Methods for generating high affinity antibodies to fentanyl

Abstract

The present invention relates to a method for producing an antibody that specifically binds to an antigen, preferably a hapten, such as fentanyl or a derivative thereof, comprising the steps of: a) contacting a B cell sample from an animal, preferably a mouse, immunized with the antigen, with a labeled antigen; b) isolating from the sample cells that are CD19 positive and CD138 negative and that bind the labeled antigen; c) determining the nucleic acid sequence of a plurality of expressed genes, preferably the entire transcriptome for each of the individually isolated cells; d) determining the nucleic acid sequence of BhIhe41, Par, BhIhe51, Par, BhIhe61, Par, BhIhe71, Par, BhIhe81, Par, BhIhe91, Par, BhIhe1, Par, BhIhe1, Par, BhIhe2, Par, BhIhe3, Par, BhIhe4, Par, BhIhe51 ... selecting individual memory B cells among the individually isolated cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of m1, CD80, CobI, IgG1, Sspn, Ackr2, Nt5e, and Mki67; e) assembling nucleic acid sequences encoding antibody light and heavy chain variable chains from the nucleic acid sequences of the expressed genes of the selected individual memory B cells; and f) expressing the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) in a host cell to produce an antibody.

Description

[Technical field]

[0001] The present invention relates to a method for producing an antibody that specifically binds to an antigen, preferably a hapten, such as fentanyl or a derivative thereof, comprising the steps of: a) contacting a B cell sample from an animal, preferably a mouse, immunized with an antigen, with a labeled version of the antigen; b) isolating from the sample cells that are CD19 positive, CD138 negative and cells that bind the labeled antigen; c) determining the nucleic acid sequence of a plurality of expressed genes, preferably the entire transcriptome for each of the isolated individual cells; d) determining the nucleic acid sequence of a plurality of expressed genes, preferably BhIhe41, Parm1, CD80, CD138, CD19, CD20, CD21, CD12, CD138, CD22, CD139, CD24, CD25, CD26, CD27, CD28, CD29, CD30, CD31, CD32, CD33, CD34, CD35, CD36, CD37, CD38, CD39, CD40, CD41, CD42, CD43, CD44, CD45, CD46, CD47, CD48, CD49, CD50, CD51, CD52, CD53, CD54, CD55, CD56, CD57, CD58, CD59, CD60, CD61, CD62, CD63, CD64, CD65, CD65, CD66, CD67, CD68, CD69, CD70, CD71, CD72, CD73, CD74, CD75, CD75, CD76, CD77, CD78, CD79, CD80, CD90, CD91, CD92, CD93, CD94, CD95, CD96, CD97, CD98, CD99, CD100, CD101, CD102, CD103, CD104, CD105, CD106, CD107, CD a) selecting individual memory B cells from among the individual isolated cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of: CobI, IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA, IgE, Sspn, Ackr2, Nt5e, and Mki67; e) assembling nucleic acid sequences encoding antibody light and heavy chain variable chains from the nucleic acid sequences of the expressed genes of the selected individual memory B cells; and f) expressing the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) in a host cell to produce an antibody. [Background technology]

[0002] Millions of people worldwide live with opioid abuse and the resulting opioid abuse disorder, with fatal overdoses increasing annually. Intervention strategies for opioid abuse disorder consist of pharmacological agonists (methadone), partial agonists (buprenorphine), and antagonists (naloxone and naltrexone) that target opioid receptors in the brain to provide therapeutic benefit. Although opioid pharmacotherapy has substantial clinical utility in medication-assisted treatment, with naloxone being a crucial rescue medication for reversing opioid overdoses, these medications have been insufficient to curb the prevalence of opioid abuse disorder and the occurrence of overdoses. Limitations of these medications include undesirable side effects, abusability or switching of agonists, the need for detoxification before initiating antagonist treatment to avoid abrupt withdrawal symptoms, and the need for frequent dosing with a high compliance burden. Therefore, complementary or alternative therapies as well as medical preventative measures are needed to complement current medications.

[0003] Immunotherapy consisting of monoclonal antibodies and vaccines offers a promising strategy to treat opioid abuse and reduce the incidence of overdose. Monoclonal antibodies and vaccine-induced polyclonal antibodies selectively alter the pharmacokinetics of targeted drugs through binding and sequestration of drug molecules in serum, preventing drug distribution to the brain without directly affecting receptor signaling. Both antibodies and vaccines have several advantages over opioid antagonists, including fewer side effects. In addition, drug therapies may require controlled detoxification to prevent rapid withdrawal, whereas antibodies and vaccines are not expected to alter endogenous opioid signaling or require detoxification. Furthermore, antibodies typically exhibit high specificity for their targets with little cross-reactivity to structurally distinct opioid agonists or antagonists. Thus, antibodies and vaccines may be considered as an alternative and / or complement to existing small molecule therapies for opioid abuse disorder.

[0004] Direct administration of specific anti-drug monoclonal antibodies may provide effective protection against targeted drugs. Drug-targeted monoclonal antibodies have demonstrated preclinical efficacy against cocaine, fentanyl, methamphetamine, nicotine, and opioids.

[0005] However, small molecule haptens such as cocaine, fentanyl, methamphetamine, nicotine, and opioids do not generally elicit strong antibody responses, and therefore such monoclonal antibodies are typically generated by conventional hybridoma techniques that require intensive screening and testing of myelomas. To simplify hybridoma generation, it has been reported that antigen-based magnetic enrichment can be used to preselect target-specific B cells prior to hybridoma fusion. Magnetic enrichment or "baiting" is often employed to increase the desired cell population for flow cytometric analysis, and single cell sorting has been used for the isolation of antigen-specific B cells and the development of recombinant monoclonal antibodies against various targets and in multiple species. Using an antigen-based enrichment platform previously validated for flow cytometric analysis of opioid-specific B cell populations, hybridomas were isolated from mice vaccinated against three commonly misused opioids: oxycodone, heroin, and fentanyl. Monoclonal antibodies using such methods have demonstrated binding to their target drugs in vitro and in vivo efficacy in reducing opioid distribution and behavioral effects when administered to mice and rats (WO2020 / 247584; WO2020 / 018596; Baehr 2020, Journal of Pharmacology and Experimental Therapeutics 375(3):469-477; Smith 2019, J Am Chem Soc. 141(26): 10489-10503). Nevertheless, reliable generation of specific, high affinity anti-drug antibodies remains highly desirable, as many existing approaches require intensive antibody screening and testing. Summary of the Invention

[0006] The technical problem underlying the present invention can be seen as the provision of means and methods for meeting the aforementioned needs. This technical problem is solved by the embodiments characterized in the claims and in the following description.

[0007] Accordingly, the present invention provides a method for producing an antibody that specifically binds to an antigen, comprising the steps of: a) contacting a B cell-containing sample from an animal, preferably a mouse, immunized with an antigen with a labeled version of that antigen; b) from that sample - cells that are CD19 positive, - cells that are CD138 negative, - Cells that bind the labeled antigen individually isolating c) determining the nucleic acid sequence of a plurality of expressed genes, preferably the entire transcriptome, for each of said isolated individual cells; d) selecting individual memory B cells from among the individual isolated cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of BhIhe41, Parm1, CD80, CobI, IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA, IgE, Sspn, Ackr2, Nt5e, and Mki67 within the nucleic acid sequence of the plurality of expressed genes; e) assembling nucleic acid sequences encoding antibody light and heavy variable chains from the nucleic acid sequences of the expressed genes of the selected individual memory B cells; and f) expressing in a host cell the nucleic acids encoding the antibody light and heavy chains assembled in step e) to produce the antibody. [Brief description of the drawings]

