fab antibody was biotinylated as described in Example 1. Human C1q protein was conjugated to Lumi4®-terbium cryptate using a terbium cryptate labeling kit (PerkinElmer / Cisbio Bioassays, #62TBSPEA) according to the manufacturer's instructions.

[0139] Description of the test antibody Human isotype controls IgG1, IgG2, and IgG4, as well as the therapeutic IgG1 antibody Rituximab, were tested in the assay and their characteristics are detailed in Table 1. All are purified antibodies of known concentrations.

[0140] Preparation of TR-FRET assay components Immediately before use, all assay components were diluted in Buffer 2, the formulation of which is detailed in Example 1. Serial dilutions of the test antibodies were prepared at 4x, aiming for final assay concentrations comprised between 1 and 200 nM. A premix [biotinylated anti-human IgG Fab antibody / streptavidin-d2] was prepared at 4x in a ratio of 1 / 1 to obtain a final concentration of 50 nM / 50 nM. Tb 3+ Cryptate-human C1q protein was prepared at 2× to reach a final concentration of 5 nM.

[0141] TR-FRET Assay Protocol Working solutions of the assay components were dispensed sequentially into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) · 5 μL of premix [biotinylated anti-human IgG Fab antibody / streptavidin-d2] 10 μL Tb 3+ Cryptate-Human C1q Protein

[0142] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0143] TR-FRET signal analysis and data processing HTRF data was analyzed and expressed as described in Example 1. The resulting binding curves for each antibody tested are shown in Figure 8. For each binding curve, S / B was calculated as detailed in Example 1, and EC 50 The results obtained for each antibody tested are shown in Table 3.

[0144] [Table 3]

[0145] Similar to the method shown in Example 1, the TR-FRET assay was able to distinguish the C1q binding ability of various human IgG antibodies of isotypes 1, 2, and 4. The best S / B and EC 50 This isotype was shown to bind human C1q most efficiently, with a lower S / B and increased EC2 than was obtained with the human IgG2 isotype control. 50indicates that human IgG2 interacts weakly with human C1q. No significant signal was measured with the human IgG4 isotype control, indicating that this isotype does not bind to human C1q. These results are consistent with [Reference 6].

[0146] The binding profile of the therapeutic IgG1 chimeric anti-CD20 antibody rituximab was close to that obtained with the human IgG1 isotype control, with similar S / B and EC 50 had value.

[0147] Example 3: Method for determining binding of a test antibody to C1q using a biotinylated anti-human Fab antibody (ALPHA)

[0148] To determine the binding of the test antibodies to C1q, an ALPHA sandwich assay was performed as shown in FIG.

[0149] ALPHA detection was based on AlphaLISA® technology (PerkinElmer). AlphaLISA® technology corresponds to the diffusion of singlet oxygen between Alpha donor beads (donor) and AlphaLISA acceptor beads (acceptor). Each bead is covalently conjugated to a molecule (antibody or protein).

[0150] The method described herein is based on the use of Alpha streptavidin-coated donor beads (PerkinElmer, #6760002) conjugated with a biotinylated mouse monoclonal anti-human IgG Fab antibody (ThermoFisher Scientific, #SA1-19255) that was used to capture and aggregate the test antibody in solution (Figure 9). Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was detected with a mouse monoclonal anti-C1q antibody (HycultBiotech, #HM2382) conjugated to AlphaLISA acceptor beads (PerkinElmer, #6772002).

[0151] The anti-human IgG Fab antibody was biotinylated as described in Example 1.

[0152] Conjugation of anti-C1q antibodies to AlphaLISA acceptor beads The conjugation was carried out in PBS buffer pH 7.4 (ThermoFisher Scientific, #10010023) at a ratio of 5 mg AlphaLISA acceptor beads / 100 μg anti-C1q antibody in the presence of 20 mM sodium cyanoborohydride (NaBH3CN) (SIGMA, #156159) and 0.06% Tween-20 (ThermoFisher Scientific, #85113). After overnight incubation at 37°C, the reaction was stopped by adding 3.1 mg / mL carboxy-methoxylamine (CMO) (Sigma, #C13408). The beads were purified to remove excess unconjugated antibody by several washing / centrifugation steps in PBS buffer pH 7.4.

[0153] Description of the test antibody Human isotype controls IgG1, IgG2, and IgG4, as well as the therapeutic IgG1 antibody Rituximab, were tested in the assay and their characteristics are detailed in Table 1. All are purified antibodies of known concentrations.

[0154] Preparation of ALPHA assay components Alpha Streptavidin Coated Donor Beads are light sensitive. All steps using this reagent (preparation, dispensing, plate reading) were performed under subdued laboratory lighting. Immediately prior to use, all assay components were diluted in Buffer 1, the formulation of which is detailed in Example 1.

[0155] Serial dilutions of test antibodies were prepared at 8x aiming for final assay concentrations comprised between 1 and 200 nM. Premix [biotinylated anti-human IgG Fab antibody / Alpha streptavidin coated donor beads] was prepared at 2.67x to obtain a final concentration of 50 nM biotin antibody / 67 μg / mL beads. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. AlphaLISA acceptor beads conjugated to anti-C1q antibody were prepared at 4x aiming for a final concentration of 20 μg / mL.

