METHOD FOR SELECTING BIOLOGICAL BINDING MOLECULES

Abstract

The present invention relates to the field of producing, identifying, and selecting biological binding molecules, e.g. in particular antibodies or fragments thereof, which selectively bind to autonomously active B-cell receptors or B-cell receptor complexes. The method is used in order to select a biological binding molecule which specifically binds to an autonomously active or autonomously activated B-cell receptor as the target receptor, but not to an inactive or non-activated B-cell receptor, and is carried out in a cell-based system using immature B cells which are in the pro/pre-stage and cause a ‘triple knockout’ of the genes for RAG2 or RAG1, lambda5, and SLP65.

Claims

1. A method for selecting a biological binding molecule that specifically binds to an autonomously active or autonomously activated B-cell receptor as target receptor, but not to a non-active or non-activated B-cell receptor, within the scope of a cell-based system using immature B cells in the pro-/pre-stage, comprising the following steps: a. providing a plurality of biological binding molecules obtained through immunization of a mammal with B-cell receptors or their fragments and subsequent immortalization and purification; b. providing immature B cells in the pro-/pre-stage, which are not capable of expressing the native genes for RAG2 and/or RAG1 and lambda5 but which were enabled to express autonomously active or autonomously activated B-cell receptors as target receptors on their cell surface; c. providing immature B cells in the pro-/pre-stage, which are not capable of expressing the native genes for RAG2 and/or RAG1 and lambda5 but which were enabled to express non-active or non-activated B-cell receptors as reference receptors on their cell surface; d. comparatively examining the binding behavior of the binding molecules provided according to step (a) with respect to cells provided according to steps (b) and (c); e. selecting at least one binding molecule that binds specifically to cells provided according to step (b) but not to cells provided according to step (c).

2. The method according to claim 1, characterized in that the cells provided according to step (b) are also not capable of expressing the native gene SLP65.

3. The method according to claim 1, characterized in that cells are provided according to step (c), which express a non-autonomously active B-cell receptor as a reference receptor.

4. The method according to claim 2, characterized in that, in addition to determining a specific binding of the binding molecule to cells provided according to step (b), step (e) includes a confirmation through an activity measurement after the induction of SLP65.

5. The method according to claim 2, characterized in that cells are provided according to step (c), which express a non-autonomously active B-cell receptor as a reference receptor.

6. The method according to claim 3, characterized in that, in addition to determining a specific binding of the binding molecule to cells provided according to step (b), step (e) includes a confirmation through an activity measurement after the induction of SLP65.

7. The method according to claim 3, characterized in that the cells provided according to step (b) are also not capable of expressing the native gene SLP65 and in addition to determining a specific binding of the binding molecule to cells provided according to step (b), step (e) includes a confirmation through an activity measurement after the induction of SLP65.

Description

EXAMPLE 1

[0056] For the production of triple knockout cells (TKO), transgene mice that have a respective knockout for the genes lambda5, RAG2 and SLP65 are the starting point (Dühren von Minden et al., 2012, Nature 489, p. 309-313). The preparation of such mice is known to a person skilled in the art and belongs to the prior art. For isolating the cells, the bone marrow of the femur of the mice was extracted after their death. The cells thus obtained were subsequently cultured under conditions that facilitate the survival of pro-/pre-B cells (37° C., 7.5% CO.sub.2, Iscove's medium, 10% FCS, P/S, murine IL7). After several passages, a FACS sorting was carried out as a control, which sorts the pro-/pre-B cells and subsequently cultures them again. The markers used for this purpose are known to a person skilled in the art.

[0057] For the reconstitution with a ‘BCR of interest,’ the corresponding coding sequences for the heavy (HC) and light (LC) chains were synthesized and then in each case cloned in expression vectors having a CMV promotor. They were introduced by means of lipofection into the packaging cell line (Phoenix cell line). After a 36-hour incubation, the virus supernatant was taken and used for a spinfection of the TKOs. Both the work for isolating the supernatants and the spinfection of the TKOs are widely known methods and known to a person skilled in the art.

[0058] The structural special features of subset 2 B-cell receptors were taken from the corresponding literature (see above). Exemplary CLL subset 2 VH and complete LC DNA segments were synthesized by a contract manufacturer in a standard method. They were then fused with a murine IgG1 constant segment by means of PCR and cloned in a CMV vector. The sequence of the finished vector was confirmed by means of Sanger sequencing.