[0008] [Figure 1]A schematic of the workflow used to identify optimal B cell receptors (antibodies) after vaccination is shown. Spleens are harvested from vaccinated animals and splenocytes are homogenized. B cells are identified by specifically recognizing B cell surface proteins (e.g., CD19, or, as pictured, IgG) using a labeling reagent. Antigen-binding B cells are identified by contact with a fluorescently labeled version of the fentanyl hapten. Finally, single cells are sorted into 384-well plates for single cell transcriptome analysis based on their identification. [Diagram 2] Figure 1 shows the library size determined for each cell in the analysis. Top graph: cells from mouse 1. Bottom graph: cells from mouse 6. Left graph: frequency of library size values ​​for all cells in each dataset, represented as a histogram, showing the number of cells with very small library size (far left on the X-axis) relative to the number of cells with usable libraries. Cells with small library sizes are of poor quality, as RNA may not have been captured efficiently. Right graph: another depiction of the same data in a violin plot (except now individual cells are plotted) so that we can show which cells passed the quality assessment and which did not (cells in black were kept, cells in grey were eliminated from further analysis). [Diagram 3] Figure 1. Percentage of mitochondrial genes detected within each library preparation / mouse. Histograms represent the frequency of mitochondrial gene expression percentage in each dataset. The mitochondrial percentage is calculated by measuring the number of reads from the complete library that correspond to mitochondrial genes. Cells with a low percentage of mitochondrial contamination are retained during analysis. Left: Mouse 1. Right: Mouse 6. [Figure 4] Figure 1 shows the total number of features for each library. The histograms show the number of genes with non-zero counts for each dataset. Cells with a small number of unique features are removed from the analysis. [Diagram 5]Figure 1 shows the calculated variance of the standardized log-expressed values ​​for each gene in each fitted data set as a function of the mean. The curve shows the average variance of all genes. The number of genes above the average variance with an FDR below 0.05 is 62 in mouse 1 and 40 in mouse 6. [Figure 6] Figure 1 shows that fentanyl-binding B cells can be separated into distinct population clusters. UMAP plot showing the output of integration using non-negative matrix factorization applied to the filtered and joined dataset (combining data from both mice). B cells form clear clusters, with the most separated (and therefore best characterized) clusters being GC-LZ-PrePB cells (bottom left) and switched memory cells (far right). Boxed numbers (e.g., 457208_1) indicate individual cells whose antibody sequences were selected for characterization. [Figure 7-1] Figure 1 shows marker genes detected for each of the six B cell clustered subpopulations in addition to several known B cell markers. Heatmap visualizing log-standardized z-scores of the top 10 statistically significant gene markers (rows) for each population (columns) resulting from univariate differential expression adjusted for covariates, with log fold change >1.5 and adjusted p-value <0.05. Several additional known B cell markers were added to enhance the annotation of the identified subpopulations. Darker shading indicates genes that are underexpressed in a particular population, and lighter shading indicates genes that are overexpressed in a particular population. Hierarchical clustering was applied to both rows and columns to group related genes together and identify similarities between clusters based on their expression profiles. [Figure 7-2] Continued from Figure 7-1. [Figure 8]Figure 1. The most reliable memory B cell markers are represented individually. Violin plots showing log-normalized expression (y-axis) for each of the marker genes mentioned on the left, for all cells (each point is a cell), and for each discrete subpopulation (x-axis). Vertical lines represent the interquartile range. The shape of the distribution (kernel density) indicates the probability that a cell in each population will have a given value. [Figure 9] Figure 1 shows a UMAP visualizing the joint expression of Cd80, Mki67 and Ighg1 genes. Cells are replotted identically to Figure 6, except that here the coloring shows the simultaneous joint density (relative overall expression levels) of three key known memory B cell markers, which appear only in the memory population (far right). [Figure 10] Figure 1: Fentanyl-binding memory B cell compartment generates the highest affinity B cell receptors. Flow cytometry data obtained during the penultimate step of Figure 1 (Y-axis: log10 of IgG1 surface expression levels measured by mean fluorescence intensity ---X-axis: log10 of contact-mediated fluorescent fentanyl binding capacity) have been replotted here to incorporate the output of the clustering algorithm. Coloring represents different subclusters. Cells present in the upward and right-shifted population have the highest fentanyl affinity and represent cells that have "switched" (via class switch recombination) to IgG1 expression. This population is almost entirely represented by cells from the memory compartment. [Figure 11] Figure 1 shows heavy chain constant region representation within each B cell subpopulation. All bars sum up to 1.0, which represents 100% of all analyzed antibody sequences from that particular population. Populations are shown along the bottom axis, with Ig subclass keys shown on the far right. IgG1 (black) and IgG2b (dark grey) are the two most prominent over-represented classes that are almost uniquely present in memory populations. [Figure 12]Figure 1 shows variable gene usage by antibodies in the memory compartment after vaccination. Circos plots are shown. Each circos plot shows the pairing of the heavy (lower half) and light (upper half) V segments of the BCR of a single memory B cell from a memory B cell subpopulation. Left: Mouse 1. Right: Mouse 6. Highlighted (black) genes in the outer ring are shown to show the similarity between the two BCR repertoires in the context of BCR expansion in the population. The shaded cross lines indicate the part of the antibody selected for biochemical characterization (same as shown in Figure 6). [Figure 13] A schematic of the workflow used to generate antibodies is shown. Left: The antibody sequence is synthesized and cloned into an expression vector (where the Fab acquires a HIS tag). The antibody is then transiently overexpressed in suspension HEK293F cells on a rotary shaking platform. The supernatant is collected and the antibody is then purified according to the protein properties indicated. Right: Coomassie stained non-reducing SDS PAGE analysis of the purification series revealed that the antibody was successfully generated and highly pure after purification. (*) indicates a shifted band present in the Fab 609 lane; the heavy chain of this Fab is heterogeneously N-glycosylated, leading to the size shift seen here. [Figure 14] Figure 2: Structure of the antibody-fentanyl complex. A. The overall structure (space-filling representation) of the complex of Fab (FenAb609, heavy chain colored light grey, light chain colored darker grey) with fentanyl is shown as a ribbon diagram with the two-dimensional chemical structure of fentanyl shown on the left. The CDR1, CDR2, and CDR3 regions of the antibody are highlighted in very light grey, dark grey, and light grey, respectively. The partially transparent molecular surface of the protein is shown superimposed on the structure. B. A view of the fentanyl binding pocket as a thin slice through the molecular surface of the protein (colored with a gradient from white to dark grey to reflect the increasing hydrophobicity of the surface, colored using the method of Eisenberg 1984). Fentanyl is shown as a stick model with carbon, nitrogen, and oxygen atoms represented in grey, dark grey, and white, respectively. [Figure 15]Figure 1 shows that m6A and inosine binding B cells can be separated into different population clusters. The plots are generated in an identical manner to that shown in Figure 6, except that these B cells were taken from mice vaccinated with m6A and inosine. [Figure 16] Figure 16. Transcriptome profile of m6A and inosine vaccinated B cells. UMAP plot with relative expression levels (density) of a set of genes defining key memory clusters overlaid on the output from Figure 15. [Figure 17] Figure 1: m6A and inosine binding memory B cell compartment generates the highest affinity B cell receptor. FACS data replotted from m6A and inosine responsive B cells after identifying population clusters. The majority of cells in the upper right quadrant (IgG high, antigen high) are from the memory compartment. [Figure 18] Figure 1: Variable gene usage by antibodies in the memory compartment after m6A or inosine-based vaccination. Circos plots are shown. Each circos plot shows the pairing of heavy (lower half) and light (upper half) V-segments of the BCR of a single memory B cell from a memory B cell subpopulation. Left: inosine. Right: m6A. Highlighted (black) genes in the outer ring are shown to show the similarity between the two BCR repertoires. [Figure 19A] A. Promise of computational antibody selection. UMAP plot represents cells from mice immunized with fentanyl. Memory population is shaded darker and circled, while all non-memory cells are lighter and boxed. B and C. Circos plot corresponds to the cell populations shown from the UMAP plot. Heavy chain V genes expressed in each B cell are in the top row of the circos plot, and light chain V genes are in the bottom row. Only the memory population displays enriched overrepresentation of stereotyped V gene pairings that are expressed by only a small subpopulation of cells within the memory population. [Figure 19B] Continued from Figure 19A. [Figure 19C] Continued from Figure 19B. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] It should be understood that, in the specification and claims, "a" or "an" can mean one or more of the items referenced below, depending on the context in which it is used. Thus, for example, a reference to "an" item can mean that at least one of the item can be utilized.

[0010] When used hereinafter, the terms "have", "comprise" or "include" are meant to have an open-ended or closed-ended meaning. Thus, when they have a closed-ended meaning, they can refer to a situation in which no other features are present in the described embodiment in addition to the features characterized by these terms, i.e., they have a closed-ended meaning in the sense of "consisting of" or "essentially consisting of". When they have a non-close-ended meaning, they refer to a situation in which one or more other features are present in the described embodiment in addition to the features characterized by the terms.