[0156] ALPHA Assay Protocol Working solutions of the assay components were dispensed sequentially into a 384-well light grey microplate (AlphaPlate-384, PerkinElmer, #6005350) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) · 15μL of premix [biotinylated anti-human IgG Fab antibody / Alpha streptavidin-coated donor beads] · 10 μL human C1q protein. The plate was covered with a sealer and pre-incubated for 30 minutes at 23°C in the dark. · 10 μL of AlphaLISA acceptor beads conjugated to anti-C1q antibody. The plate was covered with a sealer and incubated for a further 30 min at 23 °C in the dark.

[0157] VICTOR® Nivo using standard Alpha settings (excitation at 680 nm, reading emission at 520-620 nm). TM The ALPHA signal was recorded with a reader (PerkinElmer).

[0158] ALPHA signal analysis and data processing Binding curves were represented by plotting log[antibody](M) versus ALPHA signal and fitted in GraphPad Prism 9 using the "Sigmoidal dose-response curve-variable slope (four parameters)" model, as shown in Figure 10. For each binding curve, S / B: S / B = ALPHA signal maximum. 抗体 / ALPHA signal 陰性対照 The signal intensity was calculated by calculating EC 50 The results obtained for each antibody tested are shown in Table 4.

[0159] [Table 4]

[0160] Similar to the TR-FRET-based method shown in Example 1, the ALPHA assay was able to distinguish the C1q binding ability of various human IgG antibodies of isotypes 1, 2, and 4. The best S / B and EC 50 This isotype was shown to bind human C1q most efficiently, with a lower S / B and increased EC2 than was obtained with the human IgG2 isotype control. 50 indicates that human IgG2 interacts weakly with human C1q. No significant signal was measured with the human IgG4 isotype control, indicating that this isotype does not bind to human C1q. These results are consistent with [Reference 6].

[0161] The binding profile of the therapeutic IgG1 chimeric anti-CD20 antibody rituximab was close to that obtained with the human IgG1 isotype control, with similar S / B and EC 50 had value.

[0162] Example 4: Method for determining binding of antibodies to C1q using biotinylated antigen (TR-FRET)

[0163] An alternative method consists of performing a TR-FRET assay using a biotinylated antigen instead of a biotinylated anti-human Fab antibody, as shown in FIG.

[0164] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0165] The method described herein is based on the use of a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a biotinylated antigen (biotinylated recombinant human TNF-α protein, Abcam, #ab167747) that was used to capture and aggregate the test antibody in solution (Figure 11). Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was amplified by Eu ELISA as described in Example 1. 3+ Detection was with a mouse monoclonal anti-C1q antibody conjugated to cryptate (HycultBiotech, #HM2382). The antibody tested in this assay is the therapeutic anti-TNF-α antibody adalimumab, whose characteristics are detailed in Table 1. This is a purified antibody of known concentration.

[0166] Preparation of TR-FRET assay components Immediately before use, all assay components were diluted in Buffer 2, the formulation of which is detailed in Example 1. Serial dilutions of Adalimumab were prepared at 4x, aiming for a final concentration in the assay comprised between 1 and 100 nM. Premix [biotinylated human TNF-α / streptavidin-d2] was prepared at 4x in a ratio of 1 / 1 to obtain a final concentration of 100 nM / 100 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu 3+ Cryptate-anti-C1q antibody was prepared at 4× aiming for a final concentration of 0.6 nM.

[0167] TR-FRET Assay Protocol 4× solutions of the assay components were sequentially dispensed into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of Adalimumab (or 5 μL of buffer for negative control) · 5 μL of premix [biotinylated human TNF-α / streptavidin-d2] 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0168] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0169] TR-FRET signal analysis and data processing HTRF data was analyzed and expressed as described in Example 1. The binding curve obtained for adalimumab is shown in Figure 12. S / B was calculated as detailed in Example 1, and EC 50 The results are shown in Table 5.

[0170] [Table 5]

[0171] A TR-FRET assay using biotinylated human TNF-α allowed the detection of the interaction between human C1q and the therapeutic anti-TNF-α IgG1 antibody adalimumab (Figure 12). The S / B and EC 50 The values ​​(Table 5) are similar to those obtained in a TR-FRET assay using a biotinylated anti-human Fab antibody (Table 2).

[0172] Example 5: Test format [anti-human Fab-biotin / streptavidin-d2] vs. anti-human Fab-d2 (TR-FRET)

[0173] The TR-FRET assay format described in Example 1 and shown in FIG. 13A was compared with an alternative assay format using an anti-human IgG Fab directly labeled with d2 (FIG. 13B).

[0174] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0175] The assay format shown in Figure 13A is based on the use of a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a biotinylated mouse monoclonal anti-human IgG Fab antibody (GeneTex, #GTX27497) used to capture and aggregate the test antibodies in solution. The anti-human IgG Fab antibody was biotinylated as described in Example 1.

[0176] The assay format shown in Figure 13B is based on the use of the same mouse monoclonal anti-human IgG Fab antibody (GeneTex, #GTX27497) directly labeled with d2 using a d2 labeling kit (PerkinElmer / Cisbio Bioassays, #62D2DPEA) according to the manufacturer's instructions.