TABLE-US-00001 CLL subset 2 VH (SEQ ID NO. 5): EVQLVESGGGLVKPGGSLRLSCAASGFTFRSYSMNWVRQAPGKGLEWVSS IISSSSYIYYADSVKGRFTISRDNAKNSLYLQMNSLRAEDTALYYCARDQ NAMDVWGQGTTVTVSS CLL-Subset 2 LC (SEQ ID NO. 6): SYELTQPPSVSVAPGKTARITCAGNNIGSKSVHWYQQKPGQAPVLVIYYD SDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSGSDHPWVF FFTKLTVLRQPKAAPSVTLFPPSSEELQANKATLVCLISDFYPGAVTVAW KADSSPVKAGVETTTPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHE GSTVEKTVAPTECS

[0059] For the expression of the CLL subset 2 IgG1, a human cellular expression system based on HEK293T cells was used. A protocol based on polyethyleneimine (PEI) was applied for the transfection. After several passages, the supernatant was pooled and the medium contained in the combined cell supernatant was purified by means of protein G columns. The purity and quality of the soluble subset 2 IgG1 was determined by means of Western blot.

[0060] The monoclonal antibodies were produced according to the standard method in mice and with the subsequent generation of hybridoma cells. The screening for positive clones did not take place conventionally by means of ELISA. Since the target structure is a membranous receptor, it is of central importance to validate the binding of the potential antibodies also in a cellular system, i.e., by largely preserving the cell physiological states native for this cell type. First, groups of pooled supernatants were examined for binding events by means of FACS analysis. For this purpose, different CLL subset 2 BCR variants were expressed on the surface of a cell line (TKO) that cannot express an BCR itself. This way, it was at first possible to identify the supernatants whose antibodies showed a binding. Subsequently, the supernatants of the individual hybridoma clones were examined more thoroughly with respect to their binding in order to thus identify highly specific clones with high affinity.

[0061] For the screening method, different vectors were used within the scope of the preceding transformation for the following combinations of a heavy chain (HC) and a light chain (LC) of the corresponding CLL BCRs, wherein these combinations were used on the surface of the BCR reconstitution system: [0062] Control (transformation vector without BCR) (see FIG. 1 A) [0063] Vector with HC/LC typical for the CLL subset 2 (see FIG. 1 B) [0064] Vector with a non-CLL subset 2 HC/an LC typical for the CLL subset 2 (without target motif; epitope) (see FIG. 1 C) [0065] Vector with HC typical for the CLL subset 2/a non-CLL subset 2 LC (see FIG. 1 D) [0066] Vector with a non-CLL subset 2 HC/a non-CLL subset 2 LC (see FIG. 1 E) [0067] Vector with HC/LC typical for the CLL subset 2 (including mutation R110G (target motif)) (see FIG. 1 F).

[0068] This approach to selection is schematically shown in FIG. 1 with the example of the CLL subset 2 BCRs, wherein the designation ‘TKO’ refers to TKO cells (see above).

[0069] In the 1.sup.st selection round, the supernatants of a plurality of clones were combined and examined with respect to their binding profiles on the selection matrix. A binding profile is positive if a specific binding to the “BCR of interest” is shown. Groups which showed such a profile were isolated, and the binding profile of the individual clones was once again characterized within the scope of a second selection round on the selection matrix. The binding of the monoclonal antibodies was verified using a FACS binding assay using a fluorescence-marked anti-mouse IgG antibody. The letters indicate the following: A) no BCR (control); B) a CLL subset2 typical BCR; C) a BCR with an arbitrary heavy chain and a CLL subset2 typical light chain; D) a BCR with a CLL subset2 typical heavy chain and an arbitrary light chain; E) a BCR with arbitrary heavy and light chain (control; not CLL subset2 typical BCR); F) a CLL subset2 typical BCR with a mutation in the target motif (R110G) (control).

[0070] Based on the finding that the antibody only binds to the cells with the target structures (CLL subset2 BCR; FIG. 1B), it can be concluded that in this case an antibody is present that binds specifically to cells with autonomously active receptors.

[0071] In this case, it was found that the use of cells which are in the pro-/pre-stage of B cell development is necessary for the exact expression of the BCR required for verification. These cells are in their development genetically equipped to present new BCRs by exact folding and expression on their surface. Through the inactivation (knockout) of RAG2 and lambda5, the expression of an endogenous BCR or pre-BCR is prevented. The deletion of SLP65 and the subsequent reconstruction of an inducible SLP65 makes it possible to characterize the activity level of the “BCR of interest.”