[0011] Furthermore, in the following, the terms "preferably", "more preferably", "most preferably", "particularly", "more particularly", "typically" and "more typically" are used in conjunction with features to indicate that these features are preferred features, i.e., these terms indicate that alternative features may also be envisaged in accordance with the present invention.

[0012] Furthermore, it is understood that the term "at least one" used herein means that one or more items mentioned after the term can be used according to the present invention. For example, when a term indicates the use of at least one item, this can be understood as one item or more than one item, i.e., 2, 3, 4, 5, or any other number greater than 1. Depending on the item to which the term refers, a person skilled in the art will understand what upper limit, if any, the term can refer to.

[0013] The term "production" as used herein refers to the process of production of antibodies that specifically recognize a hapten, starting from a spleen sample of an animal immunized with the hapten, to the recombinant production of the antibody in a host cell. Production may also include further steps, such as purifying the produced antibody or formulating the antibody or the purified antibody as a pharmaceutical composition. Thus, the aforementioned method of the present invention may consist of the aforementioned steps or may further include additional steps.

[0014] The term "antibody" as used herein refers to any immunoglobulin polypeptide derived from a VDJ genomic sequence that contains an amino acid sequence stretch capable of forming a binding pocket sufficient for specific hapten binding with an equilibrium dissociation constant (Kd) in the picomolar range. Preferably, said antibody binds to the hapten with an equilibrium dissociation constant (Kd) of at most 1,000 pM, at most 800 pM, at most 600 pM, at most 400 pM, at most 200 pM, at most 100 pM or at most 75 pM.

[0015] Such antibodies may preferably be monoclonal antibodies, single chain antibodies, chimeric antibodies, humanized antibodies, or any fragment or derivative of such antibodies capable of specifically binding to the haptens referred to herein. Fragments and derivatives included within the term antibody as used herein include bispecific antibodies, synthetic antibodies, Fab, F(ab)2 Fv or scFv fragments, or chemically modified derivatives of any of these antibodies. Said antibodies, derivatives and fragments thereof can be produced by using the methods of the present invention.

[0016] The antibody according to the present invention comprises three complementarity determining regions in the chain. The term "complementarity determining region (CDR)" as used herein refers to a region in the variable domain of the heavy and light chains of an antibody that defines the binding affinity and specificity of the antibody. The heavy chain has three CDRs, CDR1-H, CDR2-H and CDR3-H, and the light chain has three CDRs, CDR1-L, CDR2-L and CDR3-L.

[0017] The three CDRs of an antibody form a binding pocket for the hapten to bind. The term "binding pocket" refers to the three-dimensional structure of an antibody required for hapten binding according to the present invention. The binding pocket comprises a sequence of amino acids with side chains that can interact with the hapten by physicochemical forces, such as van der Waals interactions, hydrogen bonds, pi-anions, pi-pi T-shapes or pi-alkyls. The binding pocket of an antibody produced according to the method of the present invention is composed of amino acids from all three complementarity determining regions (CDRs) of each chain. In addition, there may be additional amino acids from typical framework regions of the heavy and light chains that participate in the formation of the binding pocket.

[0018] Depending on the type of antibody envisaged, the antibody antigen-binding site may further comprise amino acids or amino acid sequences from framework regions. The term "framework region" (FR) refers to the amino acid sequences intervening between the CDRs, i.e., the portions of the immunoglobulin light and heavy chain variable regions that are relatively conserved among different immunoglobulins in a single species. The light and heavy chains of an immunoglobulin each have four FRs, designated FR1-L, FR2-L, FR3-L, FR4-L, and FR1-H, FR2-H, FR3-H, FR4-H, respectively. From the N-terminus to the C-terminus, both the light and heavy chain variable regions typically have these elements in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3 and FR4.

[0019] A numbering system has been established to assign numbers to the amino acids that occupy positions in each of the above domains. The complementarity determining regions and framework regions of a given antibody can be identified using the Kabat system. Typically, CDR and FR sequences are given herein according to the system described by Kabat. However, CDRs can also be redefined or otherwise determined and numbered according to alternative nomenclature based on IMGT definition.

[0020] The antibody referred to herein may also be a full-length antibody (i.e., an antibody comprising two heavy chains and two light chains). In such a case, the light chain comprises two domains or regions, a variable domain (VL) and a constant domain (CL). The heavy chain comprises four domains, a variable domain (VH) and three constant domains (CH1, CH2 and CH3, collectively referred to as CH). The variable regions of both the light chain (VL) and the heavy chain (VH) determine the binding recognition and specificity to the antigen. The constant region domains of the light chain (CL) and the heavy chain (CH) confer important biological properties, such as antibody chain association, secretion, transplacental mobility, complement binding, and binding to Fc receptors (FcR). The Fv fragment is the N-terminal portion of the Fab fragment of an immunoglobulin and consists of the variable portions of one light chain and one heavy chain. The specificity of an antibody resides in the structural complementarity between the antibody binding site and an antigenic determinant. The antibody binding site is composed of residues mainly derived from the hypervariable or complementarity determining regions. Occasionally, residues from non-hypervariable or framework regions (FR) influence the overall domain structure and thus the binding site. Human antibody light chains are generally classified as kappa and lambda light chains, each of which contains one variable region and one constant domain. Heavy chains are typically classified as mu, delta, gamma, alpha, or epsilon chains, which define the antibody isotype as IgM, IgD, IgG, IgA, and IgE, respectively. Human IgG has several subtypes, including but not limited to IgG1, IgG2, IgG3, and IgG4. Human IgM subtypes include IgM and IgM2. Human IgA subtypes include IgA1 and IgA2. In humans, IgA and IgD isotypes contain four heavy chains and four light chains, IgG and IgE isotypes contain two heavy chains and two light chains, and IgM isotypes contain 10 or 12 heavy chains and 10 or 12 light chains. Antibodies referred to herein may be IgG, IgE, IgD, IgA, or IgM immunoglobulins or fragments thereof.

[0021] A humanized antibody refers to an immunoglobulin chain or fragment thereof (e.g., Fab, Fab', F(ab')2, Fv, or other antigen-binding subsequence of an antibody) that contains minimal sequence (but typically still at least a portion) derived from a non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (recipient antibody) in which CDR residues of the recipient antibody have been replaced by CDR residues from a non-human species, such as mouse, rat or rabbit immunoglobulin (donor antibody) having the desired specificity, affinity, and capacity. Thus, at least a portion of the framework sequences of said antibodies or fragments thereof may be human consensus framework sequences. In some cases, Fv framework residues of the human immunoglobulin must be replaced by corresponding non-human residues to increase specificity or affinity. Furthermore, humanized antibodies may comprise residues that are found neither in the recipient antibody nor in the imported CDR or framework sequences. These modifications are made to further refine and maximize antibody performance. Generally, a humanized antibody will comprise substantially all of at least one, and typically at least two, variable domains, with all or substantially all of the CDR regions corresponding to those of a non-human immunoglobulin and all or substantially all of the framework regions being of a human immunoglobulin consensus sequence. A humanized antibody optimally also will comprise at least a portion of an immunoglobulin constant region, typically that of a human immunoglobulin, which (e.g., human) immunoglobulin constant region may be modified (e.g., by mutation or glycoengineering) to optimize one or more properties of such region and / or to improve the function of the (e.g., therapeutic) antibody, e.g., to increase or decrease Fc effector function or to increase serum half-life.

[0022] Chimeric antibodies refer to antibodies constructed, typically by genetic engineering, from immunoglobulin variable and constant regions whose light and / or heavy chain genes are identical or homologous to the corresponding sequences of different species, e.g., mouse and human. Alternatively, the variable region genes are from a particular antibody class or subclass, while the remainder of the chain is from another antibody class or subclass of the same or different species. It also includes fragments of such antibodies. For example, a typical therapeutic chimeric antibody is a hybrid protein composed of the variable or antigen-binding domain from a mouse antibody and the constant or effector domain from a human antibody, although other mammalian species can be used.

[0023] The term "antigen" as used herein includes any type of compound or structure capable of eliciting an immune response in a host, preferably an animal as specified herein. Typically, an antigen may be a protein, peptide, small molecule, sugar, lipid or any structure that exposes said protein or peptide, such as a microorganism or virus. The immune response referred to in this context includes a humoral immune response, i.e., it is assumed that B cells are involved and preferably undertake VDJ recombination events.