[0177] In both assay formats, human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was purified by Eu 3+ Detection was with mouse monoclonal anti-C1q antibody conjugated to cryptate (HycultBiotech, #HM2382).

[0178] The antibody tested in this assay was the therapeutic antibody rituximab, whose characteristics are detailed in Table 1. Rituximab was a purified antibody of known concentration.

[0179] Preparation of TR-FRET assay components Immediately before use, all assay components were diluted in buffer 2, the formulation of which is detailed in Example 1. Serial dilutions of Rituximab were prepared at 4x, aiming for a final concentration in the assay comprised between 1 and 100 nM. Premix [biotinylated anti-human IgG Fab / streptavidin-d2] was prepared at 4x in a ratio of 1 / 1 to obtain a final concentration of 50 nM / 50 nM. Anti-human IgG Fab-d2 was prepared at 4x, aiming for a final concentration of 50 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu 3+ Cryptate-anti-C1q antibody was prepared at 4× aiming for a final concentration of 0.6 nM.

[0180] TR-FRET Assay Protocol 4× solutions of the assay components were sequentially dispensed into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of Rituximab (or 5 μL of buffer for negative control) 5 μL of premix [biotinylated anti-human IgG Fab / streptavidin-d2] or 5 μL of anti-human IgG Fab-d2 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0181] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0182] TR-FRET signal analysis and data processing "HTRF Ratio" and "Delta Ratio" values ​​were calculated as detailed in Example 1. The normalized signal "Delta F%" was then calculated using the following formula: Delta F%=Delta Ratio 抗体 / HTRF 陰性対照 Binding curves were represented by plotting log[antibody](M) versus delta F% and fitted in GraphPad Prism 9 using the "Sigmoidal dose-response curve-variable slope (4 parameters)" model, as shown in Figure 14. S / B was calculated as detailed in Example 1, and EC 50 The results are shown in Table 6.

[0183] [Table 6]

[0184] The S / B obtained in the rituximab binding curve using anti-human IgG Fab directly labeled with d2 was 3-fold lower than that obtained using the complex [anti-human IgG Fab-biotin / streptavidin-d2]. Furthermore, the replacement of the complex [biotin-antibody / streptavidin-d2] with d2-antibody reduced the EC 50 values ​​increased by a factor of 2.1. These data suggest that anti-human IgG Fab-d2 alone (not complexed to streptavidin) is unable to adequately induce the aggregation of rituximab and, consequently, its proper binding to human C1q.

[0185] Example 6: Sandwich method (TR-FRET) to determine the affinity of anti-human Fab antibodies to human IgG of various isotypes (IgG1, IgG2, and IgG4)

[0186] As shown in FIG. 15, a TR-FRET sandwich assay was set up to perform saturation binding experiments to determine the affinity of anti-human Fab antibodies to human IgG of various isotypes (IgG1, IgG2, and IgG4).

[0187] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0188] Two different binding assay formats were tested (Figures 15A and B). Both assay formats were performed using a Europium Cryptate labeling kit (PerkinElmer / Cisbio Bioassays, #62EUSUEA) according to the manufacturer's instructions. 3+ It is based on the use of cryptate-labeled human IgG (IgG1, IgG2, or IgG4) antibodies. The human IgG antibodies used in the assay are listed in Table 7.

[0189] [Table 7]

[0190] The binding assay format shown in Figure 15A uses increasing concentrations of anti-human IgG Fab antibodies directly labeled with d2 using the d2 labeling kit (PerkinElmer / Cisbio Bioassays, #62D2DPEA) according to the manufacturer's instructions. The binding assay format shown in Figure 15B uses increasing concentrations of biotinylated anti-human IgG Fab antibodies conjugated to streptavidin-d2 (PerkinElmer / Cisbio Bioassays, #610SADLF). The anti-human IgG Fab antibodies were biotinylated as described in Example 1.

[0191] The characteristics of the three different anti-human IgG Fab antibodies tested in the assay are shown in Table 8.

[0192] [Table 8]

[0193] Preparation of TR-FRET assay components Immediately prior to use, all assay components were diluted in Buffer 3 (Tris-HCl 50 mM, 0.1% BSA, 100 mM KF, pH 7.4). 3+ Cryptate-labeled human IgG (IgG1, IgG2, or IgG4) antibodies were prepared at 4x to reach a final concentration of 0.3 nM. A 2x solution of the corresponding unlabeled human IgG was prepared to obtain a final concentration of 300 nM (used in large excess to compete with the labeled human IgG). Serial dilutions of anti-human IgG Fab-d2 were prepared at 4x aiming for a final assay concentration comprised between 0.02 and 50 nM. A premix [biotinylated anti-human IgG Fab / streptavidin-d2] was prepared at 4x to obtain a final assay concentration comprised between 0.02 nM / 0.02 nM and 50 nM / 50 nM and serially diluted while maintaining the ratio 1 / 1.