[0072] For determining the amino acid sequence of the monoclonal antibodies selected by means of selection, the mRNA was isolated from the individual hybridoma clones, and cDNA was generated therefrom, which was amplified by means of anchored PCR (Rapid expression cloning of human immunoglobulin Fab fragments for the analysis of antigen specificity of B cell lymphomas and anti-idiotype lymphoma vaccination; Osterroth F, Alkan O, Mackensen A, Lindemann A, Fisch P, Skerra A, Veelken H., J Immunol Methods 1999 Oct. 29; 229 (1-2): 141-53).

[0073] After identification and sequence determination of the regions important for the binding (CDRs), they were transferred by means of PCR to a human antibody structure. For this purpose, the VH sequence was generated in silico from the human FR regions and the murine CDR regions and subsequently synthesized as DNA fragments. They were subsequently fused by means of PCR with a human IgG1 and cloned in a vector suitable for the expression.

[0074] For generating the monoclonal antibodies, synthetic peptides, in addition to the complete immunoglobulins, were also used, which represent the regions for the capacity of an autonomous signal.

[0075] The specific monoclonal antibody against subset 2 was sequenced. In the process, the following amino acid sequences were determined, wherein the SEQ ID NO. 9 relates to the variable part of the heavy chain (HC), and the SEQ ID NO. 10 relates to the variable part of the light chain (LC), and wherein the marked regions—in the specified order—designate CDR 1, 2 and 3.

TABLE-US-00002 (AVA-mAb01 HC) SEQ ID NO. 9 QVQLLQQSGPGLVQPSQSLSITCTVS IHWVRQSPKGKGL EWLGV DSNAAFMSRLSITKDNSKSQVFFKMNSLQADDTAI YYC WGQGTSVTVSS (AVA-mAb01 LC) SEQ ID NO. 10 QIVLTQSPASLSASVGETVTITCRAS LAWYQQKQGKSPQLLV Y TLADGVPSRFSGSGSGTQYSLKINSLQPEDFGSYYC FGAGTKLELK

[0076] The partial sequences of the heavy chain corresponding to CDR1, CDR2 and CDR3 according to SEQ ID NO. 9 are specified in SEQ ID NOS. 11 to 13, while the partial sequences of the light chain corresponding to CDR1, CDR2 and CDR3 according to SEQ ID NO. 10 are shown in SEQ ID NOS. 14 to 16.

TABLE-US-00003 (AVA-mABO1 CDR1 HC) SEQ ID NO. 11 GFSLTSYG (AVA-mABO1 CDR2 HC) SEQ ID NO. 12 IWRGGGT (AVA-mABO1 CDR3 HC) SEQ ID NO. 13 ARSRYDEEESMNY (AVA-mABO1 CDR1 LC) SEQ ID NO. 14 GNIHSY (AVA-mABO1 CDR2 LC) SEQ ID NO. 15 NAKT (AVA-mABO1 CDR3 LC) SEQ ID NO. 16 QHFWNTPPT

[0077] The above-described approach is exemplary for the generation of specific antibodies with respect to CLL subset 2. The same process was also carried out using specific sequences and isotypes for subset 4.

[0078] Exemplary CLL subset 4 VH and complete LC DNA segments were synthesized by a contract manufacturer in a standard method. They were then fused with a murine IgG1 constant segment by means of PCR and cloned in a CMV vector. The sequence of the finished vector was confirmed by means of Sanger sequencing.

TABLE-US-00004 CLL subset 4 HC (SEQ ID NO. 7): QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWTWIRQSPGKGLEWIGE INHSGSTTYNPSLKSRVTISVDTSKNQFSLKLNSVTAADTAVYYCARGYG DT MDVWGQGTTVTVSSASTKGPSVFPLAPSSKSTSGGTAA LGCLVKDYFPEPVTVSWNSGALTSGVHTFPACLQSSGLYSLSSVVTVPSS SLGTQTYICNV DKKC

[0079] The regions in bold type denote the target sequences (epitopes) of the variable part of the heavy chain of the BCR of subset 4, which are responsible for its autonomously active state (cf. SEQ ID NOS. 3 and 4).

TABLE-US-00005 CLL subset 4 LC (SEQ ID NO. 8): DIVMTQSPLSLPVTLGQPASISCRSSQSLVHSDGNTYLNWFQQRPGQSPR RLIYKVSDRDSGVPDRFSGSGSGTDFTLKISRVEAEDVGLYYCMQGTHWP PTFGQGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAK VQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACE VTHQGLSSPVTKSFNRGEC.