[0024] Preferably, the antigen according to the present invention is a disease-associated antigen, such as a peptide, protein, small molecule, sugar or lipid, associated with the onset or progression of a disease. Typically, a disease in this context may be selected from the group consisting of a proliferative disorder, an infectious disease, an inflammatory disease, an immunodeficiency disorder and an autoimmune disorder.

[0025] Preferably, the antigen may also be associated with a pathogen, for example a microorganism, such as a bacterial, fungal, algae, parasitic or protozoan pathogenic organism, or a virus, viroid or prion.

[0026] Preferably, the antigen may be any of the aforementioned compounds found or released in the environment. Such compounds in the environment may be of natural origin, i.e., released into the environment for natural resources, including non-living and living natural resources, such as organisms. Antigens may also be released into the environment from artificially created sources.

[0027] Preferably, the antigen may be a hapten. The term "hapten" as used herein refers to a small molecule compound. Such small molecules typically do not induce an immune response in a physiological environment due to their size and other properties. However, there are small molecule compounds, e.g., drugs, sugars, lipids, nucleotides, for which it is highly desirable to have an antibody that specifically binds, either as a therapeutic or for diagnostic purposes. For example, when bound to a hapten, the antibody may also neutralize some or all of the biological effects caused by the small molecule hapten. There are techniques that allow for efficient immunization of animals with haptens. Suitable immunization techniques are described in more detail elsewhere herein. Based on the immunized animals and the methods of the present invention, it is possible to produce antibodies against the aforementioned haptens.

[0028] Haptens as referred to herein are small molecule compounds, preferably less than 50 kD, more preferably less than 10 kD. Typically, such haptens are small molecule drugs, sugars, lipids, nucleotides and their derivatives, etc.

[0029] More preferably, the hapten is (i) delta-9-tetrahydrocannabinol (THC) or synthetic cannabinoids, such as classical cannabinoids, non-classical cannabinoids, hybrid cannabinoids, aminoalkyl indoles, and eicosanoids; for example, Δ9-THC HU-210, (C8) CP 47,497, JWH-018, AM-2201 (fluorinated JWH-018), UR-144, XLR11 (fluorinated UR-144), APICA, STS-135 (fluorinated APICA), AB-PINACA, PB-22, 5F-PB-22 (fluorinated PB-22); (ii) Methamphetamine and its derivatives, such as 3,4-methylenedioxy-methamphetamine (MDMA) Ecstasy / Molly; (iii) synthetic cathinone-like 5alpha-pyrrolidinopentiophenone (alpha-PVP); (iv) Opioids such as heroin, synthetic opioids, e.g., fentanyl or related compounds, e.g., carfentanil, and other opioid analgesics, e.g., oxycodone (OxyContin®), hydrocodone (Vicodin®), codeine, morphine, desomorphine (Krokodil); (v) steroids (anabolic substances); and (vi) Nicotine The addictive substance is selected from the group consisting of:

[0030] The opium alkaloids and derivatives according to the present invention are selected from the group consisting of phenanthrenes such as codeine; morphine; thebaine; mixed opium alkaloids such as oripavine or papaveretum; esters of morphine such as diacetylmorphine (morphine diacetate; heroin); nicomorphine (morphine dinicotinate); dipropanoylmorphine (morphine dipropionate); diacetyldihydromorphine; acetylpropionylmorphine; dma desomorphine; methyldesorfin; dibenzoylmorphine; ethers of morphine such as dihydrocodeine; ethylmorphine; and heterocodeine.

[0031] Also included are semi-synthetic alkaloid derivatives, such as buprenorphine; etorphine; hydrocodone; hydromorphone; oxycodone; oxymorphone. Also included are synthetic opioids, such as anilidopiperidines, such as fentanyl; alpha-methylfentanyl; alfentanil; sufentanil; remifentanil; carfentanil; omfentanil; and 20 phenylpiperidines, such as pethidine (meperidine); ketobemidone; MPPP; allylprodine; prodine; PEPAP; promedol. Included are diphenylpropylamine derivatives, such as propoxyphene; dextropoxyphene; dextromoramide; bezitramide; piritramide; methadone; dipipanone; levomethadyl acetate (LAAM); difenoxine; diphenoxylate; loperamide. Additionally, these include benzomorphan derivatives such as dezocine, pentazocine, and phenazocine; oripavine derivatives such as buprenorphine, dihydroetorphine, and etorphine; morphine derivatives such as butorphanol; nalbuphine; levorphanol; levomethorphan; racemethorphan; others such as lephetamine; menthol (kappa-opioid agonist); meptazinol; mitragynine; tilidine; tramadol; tapentadol; eluxadoline; AP-237; and 7-hydroxymitragynine.

[0032] Haptens used in the context of the present invention may also preferably be derivatives or variants of nucleic acids or nucleobases, for example variants of RNA or DNA nucleobases against which antibodies are desired.

[0033] Most preferably, the hapten according to the present invention is selected from the group consisting of fentanyl or a derivative thereof, N6-methyladenosine (m6A), and inosine.

[0034] Fentanyl as used herein refers to the compound N-phenyl-N-[1-(2-phenylethyl)piperidin-4-yl]propanamide. Fentanyl is listed under CAS number 437-38-7. Fentanyl is an opioid typically used as a pain medication or for anesthesia. It is also abused as a recreational drug and can lead to drug addiction. Fentanyl can be administered by various routes, for example, injection, nasal spray, transdermal (e.g., by skin patch), transmucosal, lozenge or tablet. Derivatives of fentanyl contemplated according to the present invention include structurally and / or functionally related derivatives of fentanyl. Typically, fentanyl derivatives according to the present invention are alfentanil, sufentanil, remifentanil and carfentanil. Preferably, the hapten contemplated according to the present invention is fentanyl.

[0035] m6A refers to N6-methyladenosine, a nucleoside obtained by methylation of adenosine and which can be found in mRNA, tRNA, rRNA or snRNA in various species. 11 H 15 N 5 O 4 and is listed under the CAS number 1867-73-8.

[0036] Inosine refers to the nucleoside produced when hypoxanthine is attached to a ribose ring (also known as ribofuranose) via a β-N9-glycosidic bond. Inosine can be found in tRNA and is essential for proper translation of the genetic code in wobble base pairs. It has the general formula C 10 H 12 N 4 O 5 and is listed under the CAS number 58-63-9.

[0037] It is understood that the antigens referred to herein may also be classes of similar molecules, such as sugars or lipids, or proteins or peptides sharing a common domain that is structurally related and therefore recognized by an antibody.

[0038] The phrase "specifically binds" as used in accordance with the present invention means that the antibody does not significantly cross-react with components other than the antigen, i.e., with molecules other than the specific antigen molecule or with classes of molecules other than that of the antigen class. The cross-reactivity of the antibodies referred to herein can be tested by the skilled artisan by various techniques including immunological techniques, such as Western blotting, ELISA or RIA based assays, or measuring binding affinity, for example using Biacore technology.

[0039] The term "labeled antigen" as used herein refers to an antigen linked to a label that can be used to isolate cells.Typically, the label referred to herein is a fluorescent dye that can be determined by FACS, a magnetic label that can be determined by MACS, or any other method for isolating single cells as described herein. Preferably, fluorescent dyes that may be used according to the present invention as labels for antigens are (i) single dyes, such as DyLight 405, Alexa Fluor 405, Pacific Blue, Alexa Fluor 488, FITC, DyLight 550, PE, APC, Alexa Fluor 647, DyLight 650, PerCP, or Alexa Fluor 700, (ii) StarBright dyes, such as StarBright Violet 440, 515, 610 or 670, or StarBright Blue 700, (iii) tandem dyes capable of FRET, such as PE-Alexa Fluor® 647, PE-Cy5, PerCP-Cy5.5, PE-Cy5.5, PE-Alexa Fluor® 750, PE-Cy7, or APC-Cy7, or (iv) fluorescent proteins, such as EGFP, CFP, EGFP, YFP, RFP, or mCHERRY. Preferably, the magnetic label that can be used according to the present invention as a label for antigens is Dynabeads. The label can be linked to the antigen via a permanent or reversible linkage, i.e. via a chemical bond or via a reversible chemical interaction, e.g. electrostatic interaction. The label can be linked to the antigen by a linker molecule. Depending on the nature of the label and / or the antigen, the skilled person is well aware which linker can be used.