[0194] TR-FRET Assay Protocol Working solutions of the assay components were dispensed sequentially into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 10 μL of Buffer 3 (or 10 μL of unlabeled human IgG to determine non-specific signals) 5 μL Eu 3+ Cryptate-Human IgG 5 μL of serial dilutions of anti-human IgG Fab-d2 or 5 μL of serial dilutions of premix [biotinylated anti-human IgG Fab / streptavidin-d2]

[0195] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0196] TR-FRET signal analysis and data processing The "HTRF ratio" was calculated as detailed in Example 1. The total signal corresponds to the HTRF ratio obtained in the absence of unlabeled human IgG. The non-specific signal corresponds to the HTRF ratio obtained in the presence of unlabeled IgG. The specific signal was calculated as follows: Specific signal = HTRF ratio 総シグナル HTRF ratio 非特異的シグナル Saturation binding curves (of total and specific signals) were represented by plotting [anti-human IgG Fab antibody] (nM) versus HTRF ratio and fitted in GraphPad Prism 9 using the "binding-saturation (one site)" model. Non-specific signals were fitted using the "simple linear regression" model. 3+ Cryptate-Human IgG1, Eu 3+ Cryptate-human IgG2, and Eu 3+ The results obtained with cryptate-human IgG4 are shown in Figures 16, 17, and 18, respectively. 3+ Saturation binding experiments with cryptate-human IgG2 were performed only in the assay format based on biotinylated anti-human Fab complexed to streptavidin-d2 (Figure 17). Kd values ​​determined from specific binding saturation curves are shown in Table 9 (nd = not determined).

[0197] [Table 9]

[0198] The Kd values ​​determined for each assay format (anti-human IgG Fab-d2, or biotinylated anti-human IgG Fab conjugated to streptavidin-d2) were relatively close, differing by a maximum of 2.3-fold. Anti-human Fab1 and 2 were Eu 3+ Cryptate-human IgG1, IgG2, and IgG4 antibodies gave similar Kd values ​​ranging from 0.16 nM to 0.47 nM, indicating that these mouse monoclonal antibodies recognize different human IgG isotypes (IgG1, IgG2, and IgG4) similarly and have good affinities in the subnanomolar range. Conversely, anti-human Fab3 did not recognize Eu 3+Compared with the affinity obtained with cryptate-human IgG1 (Kd values ​​of approximately 0.6–1 nM), Eu 3+ Cryptate-exhibited lower affinity for human IgG2 and IgG4 (Kd values ​​between 3.3 and 9.6 nM), indicating that this goat polyclonal antibody does not recognize different human IgG isotypes (IgG1, IgG2, and IgG4) in the same way, and has approximately 1 / 6-fold lower affinity for IgG2 compared to IgG1, and approximately 1 / 9-1 / 7-fold lower affinity for IgG4 compared to IgG1.

[0199] Example 7: Competitive method (TR-FRET) to determine the affinity of anti-human Fab antibodies to fully human, humanized, and chimeric IgG of various isotypes (IgG1, IgG2, and IgG4)

[0200] Based on the TR-FRET sandwich assay shown in Example 6, a competitive binding assay was performed to determine the affinity of anti-human Fab antibodies to fully human, humanized, and chimeric IgG antibodies of various isotypes (IgG1, IgG2, and IgG4), as shown in FIG.

[0201] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0202] Based on the two binding assay formats described in Example 6 (Figures 15A and B, Tables 7, 8), two different competitive binding assay formats were tested (Figure 19). The concentrations of anti-human Fab (labeled with d2 or biotinylated and complexed with streptavidin-d2) were determined based on the Kd values ​​determined in Example 6. Competitive binding experiments were performed by adding increasing concentrations of unlabeled IgG antibodies of different isotypes (IgG1, IgG2, and IgG4) and different formats (fully human, humanized, and chimeric). The characteristics of the IgG antibodies tested in the assays are listed in Table 10. All of them are purified antibodies of known concentrations.

[0203] [Table 10]

[0204] Preparation of TR-FRET assay components Immediately prior to use, all assay components were diluted in Buffer 3, the formulation of which is described in Example 6. 3+ Cryptate-labeled human IgG (IgG1, IgG2, or IgG4) antibodies were prepared at 4x to reach a final concentration of 0.3 nM. Anti-human IgG Fab (labeled with d2 or biotinylated and complexed with streptavidin-d2 in a 1 / 1 ratio) was prepared at 4x to a final concentration corresponding to 2x the Kd values ​​determined in Example 6 (Table 9). Serial dilutions of unlabeled IgG antibodies were prepared at 2x to a final concentration in the assay comprised between 0.03 and 150 nM.

[0205] TR-FRET Assay Protocol Working solutions of the assay components were dispensed sequentially into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 10 μL of serial dilutions of unlabeled IgG antibody (or 10 μL of Buffer 3 to determine the signal maximum) 5 μL Eu 3+ Cryptate-Human IgG 5 μL of anti-human IgG Fab-d2 or 5 μL of premix [biotinylated anti-human IgG Fab / streptavidin-d2]

[0206] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0207] TR-FRET signal analysis and data processing "Delta F%" values ​​were calculated as detailed in Example 5. Delta F% 最大値 corresponds to the delta F% obtained in the absence of unlabeled IgG antibody. The % of signal maximum was calculated for each concentration of unlabeled IgG antibody as follows: % of signal maximum = delta F% 非標識IgG抗体 / Delta F% 最大値 ×100. Competitive binding curves were represented by plotting log[unlabeled IgG antibody](M) versus % of signal maximum and fitted in GraphPad Prism 9 using the "dose response-inhibition (variable slope-four parameters)" model. Competitive binding curves obtained with unlabeled IgG1, IgG2, and IgG4 antibodies are shown in Figures 20, 21, and 22, respectively. IC values ​​calculated from competitive binding curves 50 According to the value, the Cheng-Prusoff equation: Ki=IC 50 Ki values ​​were calculated using the formula: / (1+([labeled human IgG] / Kd)). IC obtained with unlabeled IgG1, IgG2, and IgG4 50 and Ki values ​​are shown in Tables 11, 12, and 13, respectively (nd=not determined).