[0040] The term "contacting" as used herein refers to bringing the labeled antigen into physical proximity with the cells contained in the spleen sample so that the cells capable of specifically binding to the labeled antigen can do so. Thus, the contact should be performed for a time and under conditions that allow specific binding of the labeled antigen to the cells. Typically, the spleen sample is contacted for a time within the range of about 15 to about 60 minutes, preferably about 30 to about 45 minutes, more preferably about 45 minutes. Typically, the conditions for contacting are: (i) staining with a live / dead stain (e.g., Thermo Scientific live / dead blue dye or propidium iodide) to remove dead cells from the analysis; (ii) blocking Fc receptors on the cells for 15 minutes to prevent non-specific antibody binding; (iii) contacting with a decoy label (a conjugate of a fluorescent label and another fluorescent label of a different color, where the former is the same label used in contacting with the antigen in the following step) for 10 minutes; (vi) contacting with a labeled hapten, e.g., fluorescent fentanyl at a 1:2000 dilution to the staining volume, for about 45 minutes; (v) staining with all primary B cell identification antibodies (e.g., anti-CD138 antibody, anti-CD19 antibody) for 45 minutes; (vi) staining with the required secondary antibodies for 15 minutes. All steps are preferably performed on ice. Washing steps between the preceding steps (i) or vi) can be performed similarly, including centrifugation and resuspension of the cell pellet in an appropriate washing solution. Most preferably, the contacting is carried out as described in the accompanying examples below.

[0041] "B cell sample" as used herein refers to a sample from an animal, as specified elsewhere herein, that contains B cells. Preferably, such a sample may be a body fluid or tissue sample that contains B cells. Preferably, the B cell sample may be a bone marrow sample or a spleen sample. The term "spleen sample" as used herein refers to a sample derived from the spleen that contains antibody-producing cells, preferably different types of B cells. The sample is typically a tissue sample that may or may not have been pre-treated to remove single cells from the spleen tissue. Preferably, the spleen sample is a homogenized whole spleen sample. The skilled person is well aware how such a spleen sample can be obtained, for example, by biopsy of a part of the spleen or by splenectomy.

[0042] The term "animal" as used herein refers to a non-human animal suitable for immunization and antibody production and from which B cell samples can be taken to isolate antibody-producing cells, preferably different types of B cells. Thus, an animal has a humoral immune system. Preferably, a suitable animal is a mammal, more preferably a laboratory animal such as a rodent, most preferably a mouse, or a domestic animal, such as a goat, sheep, pig or cow.

[0043] The term "isolate" as used herein refers to the physical separation of individual cells at the single cell level from a sample. Isolation of said cells at the single cell level can be achieved by cell sorting techniques, including, for example, fluorescence-activated cell sorting (FACS) or magnetic-activated cell sorting (MACS). Typically, cells contained in a sample are individually separated by cell sorting techniques based on the determination of a label present on the surface or within said cells. Other techniques can be based on microfluidic devices that use different microfluidic channels into which cells can enter, for example, by changing the flow path or buoyancy-activated cell sorting (BACS). Once cell sorting has been performed, the individual cells are typically maintained in a microwell plate for further analysis.

[0044] The term "individual cells" as used herein refers to a collection of isolated, i.e. physically separated, single cells.

[0045] The term "CD19" as used herein refers to cluster of differentiation 19, a B-cell surface antigen, which is a transmembrane protein expressed in all B-lineage cells, including plasma cells. CD19 plays two major roles in B cells: (i) it acts as an adaptor protein that recruits cytoplasmic signaling proteins to the membrane, and (ii) it acts within the CD19 / CD21 complex to lower the threshold of the B-cell receptor signaling pathway. Because it is present on all B cells, it is a biomarker for B-cell development, lymphoma diagnosis, and can be used as a target for leukemia immunotherapy. Human CD19 protein has been deposited under UniProt number: P15391, and mouse CD19 under UniProt number: P25918. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion, and / or addition. Typically, such variants, such as homologs, orthologs or paralogs, have amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned. The sequence identity between two amino acid sequences mentioned herein can generally be determined by alignment of the two sequences, either over the entire length of one of the sequences or within a comparison window. The percentage is calculated by determining the number of positions in both sequences where identical amino acid residues occur to obtain the number of matched positions, dividing the number of matched positions by the total number of positions in the comparison window, and multiplying the result by 100 to obtain the percentage of sequence identity. Optimal alignment and sequence identity calculation can be performed by using published techniques or computer programs, such as methods codified in BLASTP, BLASTN or FASTA. The sequence identity percentage value is preferably calculated over the entire amino acid sequence. A range of programs based on different algorithms are available to the skilled person for comparing different sequences. In this context, the algorithms of Needleman and Wunsch or Smith and Waterman give particularly reliable results.To perform sequence alignment, the program PileUp or the programs Gap and BestFit, which are part of the GCG software package (Genetics Computer Group, US), can be used. The sequence identity values ​​listed above in percentage (%) are in another embodiment of the present invention determined using the program GAP over the entire sequence region with the following settings: gap weight: 50, length weight: 3, average match: 10.000 and average mismatch: 0.000, which are always used as standard settings for sequence alignment unless otherwise stated. Antibodies that specifically bind to CD19 are available in the prior art and are described, for example, in Triller 2017, Immunity 47(6): 1197-1209 (human anti-CD19 antibodies) or Cho 2018, Nat. Commun. 9(1): 2757 (mouse anti-CD19 antibodies). They are commercially available from Thermo Fisher Scientific, US.

[0046] The term "CD138" or syndecan-1, as used herein, refers to a transmembrane (type I) heparan sulfate proteoglycan and is a member of the syndecan proteoglycan family. Syndecan-1 protein functions as an integral membrane protein and is involved in cell proliferation, cell migration and cell-matrix interactions through its receptors for extracellular matrix proteins. Syndecan-1 is a sponge for growth factors and chemokines, which it binds primarily through heparan sulfate chains. Syndecans mediate cell binding, cell signaling, and cytoskeletal organization, and the syndecan receptor is required for the internalization of HIV-1 tat protein. Alterations in syndecan-1 expression have been detected in several different tumor types. Syndecan-1 can be a marker for plasma cells. Human CD138 protein has been deposited under UniProt number: P18827, and mouse CD138 under UniProt number: P18828. However, it is understood that the term also encompasses variant proteins that differ from the proteins having the aforementioned amino acid sequences by at least one amino acid exchange, deletion and / or addition. Typically, such variants, such as homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned. Antibodies that specifically bind to CD138 are available in the prior art and are described, for example, in Cho 2018, Nat. Comun. 16(9): 2757. They are commercially available from Thermo Fisher Scientific, US.

[0047] The term "determining nucleic acid sequence" as used herein refers to determining the order of nucleotides of a nucleic acid, i.e., their sequence. Said nucleic acid sequence determination can be performed by any known DNA or RNA sequencing technology, including Sanger sequencing, pyrosequencing, next generation sequencing, sequencing by reversible terminator chemistry, sequencing by ligation mediated by ligase enzyme, phosphorus-linked fluorescent nucleotides or real-time sequencing, etc. Various technology platforms are commercially available, for example, from Roche, Illumina, or Life Technologies. Preferably, one-end sequencing of mRNA is performed according to Illumina NGS, followed by the SMART seq 2.5 library preparation protocol developed by Picelli 2014, Nature Protocols 9, 171-181 and modified by Single-cell Open Lab (scOpenLab).

[0048] The term "plurality" as used herein generally refers to a larger number of items, e.g., expressed genes referred to according to the present invention. Thus, a plurality according to the present invention refers to at least 100, at least 1,000, at least 10,000, at least 100,000 or at least 1,000,000 expressed genes. More specifically, it is envisaged that the plurality of expressed genes corresponds to the entire detectable transcriptome, i.e., the entire expressed genes of the cell interrogated by the method of the present invention, which can be detected by sequencing.

[0049] The term "expressed gene", as used herein, refers to any gene of a cell that is expressed by the cell, i.e., for which RNA, typically mRNA, can be found in the cell. In contrast to expressed genes, there are non-expressed genes, i.e., genes that are only present in the genome of the cell but are not expressed, and therefore no RNA is found in the cell.

[0050] The term "selecting", as used herein, refers to identifying an isolated individual cell and a dataset obtained therefrom, for example a dataset comprising nucleic acid sequences determined in said cell, and further evaluating said dataset of said cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of BhIhe41, Parm1, CD80, CobI, IgG1, Sspn, Ackr2, Nt5e, and Mki67 within the nucleic acid sequences of a plurality of expressed genes.

[0051] The term "memory B cells" as used herein refers to a type of B cell resulting from cell differentiation from naive B cells. Memory B cells can differentiate into plasma cells upon a second contact with an antigen. The differentiation is typically faster than that of naive B cells into plasma cells, allowing for a faster humoral immune response in a second infection. Memory B cells can survive for decades in an organism, thus acting as a memory reservoir. B cells typically have undergone class switching and can therefore express a range of immunoglobulin molecules. IgM-expressing memory B cells are typically found concentrated in the tonsils, Peyer's patches, and lymph nodes. This subset of memory B cells is likely to proliferate and re-invade germinal centers during a secondary immune response. IgG-expressing memory B cells typically differentiate into plasma cells. IgE-expressing memory B cells are very rare in healthy individuals. This may be because IgE-expressing B cells differentiate into plasma cells more frequently than memory B cells. Memory B cells expressing IgD are very rare. Concentrations of B cells with IgD only are found in the tonsils. Memory B cells as referred to in accordance with the present invention are typically those that exhibit the characteristics used for their isolation from spleen samples, i.e., they are CD19 positive, CD138 negative, and capable of specifically binding to a labeled hapten. Furthermore, memory B cells as envisaged in accordance with the present invention are those that express at least one biomarker selected from the group consisting of BhIhe41, Parm1, CD80, CobI, IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA, IgE, Sspn, Ackr2, Nt5e, and Mki67. More preferably, memory B cells as envisaged in accordance with the present invention are those that express all of the aforementioned biomarkers.

[0052] The term "BhIhe41" as used herein refers to a basic helix-loop-helix transcription factor repressor protein in various tissues in both humans and mice. BHLHE41 is also known for its role in circadian molecular mechanisms affecting the amount of sleep, as well as its role in immune function and maturation of the T helper 2 type cell lineage associated with humoral immunity. The human protein has been deposited under UniProt number: Q9C0J9, and the mouse protein under UniProt number: Q99PV5. However, it is understood that the term also encompasses variant proteins that differ from the proteins having the aforementioned amino acid sequences by at least one amino acid exchange, deletion and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0053] The term "Parm1" as used herein refers to prostate androgen-regulated mucin-like protein 1. The human protein has been deposited under UniProt number: Q6UWI2, and the mouse protein under UniProt number: Q923D3. However, it is understood that this term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion and / or addition. Typically, such variants, such as homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0054] The term "CD80" as used herein refers to cluster of differentiation 80 (also CD80 and B7-1), which is a type I membrane protein of the immunoglobulin superfamily, B7, with an extracellular immunoglobulin constant-like domain and a variable-like domain required for receptor binding. CD80 can be found on the surface of various immune cells, including B cells, monocytes, or T cells, most typically on antigen-presenting cells (APCs), such as dendritic cells. CD80 plays a vital role in regulating T cell immune function as a checkpoint protein in the immunological synapse. Expression of CD80 on B cells is associated with T cell-dependent activation in the case of T-dependent antigens. The human protein has been deposited under UniProt number: P33681, and the mouse protein under UniProt number: Q00609. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion, and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences referred to above.

[0055] The term "CobI" as used herein refers to a Cordon-bleu protein that has been demonstrated to be a brain-enriched, Wiskott-Aldrich homology 2 WH2 domain-based actin nucleator, which plays a key role in the morphogenetic process that gives rise to the complex dendrites of neurons in the vertebrate central nervous system (CNS). The human protein has been deposited under UniProt number: O75128, and the mouse protein under UniProt number: Q5NBX1. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0056] The terms "IgG1", "IgG2A", "IgG2B", "IgG3", "IgG4", "IgA" and "IgE" as used herein refer to the corresponding immunoglobulin, e.g., IgG1 refers to immunoglobulin G1, etc. All these immunoglobulin subtypes are well known. The amino acid sequences of human and mouse immunoglobulin subtypes are well known and vary depending on the antigen target. Further details regarding immunoglobulins or antibodies are also found elsewhere herein.

[0057] The term "Sspn" as used herein refers to the K-ras related polypeptide sacrospan. It is a member of the dystrophin-glycoprotein complex that spans the muscle cell membrane and is composed of dystrophin, syntrophin, alpha and beta-dystroglycan, and sarcoglycan. The human protein has been deposited under Genbank accession number XP_011519155.1. However, it is understood that the term also encompasses variant proteins that differ from the proteins having the aforementioned amino acid sequences by at least one amino acid exchange, deletion and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0058] The term "Ackr2" as used herein refers to atypical chemokine receptor 2, a beta chemokine receptor, which is a seven-transmembrane protein similar to G protein-coupled receptors. It is expressed in various tissues and hematopoietic cells. Expression of this receptor in lymphatic endothelial cells and overexpression in vascular tumors suggested its function in chemokine-driven recirculation of leukocytes and possible chemokine action on the development and growth of vascular tumors. This receptor appears to bind most of the beta-chemokine family members, but its specific function remains unclear. The human protein has been deposited under UniProt number: O00590. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences referred to above.

[0059] The term "Nt5e" as used herein refers to 5'-nucleotidase (5'-NT), also known as ecto-5'-nucleotidase or CD73 (cluster of differentiation 73). Nt5e is an enzyme capable of converting AMP to adenosine. The human protein has been deposited under UniProt number: P21589, and the mouse protein under UniProt number: Q61503. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion and / or addition. Typically, such variants, e.g., homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0060] The term "Mki67" as used herein refers to a nuclear protein associated with proliferation. The human protein has been deposited under UniProt number: P46013, and the mouse protein under UniProt number: E9PVX6. However, it is understood that the term also encompasses variant proteins that differ from the protein having the aforementioned amino acid sequence by at least one amino acid exchange, deletion and / or addition. Typically, such variants, such as homologs, orthologs or paralogs, have an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the specific sequences previously mentioned.

[0061] The term "assembling" as used herein refers to establishing the amino acid sequences of the antibody light and heavy chains from the determined nucleic acid sequences of individual isolated cells, i.e., sequence data sets. Preferably, assembling refers to establishing at least the variable amino acid sequences of the light and heavy chains, more preferably the entirety of the light and heavy chain amino acid sequences, based on the sequence data sets of individual cells. The process of assembling the amino acid sequences of said antibody light and heavy chains may include using bioinformatics tools, such as the BASIC software (Canzar 2017) described elsewhere herein, as well as pre-collected sequence data for the variable and constant regions to facilitate and / or improve the assembling process.

[0062] The term "antibody light and heavy variable chains" as used herein refers to the immunoglobulin heavy chain (IgH), which is the large polypeptide subunit of an antibody. In the human genome, the IgH locus is located on chromosome 14. An antibody is typically composed of two immunoglobulin (Ig) heavy chains and two Ig light chains, which are the small polypeptide subunits of an antibody. There are several different types of heavy chains that define the class or isotype of an antibody. The type of heavy chain varies between animals. All heavy chains contain a series of immunoglobulin domains, usually with one variable domain (VH) and several constant domains (CH1, CH2, etc.) that are important for binding to an antigen. There is only one type of light chain in a typical antibody, and therefore the two light chains of an individual antibody are identical. Each light chain is composed of two tandem immunoglobulin domains, one constant (CL) domain and one variable domain (VL), which are important for binding to an antigen. The antibody heavy and light chains assembled by the methods of the invention can be used as assembled, or their amino acid sequences can be further modified to generate antibody derivatives, such as humanized or chimeric antibodies.

[0063] The term "express" as used herein refers to the transcription and translation of nucleic acid encoding antibody light and heavy chains in a host cell such that a functional antibody is produced and secreted from the host cell. A functional antibody referred to in this context is an antibody that is capable of specifically recognizing a hapten.