[0208] Competitive binding experiments with fully human, humanized, and chimeric IgG1 antibodies were performed with biotinylated anti-human Fab1 conjugated to streptavidin-d2 (Figure 20A) and with anti-human Fab2 and 3 directly labeled with d2 (Figures 20B and C). Anti-human Fab3-d2 was not tested with any IgG1 antibodies.

[0209] [Table 11]

[0210] The competition curves obtained with all tested IgG1 antibodies, whatever the anti-human Fab used, were overlaid. The Ki values ​​corresponding to the affinity of each anti-human Fab for each tested IgG1 antibody were very close, averaging about 0.2 nM, whatever the anti-human Fab used (Table 11). This shows that anti-human Fabs 1, 2, and 3 recognize fully human, humanized, and chimeric IgG1 antibodies similarly, with affinities in the subnanomolar range.

[0211] [Table 12]

[0212] Competitive binding experiments with fully human IgG2 and chimeric IgG2 antibodies were performed using only biotinylated anti-human Fab1 conjugated to streptavidin-d2 (Figure 21). The competition curve profiles obtained with all tested IgG2 antibodies were similar, with Ki values ​​ranging from 0.27 to 0.72 nM (Table 12). This indicates that anti-human Fab1 recognizes fully human and chimeric IgG2 antibodies similarly, with an average affinity of 0.5 nM.

[0213] [Table 13]

[0214] Competitive binding experiments with fully human, humanized, and chimeric IgG4 antibodies were performed with biotinylated anti-human Fab1 conjugated to streptavidin-d2 (Figure 22A) and with anti-human Fab2 and 3 directly labeled with d2 (Figures 22B and C). Anti-human Fab3-d2 was not tested with any IgG4 antibodies.

[0215] The profiles of the competition curves obtained with all tested IgG4 antibodies were similar whether anti-human Fab1 or anti-human Fab2 was used (Figures 22A and B). The Ki values ​​determined with each of the two anti-human Fabs were very close, averaging 0.1 nM for anti-human Fab1 and 0.4 nM for anti-human Fab2 (Table 13). This indicates that these mouse monoclonal anti-human Fab antibodies bind similarly to fully human, humanized, and chimeric IgG4 antibodies with affinities in the sub-nanomolar range. The competition binding curves obtained with anti-human Fab3 showed different profiles depending on the nature of the antibody tested (Figure 22C). In fact, the human IgG4 isotype control antibody induced a total inhibition of the signal with a Ki value of 0.13 nM. Conversely, competition with the therapeutic antibodies pembrolizumab (humanized IgG4) and nivolumab (fully human IgG4) was only partial, with only 31-32% inhibition of the signal. These results suggest that the goat polyclonal anti-human Fab3 antibody does not adequately recognize the therapeutic IgG4 antibody.

[0216] Example 8: Optimization of the concentration of biotinylated anti-Fab ligand

[0217] To optimize the concentration of the biotinylated anti-Fab ligand, the TR-FRET assay described in Example 1 was performed using various concentrations of the anti-Fab ligand.

[0218] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0219] The assay format uses a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a mouse monoclonal anti-human IgG Fab antibody conjugated to biotin (ThermoFisher Scientific, #SA1-19255) as described in Example 1. Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was purified by Eu 3+ Detection was with mouse monoclonal anti-C1q antibody conjugated to cryptate (HycultBiotech, #HM2382).

[0220] The antibodies tested in the assay were the therapeutic antibody Atezolizumab and a human IgG2 isotype control, whose characteristics are detailed in Table 1. Both are purified antibodies of known concentrations.

[0221] Preparation of TR-FRET assay components Immediately before use, all assay components were diluted in buffer 2, the formulation of which is detailed in Example 1. Serial dilutions of the tested antibodies were prepared at 4x, aiming for final concentrations in the assay comprised between 0.5 and 100 nM. Three different solutions of premix [biotinylated anti-human IgG Fab / streptavidin-d2] were prepared at 4x in a ratio of 1 / 1 to obtain final concentrations of 30 nM / 30 nM, 40 nM / 40 nM, or 50 nM / 50 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu 3+ Cryptate-anti-C1q antibody was prepared at 4× aiming for a final concentration of 0.6 nM.

[0222] TR-FRET Assay Protocol 4× solutions of the assay components were sequentially dispensed into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) 5 μL of premix [biotinylated anti-human IgG Fab / streptavidin-d2] tested at three different concentrations 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0223] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0224] TR-FRET signal analysis and data processing "Delta F%" values ​​were calculated and dose-response binding curves were fitted as described in Example 5. Binding profiles using atezolizumab (IgG1) and IgG2 isotype controls with various concentrations of biotinylated anti-human IgG Fab are shown in Figures 23A and 23B, respectively. S / B was calculated as detailed in Example 1, and EC 50 The values ​​were also determined, and the results are shown in Table 14.