[0064] The term "host cell" as used herein refers to a cell that allows the recombinant production of an antibody. Typically, such cells are genetically modified by introducing nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) of the method of the invention. The nucleic acids encoding said antibody light and heavy chains may be contained in a plasmid that can be introduced into the host cell by well-known transfection or transformation techniques. Such techniques include, for example, electroporation, lipofection, CsCl-based transfection, RbCl-based transfection, viral transfection, etc., depending on the nature of the host cell. Particularly well suited plasmids are expression plasmids, well known in the art. Preferably, the host cell referred to by the present invention may be a bacterial cell, a fungal cell or a eukaryotic cell. Most preferably, said eukaryotic cell is selected from the group consisting of HEK293T cells, CHO cells, BHK cells, NSO-GS cells, and cell lines derived from any of the aforementioned cells. The skilled person is well aware how a host cell can be cultured to obtain expression of the introduced nucleic acid of interest.

[0065] Advantageously, it has been found according to the present invention that a method comprising isolating B cells comprising a memory B cell subpopulation and individually analyzing their transcriptomes to identify individual memory B cells can be used to assemble antibody light and heavy chain variable region sequences suitable for recombinant production of antibodies that efficiently recognize antigens that contain haptens, such as fentanyl, m6A or inosine. The method allows the generation of high affinity monoclonal antibodies specific for antigens that bind with affinities in the picomolar range that can be administered at low doses. This is achieved in the method of the present invention by combining a cell sorting approach with bioinformatics to identify a subpopulation of memory B cells that produces the desired type of antibody. The method of the present invention further requires the assembly of at least a sequence for the antigen-binding structure of such an antibody, and the recombinant production of an antibody comprising such assembled sequence.

[0066] Thanks to the present invention, it is possible to efficiently provide highly specific antibodies against various antigens that can be used in antibody-based therapeutic approaches, for diagnostic or any other purpose, and to avoid the laborious screening of a large number of hybridomas, as may occur in conventional approaches for antibody generation.

[0067] In a preferred embodiment of the method of the invention, the animal is (i) administering at least one immunogenic particle displaying a plurality of antigen molecules as a priming step; and (ii) administering at least one carrier molecule displaying a single antigen molecule as a boosting step; The subject has been immunized with the antigen using an immunization method comprising:

[0068] Preferably, it is envisaged to use antigenic particles displaying VSG protein on their surface as carriers that can be coupled to antigens by suitable chemical linkages. Coupling chemistry, e.g. "click" chemistry, can be used to link antigen molecules to VSG proteins. Alternatively, modified VSG proteins can be used as carriers allowing sortagging of antigen molecules to VSG. Sortagging is an enzymatic coupling technique based on the linkage between a sortagging donor peptide and a sortagging acceptor peptide established by the enzyme sortase. This process is well established in the art and details can be found, for example, in WO 2021 / 214043. Preferably, the immunogenic particles are selected from the group consisting of liposomes, micelles, beads, vesicles and cells. Preferably, said cells are trypanosome cells, preferably inactivated, e.g. UV-inactivated, trypanosome cells, or any membrane fragments thereof. The immunogenic particles carry a plurality of antigen molecules, preferably at least 5, at least 10, at least 20, at least 30, at least 40 or at least 50 antigen molecules, via a carrier protein displayed on their surface.

[0069] The immunogenic particles referred to herein are used for priming, i.e. the first administration of the antigen to be used for immunization. Priming can be performed at one or several time points, typically twice, with 30 days between both priming immunizations. In the boosting step, i.e. step b) of the above immunization method, the antigen is administered without immunogenic particles. Typically, the VSG carrier protein coupled to the antigen is administered in a "soluble" form, i.e. not immobilized on a larger immunogenic particle. Boosting can be performed at one or several time points, typically twice, with 30 days between both boosting immunizations and 30 days between the last priming and the first boosting immunization. Further details on the method of performing immunization are described in International Publication No. WO2021 / 214043.

[0070] In a preferred embodiment of the method of the present invention, said step of isolating individual cells in step b) is carried out by using a single cell sorting technique.

[0071] Preferably, FACS is used as the single cell sorting technique, more preferably FACS as described in the accompanying examples below.

[0072] In a further preferred embodiment of the method of the present invention, the step of isolating individual cells in step b) further comprises the step of isolating individual cells that are viable cells. More preferably, the viable cells are negative for 7-aminoactinomycin (7AAD) staining.

[0073] 7AAD is a double-stranded DNA intercalating agent. It is known that 7AAD penetrates and stains cells with defective cell membranes, such as dead cells. Thus, 7AAD staining results in the staining of dead cells, but not live cells. Dead cells are preferably the ones that are selected in the method of the present invention. The selection can be preferably performed by FACS, more preferably as described in the accompanying examples below.

[0074] In a preferred embodiment of the method of the invention, said method further comprises the step of determining whether the isolated cells in step b) display IgG on their surface.

[0075] In a preferred embodiment of the method of the present invention, said step of determining the nucleic acid sequences of the plurality of expressed genes in step c) is carried out by using single cell sequencing techniques.

[0076] In a preferred embodiment of the method of the present invention, said step of determining the nucleic acid sequences of the plurality of expressed genes in step c) comprises a bioinformatics evaluation of the determined nucleic acid sequences. More preferably, said bioinformatics evaluation comprises generating a data set for each individual cell containing data reflecting an in vivo expression profile.

[0077] Also more preferably, the step of generating a dataset for each individual cell containing data reflecting said expression profile comprises: (i) removing low quality sequence reads; (ii) removing adapter sequences from the single cell sequencing; (iii) aligning the nucleic acid sequence to an indexed reference genome; (iv) compiling BAM files for the multiple nucleic acid sequences of expressed genes aligned to the reference genome for each individual cell and assigning each BAM file a quality score for the quality of the alignment of the sequences to the indexed reference genome; (v) removing low quality sequence reads based on assigned quality scores, preferably Phred scores; (vi) annotating the aligned sequences of the BAM file to chromosomal loci; (vii) compiling a count matrix dataset comprising identifiers for individual cells and identifiers for expressed genes; (viii) removing low quality matrix datasets (representing individual cells) based on the following criteria: a high percentage of mitochondrial genes in individual cells, the total number of nucleic acid sequence reads and the number of expressed genes; and (ix) removing data for individual genes that are significantly under-expressed; Includes.

[0078] The above steps can preferably be carried out as described below: (i) Removing low-quality sequence reads Generally, low quality sequence reads must be removed in this step. Low quality sequences are those with nucleotide sequences that cannot be called reliably / confidently or are too short for the final genome alignment to be reliable. This can be preferably done by using bioinformatics tools, for example the tools "trim_galore (version 0.6.4_dev)" and "cutadapt (version 1.18)".

[0079] (ii) removing adapter sequences from single-cell sequencing The sequencing process (e.g., RNAseq) adds adapter sequences to all reads to allow the sequencing platform to recognize only the nucleic acid sequences generated by the sequencing PCR. These adapter sequences are removed using bioinformatics, as they may disrupt the final genome alignment. This can also be preferably done by using bioinformatics tools, for example the tools "trim_galore (version 0.6.4_dev)" and "cutadapt (version 1.18)".

[0080] (iii) aligning the nucleic acid sequence to an indexed reference genome. As a result of the previous steps, a pool of high-quality sequence read data for each cell is obtained. In the next step, the sequenced reads are aligned to a reference genome in order to identify expressed genes. For this purpose, each read is aligned to a reference genome (during said alignment, each read obtains the chromosomal coordinates of the reference genome to which they are aligned). In this process, the bioinformatics tool "STAR" (version STAR_2.6.0a) is typically used. Preferably, for sequences derived from mouse samples, the "Release M25 (GRCm38.p6)" mouse genome and for sequences derived from human samples, the "Release 38 (GRCh38.p13)" human genome can be used as reference genomes. Genome indexing is a step that is performed only once, and the indexed genome can then be used for all future datasets for reads with the same sequencing length. However, indexing must be repeated if a new reference genome or version of the reference genome is used, or if the sequencing length of the reads changes.

[0081] (iv) compiling BAM files for the multiple nucleic acid sequences of expressed genes aligned to the reference genome for each individual cell and assigning each BAM file a quality score for the quality of the alignment of the sequences to the indexed reference genome; A compilation of multiple nucleic acid sequences of expressed genes aligned to the reference genome for each individual cell generates a genome-aligned "BAM" file for each cell that is used in the next step. These files contain the individual read information and alignment to the genome, as well as some quality criteria, such as the Phred score.