[0225] [Table 14]

[0226] For both IgG1 and IgG2 antibodies, the binding curves obtained with 30 nM biotinylated anti-human IgG Fab showed a left-shifted EC compared to the binding curves obtained with 50 nM biotinylated anti-Fab ligand. 50 The binding curves obtained with 40 or 50 nM biotinylated anti-human IgG Fab were nearly superimposable, with the signal intensity (S / B) being slightly better with 50 nM biotinylated anti-Fab ligand.

[0227] These results indicate that the optimal concentration of biotinylated anti-human IgG Fab used to reliably function at a saturating dose is 50 nM, confirming that this concentration is approximately 100-fold higher and in large excess than the affinities previously determined in Example 7 (approximately 0.2 nM for IgG1 and approximately 0.5 nM for IgG2).

[0228] Example 9: Optimization of salt concentration

[0229] To optimize the salt concentration in the assay, the TR-FRET assay described in Example 1 was performed in 4 mM PO4 buffer supplemented with various concentrations of NaCl.

[0230] TR-FRET detection is based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0231] The assay format uses a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a mouse monoclonal anti-human IgG Fab antibody conjugated to biotin (ThermoFisher Scientific, #SA1-19255) as described in Example 1. Human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was purified by Eu 3+ Detection was with mouse monoclonal anti-C1q antibody conjugated to cryptate (HycultBiotech, #HM2382).

[0232] The antibodies tested in this assay were the therapeutic antibodies rituximab and obinutuzumab, as well as the rituximab IgG2 isotype variant. Their characteristics are detailed in Table 1. All are purified antibodies of known concentration.

[0233] Preparation of TR-FRET assay components 4 mM PO4 buffer (Na2HPO4 3 mM, KH2PO4 1 mM, BSA 0.1%, Tween-20 0.05%, ProClin-300 0.01%, pH 7.4) was prepared and supplemented with various concentrations of NaCl ranging from 125 to 155 mM. Immediately before use, all assay components were diluted in the respective buffers containing different concentrations of NaCl. Serial dilutions of the tested antibodies were prepared at 4x aiming for a final concentration in the assay comprised between 1 and 100 nM. A premix [biotinylated anti-human IgG Fab / streptavidin-d2] was prepared at 4x in the ratio 1 / 1 to obtain a final concentration of 50 nM / 50 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu 3+ Cryptate-anti-C1q antibody was prepared at 4× aiming for a final concentration of 1.2 nM.

[0234] TR-FRET Assay Protocol 4× solutions of the assay components prepared in each buffer were dispensed sequentially into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) · 5 μL of premix [biotinylated anti-human IgG Fab / streptavidin-d2] 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0235] The plate was covered with a sealer and incubated overnight at room temperature. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording signal emission at 665 nm and 620 nm).

[0236] TR-FRET signal analysis and data processing "Delta Ratio" and "Delta F%" values ​​were calculated as described in Examples 1 and 5, respectively. Binding curves were fitted as described in Example 5.

[0237] The maximum delta ratio values ​​obtained for 50 nM rituximab and 50 nM obinutuzumab in buffers supplemented with various concentrations of NaCl are shown in Figure 24. For each concentration of NaCl, the corresponding osmolality of the buffer was calculated and further the signal maximum was used to evaluate the ability of the assay to discriminate the C1q binding ability of the two anti-CD20 antibodies. リツキシマブ / signal maximum オビヌツズマブ The ratio was calculated (Table 15).

[0238] [Table 15]

[0239] In the presence of NaCl concentrations ranging from 135 to 155 mM (corresponding to osmolalities of 282 mOsm / L to 322 mOsm / L), the ratio was greater than 3, indicating that the assay adequately discriminates between the ability of the type I anti-CD20 antibody rituximab and the type II anti-CD20 antibody obinutuzumab to interact with human C1q. Below this NaCl concentration (osmolality less than 282 mOsm / L), the ratio was less than 3, indicating less good discrimination.

[0240] The binding curves (expressed as delta F%) for rituximab, obinutuzumab, and rituximab IgG2 isotype variants obtained in the presence of 155 mM and 135 mM NaCl are shown in Figures 25A and B, respectively. For each curve, S / B was calculated as detailed in Example 1, and EC 50 These values ​​are shown in Table 16.

[0241] [Table 16]

[0242] These results showed that NaCl concentration also influenced the assay sensitivity. At 155 mM NaCl, the S / B obtained with obinutuzumab and rituximab IgG2 isotype variants was very low (included between 1.2 and 1.6). When 135 mM NaCl was used, the discrimination between the different antibodies was still good and the S / B was more robust (above 1.7). Therefore, this concentration of NaCl was optimal to investigate the binding of antibodies to human C1q.

[0243] Example 10: Test format [Anti-human Fab-biotin / streptavidin-d2] versus [Protein L-biotin / streptavidin-d2] (TR-FRET)

[0244] The TR-FRET assay format described in Example 1 and shown in FIG. 29A was compared with an alternative assay format that uses recombinant Protein L instead of an anti-human Fab antibody (FIG. 29B).

[0245] TR-FRET detection was based on HTRF® technology (PerkinElmer / Cisbio Bioassays), as described in Example 1.

[0246] The assay format shown in Figure 29A was based on the use of a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and a biotinylated mouse monoclonal anti-human IgG Fab antibody (GeneTex, #GTX27497) that was used to capture and aggregate the test antibodies in solution. The anti-human IgG Fab antibody was biotinylated as described in Example 1.

[0247] The assay format shown in Figure 29B was based on a complex of d2-labeled streptavidin (PerkinElmer / Cisbio Bioassays, #610SADLF) and biotinylated recombinant Protein L (ThermoScientific, #21189) that was used to capture and aggregate the test antibodies in solution. Recombinant Protein L was biotinylated using the same protocol used for the anti-human Fab antibody described in Example 1.

[0248] In both assay formats, human C1q protein (Sigma-Aldrich, #C1740) bound to the test antibody was purified by Eu 3+ Detection was with a mouse monoclonal anti-C1q antibody conjugated to cryptate (Hycult Biotech, #HM2382).

[0249] The antibodies tested in this assay were the therapeutic monoclonal antibodies adalimumab and trastuzumab. The characteristics of adalimumab are detailed in Table 1. Trastuzumab is a humanized IgG1 antibody that targets the HER2 receptor. Both were purified antibodies of known concentrations.

[0250] Preparation of TR-FRET assay components Immediately before use, all assay components were diluted in Buffer 1, the formulation of which is detailed in Example 1. Serial dilutions of each antibody were prepared at 4x, aiming for a final concentration in the assay comprised between 1 and 400 nM. Premixes [biotinylated anti-human IgG Fab / streptavidin-d2] or [biotinylated protein L / streptavidin-d2] were prepared at 4x in a ratio of 1 / 1 to obtain a final concentration of 50 nM / 50 nM. Human C1q protein was prepared at 4x to reach a final concentration of 5 nM. Eu 3+ Cryptate-anti-C1q antibody was prepared at 4× aiming for a final concentration of 0.6 nM.

[0251] TR-FRET Assay Protocol 4× solutions of the assay components were sequentially dispensed into a 384-well low volume white microplate (Proxiplate Plus, PerkinElmer, #6008280) as follows: 5 μL of serial dilutions of the test antibody (or 5 μL of buffer for negative control) 5 μL of premix [biotinylated anti-human IgG Fab / streptavidin-d2] or [biotinylated protein L / streptavidin-d2] 5 μL human C1q protein 5 μL Eu 3+ Cryptate-anti-C1q antibody

[0252] The plate was covered with a sealer and incubated at room temperature for 3 hours. HTRF signals were recorded on a PHERAstar FS microplate reader (BMG Labtech) using the HTRF detection module (excitation with a 337 nm flash lamp, recording of signal emission at 665 nm and 620 nm).

[0253] TR-FRET signal analysis and data processing "HTRF Ratio" and "Delta Ratio" values ​​were calculated as detailed in Example 1. The normalized signal "Delta F%" was then calculated as detailed in Example 5. Binding curves were represented by plotting log[antibody](M) versus Delta F% and fitted in GraphPad Prism 9 using the "Sigmoidal dose response curve - variable slope (4 parameters)" model, as shown in Figure 30. S / B was calculated as detailed in Example 1, and EC 50 The results are shown in Table 17.

[0254] [Table 17]

[0255] The S / B obtained for the adalimumab and trastuzumab binding curves using the complex [Protein L-biotin / streptavidin-d2] was 3.4 times lower than that obtained using the complex [anti-human Fab-biotin / streptavidin-d2] (Table 17). Furthermore, when the anti-human Fab antibody was replaced by recombinant Protein L, the EC 50 The values ​​increased nine-fold (Table 17).

[0256] These data demonstrate that biotinylated anti-human Fab antibody, but not biotinylated Protein L, induces the appropriate signal.

[0257] References [Reference 1] Kaul et al., Dissection of C1q Capability of Interacting with IgG, THE JOURNAL OF BIOLOGICAL CHEMISTRY, 1997 [Reference 2] Patel et al., IgG subclass specificity to C1q determined by surface plasmon resonance using Protein L capture technique, Analytical biochemistry, 2015 [Reference 3] Lilienthal et al., Potential of Murine IgG1 and Human IgG4 to Inhibit the Classical Complement and Fcγ Receptor Activation Pathways, Frontiers in Immunology, 2018 [Reference 4] Diebolder et al., Complement Is Activated by IgG Hexamers Assembled at the Cell Surface,Science,2014 [Reference 5] Herter et al., Preclinical Activity of the Type II CD20 Antibody GA101(Obinutuzumab) Compared with Rituximab and Ofatumumab In Vitro and in Xenograft Models, Molecular Cancer Therapeutics, 2013 [Reference 6] Almagro et al., Progress and Challenges in the Design and Clinical Development of Antibodies for Cancer Therapy, Frontier in Immunology, 2018 [Reference 7] Zhou et al., Characterization of antibody-C1q interactions by Biolayer Interferometry, Analytical Biochemistry, 2018 [Reference 8] Stubenrauch et al., Characterization of murine anti-human Fab antibodies for use in an immunoassay for generic quantification of human Fab fragments in non-human serum samples including cynomolgus monkey samples, Journal of Pharmaceutical and Biomedical Analysis, 2013

Claims

1. 1. An in vitro method for determining binding of an antibody (test antibody) to complement component 1q (C1q), comprising: a) the measuring medium, A test antibody, a biotinylated anti-Fab ligand capable of binding to the Fab region of the test antibody; Streptavidin, directly or indirectly labeled with a first member of a pair of HPA (Homogeneous Proximity Assay) partners; and C1q directly or indirectly labeled with a second member of the HPA partner pair and b) measuring an HPA signal in the measurement medium, the presence of which indicates binding of the test antibody to the C1q; The method includes:

2. 2. The method of claim 1, wherein the HPA is selected from (i) chemically amplified luminescent oxygen channeling immunoassays (LOCI), such as chemically amplified luminescent proximity homogeneous assays (ALPHA), (ii) resonance energy transfer (RET), and (iii) spatial proximity analytical reagent capture luminescence (SPARCL).

3. 3. The method of claim 1 or 2, wherein the test antibody is an IgG, e.g., IgG1, IgG2, or IgG4, preferably IgG1 or IgG2.

4. The method of any one of claims 1 to 3, wherein the biotinylated anti-Fab ligand is a biotinylated anti-Fab antibody or antibody fragment, such as a biotinylated anti-Fab mouse antibody or antibody fragment, such as a biotinylated anti-Fab mouse antibody or antibody fragment.

5. The method of any one of claims 1 to 3, wherein the biotinylated anti-Fab ligand is a biotinylated antigen or antigen fragment.

6. 6. The method of any one of claims 1 to 5, wherein the streptavidin is directly labelled with a first member of an HPA partner pair, such as a first member of a RET partner pair or a first member of a LOCI partner pair.

7. 7. The method of any one of claims 1 to 6, wherein the C1q is indirectly labeled with an anti-C1q ligand labeled with a second member of an HPA partner pair, such as a second member of a RET partner pair or a second member of a LOCI partner pair.

8. (i) a first member of the HPA partner pair is an acceptor and a second member of the HPA partner pair is a donor; or 8. The method of any one of claims 1 to 7, wherein (ii) a first member of the HPA partner pair is a donor and a second member of the HPA partner pair is an acceptor.

9. The HPA is RET and the donor is a FRET (Förster Resonance Energy Transfer) donor selected from europium cryptates, europium chelates, terbium chelates, terbium cryptates, ruthenium chelates, quantum dots, allophycocyanins, rhodamines, cyanines, squaraines, coumarins, proflavines, acridines, fluoresceins, boron-dipyrromethene derivatives, and nitrobenzoxadiazoles, or A BRET (Bioluminescence Resonance Energy Transfer) donor selected from luciferase (luc), Renilla luciferase (Rluc), a mutant of Renilla luciferase (Rluc8), and firefly luciferase. The method of claim 8, wherein the donor compound is selected from luminescent or fluorescent donor compounds such as

10. The HPA is RET and the acceptor is allophycocyanin, rhodamine, cyanine, squaraine, coumarin, proflavine, acridine, fluorescein, boron-dipyrromethene derivatives, nitrobenzoxadiazole, quantum dots, GFP mutants selected from GFP, GFP10, GFP2, and eGFP, YFP mutants selected from YFP, eYFP, YFP topaz, YFP citrine, YFP venus, and YPet, FRET acceptors selected from mOrange, DsRed, or allophycocyanin, rhodamine, cyanine, squaraine, coumarin, proflavine, acridine, fluorescein, boron-dipyrromethene derivatives, nitrobenzoxadiazole, quantum dots, GFP mutants selected from GFP, GFP10, GFP2, and eGFP, YFP mutants selected from YFP, eYFP, YFP topaz, YFP citrine, YFP venus, and YPet, BRET acceptors selected from mOrange and DsRed.

10. The method according to claim 8, wherein the acceptor is selected from a fluorescent acceptor compound such as or a non-fluorescent acceptor compound (quencher).

11. The method of claim 8 , wherein the HPA is LOCI and the donor is a phthalocyanine.

12. 12. The method of claim 11, wherein the HPA is LOCI and the acceptor comprises (i) thioxene, anthracene, and rubrene, or (ii) thioxene and europium chelates.

13. Steps (a) and (b) are repeated using various concentrations of the test antibody, and preferably the method further comprises determining the dissociation constant (Kd) and / or EC ５０ The method of any one of claims 1 to 12, comprising the additional step (c) of determining:

14. The method according to any one of claims 1 to 13, wherein the osmolality of the measurement medium is between 250 mOsm / L and 500 mOsm / L.

15. A reagent kit for carrying out the method according to any one of claims 1 to 14, comprising: (i) a biotinylated anti-Fab ligand; (ii) streptavidin or streptavidin directly labeled with a first member of a pair of HPA partners; (iii) C1q or C1q directly labeled with a second member of the HPA partner pair; (iv) if the streptavidin is not directly labeled with the first member of the HPA partner pair, an anti-streptavidin ligand directly labeled with the first member of the HPA partner pair; (v) if the C1q is not directly labeled with the second member of the HPA partner pair, an anti-C1q ligand directly labeled with the second member of the HPA partner pair; Kit including:

Images