[0082] The BAM file is then typically sorted and filtered to facilitate downstream steps. This process can be done using the bioinformatics tool "Samtools" (version Samtools 1.9). The aligned transcripts at this point are mixed in the BAM file. Since information about the chromosomal location is known, the file can be sorted to group all the transcripts based on their chromosomal localization. This facilitates subsequent filtering steps of the file.

[0083] (v) removing low-quality sequence reads based on assigned quality scores; At this stage, lower quality sequences recognized at this step are removed because they cannot be fully aligned to the genome. These sequences are typically identified based on the Phred score, which is the quality score calculated in step (iii), for example during STAR alignment. Furthermore, at this stage, duplicated reads are removed. These reads have the same chromosomal coordinates, and are probably the result of PCR bias. Their presence does not necessarily reflect the actual enrichment on the mRNA level, and are removed for these reasons. Furthermore, reads that are not mapped to any position are also removed.

[0084] (vi) annotating the aligned sequences in the BAM file to chromosomal loci. Next, the aligned reads in the BAM file need to be annotated (reads identified as mapping to chromosome 1, position X need to be mapped to the identified gene ID instead). This requires the use of a number of different tools, listed in Annotation. Those skilled in the art are well aware of several options to use. For example, exons, intronic regions, or both can be mapped. Preferably, exons can be used.

[0085] (vii) compiling a matrix dataset comprising identifiers for individual cells and identifiers for expressed genes; At this point, a matrix of data has been created with the ID of the cell in each column and the ID of the gene in each row. The values ​​in the matrix are the number of reads that belong to each gene for each cell.

[0086] (viii) Removing low quality matrix datasets (representing individual cells) based on the following criteria: high percentage of mitochondrial genes in individual cells, total number of nucleic acid sequence reads, and number of expressed genes. Next, individual cells (columns) of the count matrix with poor quality are eliminated. These cells are identified as "poor quality cells" for the following reasons: Cells with a high percentage (Z) of transcript reads are mapped and annotated as mitochondrial genes. This indicates that the cells are undergoing cell death. They are not useful for further analysis, especially for characterizing the developmental stage of B cells (this is because the genes used to identify memory B cells are genomic, not mitochondrial). Also, leaving them in would pull the analysis in the wrong direction and mischaracterize some populations. Further quality control metrics are library size (total number of transcripts / reads) and number of genes identified in each cell. Cells presenting less than (X) total "features" (individual genes annotated in the transcriptome) and less than (Y) number of reads are filtered under the tendency that they are "incomplete" and therefore not informative enough to carry statistical relevance. X, Y and Z are calculated for each cell and then outlier cells (based on the median absolute deviation (MDA=3) of X, Y and Z) are removed.

[0087] (ix) removing data for individual genes that are significantly under-expressed; Next, based on representation, individual annotated genes are eliminated from the remaining individual transcriptome data sets. For example, a cell with a "good" overall transcriptome (assessed in the previous step) may still have certain genes that are not sufficiently represented to be considered statistically relevant. For example, a gene that is annotated as being present in an individual cell's transcriptome but has only one or two reads (the cutoff value applied in practice is three reads in less than five cells), that gene and its reads are eliminated. Furthermore, at this stage, transcripts that map to variable immunoglobulin (antibody) genes are eliminated. This is because antibody transcripts are highly abundant in the B cell transcriptome, thus resulting in an unfair weighting of the overall transcriptome.

[0088] In a preferred embodiment of the method for producing an antibody of the invention as described above, said bioinformatics evaluation comprises a cluster analysis of individual cells based on a data set of each individual cell containing data reflecting an expression profile. More preferably, said cluster analysis comprises: (i) normalizing expression levels for each gene between individual cell data sets by deconvolution; (ii) identifying reference genes to be clustered by modeling gene expression variability of expressed genes of cells; (iii) performing a dimensionality reduction analysis on the dataset of individual cells, preferably a principal component analysis on the dataset of individual cells which results in clustering of individual cells in different subpopulations based on features derived from said dataset; (iv) merging RNAseq datasets from different sequencing plates; Clustering the B cells in different subpopulations using integrative non negative matrix factorization; and (v) identifying individual memory B cells as cells from the subpopulation of cells identified by clustering. Includes.

[0089] The above steps can preferably be carried out as described below: (i) normalizing expression levels for each gene between individual cell data sets by deconvolution; Since the entire transcriptome has been generated at this stage, the cells need to be clustered in order to identify separable B cell populations. The first step towards this direction is to standardize the expression levels of each gene in the cells. The goal is to correct for cell-to-cell differences, remove cell-specific biases, and ensure that downstream cell-to-cell comparisons of relative expression are valid. There can be many ways to normalize gene expression. Towards the identification of rare populations, all biological variation should be maintained after normalization. Preferably, normalization is achieved by deconvolution. Typically, a pool of cells is selected and the expression profiles for these cells are summed together. The pooled expression profile is normalized to an average reference pseudocell, which is constructed by averaging the counts across all cells. This defines the size factor of the pool as the median ratio between the count sum and the average across all genes. The scaling bias of the pool is equal to the sum of the biases of the constituent cells. The same applies to the size factors, since these are effective estimates of the biases of each cell. Repeating this process for multiple pools results in a system of simultaneous equations that can be solved to obtain the size factors of the individual cells. In this way, the pool-based factors are deconvolved to yield the relevant cell-based factors. Any systematic differences in expression across the majority of genes are treated as bias, which is incorporated into the size / normalization factor and removed during scaling. The size factors calculated within each cluster must be rescaled for comparison between clusters. This is done by normalizing between pseudo cells per cluster to identify the rescaling factor. One cluster is chosen as the "reference" and all others are normalized against it. Ideally, the reference cluster should have a stable expression profile and not be extremely different from all other clusters. The assumption here is that there is a majority of non-DE between the reference cluster and each other clusters according to Aaron Lun and Karsten Bach (Lun 2016, Genome Biol. 17:75).A log transformation to the normalized expression values ​​can then be applied.

[0090] (ii) identifying reference genes to be clustered by modeling gene expression variability of expressed genes of the cells; At this stage, each cell has multiple (approximately 20,000) genes with satisfactory normalized expression levels. Many of them are housekeeping genes or are expressed at the same level for all cells. These genes are expressed by all cells, so it is not important to group cells with similar profiles together. A method to measure the variation level of genes is to calculate the variance and mean of the logarithmic representation. More specifically, for each gene, the variance and mean of the logarithmic representation values ​​were calculated. A trend is fitted to the variance relative to the mean of all genes. The fit value of each gene is used as a proxy for the technical variation component of each gene, under the assumption that most genes show a low baseline level of variation that is not biologically interesting. The biological component of the variation of each gene is defined as the residual from the trend. Ranking genes by their biological component allows the identification of interesting genes for downstream analysis in a way that accounts for the mean-variance relationship. Log-transformed expression values ​​can be used to blunt the effect of large positive outliers and ensure that large variance is driven by strong logarithmic fold changes between cells, rather than differences in counts. Since log expression values ​​are also used in downstream analyses such as PCA, modeling them here avoids discrepancies with different quantifications of variability between analysis steps. After this step, cells in different subpopulations can be clustered based on these variably expressed genes, with genes above the mean value of the cells and variance reaching this step of the analysis.

[0091] (iii) performing a dimensionality reduction analysis on the data set of individual cells, preferably a principal component analysis on the data set of individual cells which results in clustering of individual cells in different subpopulations based on features derived from said data set. From the previous step, all genes with high variance (typically v>0 and fractal dimensionality reduction <0.005) are used as input for dimensionality reduction. The main linear technique for dimensionality reduction is Principal Component Analysis (PCA). PCA performs a linear mapping of the data to a low-dimensional space such that the variance of the data in the low-dimensional representation is maximized. In the PCA plot, each cell is plotted. Cells plotted close to each other have similar transcriptomes. The length of the distance corresponds to the similarity. The longer the distance between two cells, the more dissimilar they are. This step is performed for a preliminary inspection of the data, before the integration of the data mentioned in the next paragraph. Another way to reduce the dimensionality is to perform t-Stochastic Neighbor Embedding (t-SNE) on the cells (nonlinear) based on the PCA dimensions. The existing reduced dimensionality matrix can be taken from the PCA. By default, all dimensions are used to calculate the second reduced dimensionality set. The results can be plotted to create a tSNE map showing the different subclusters on the map. Perplexity selection: The value of the perplexity parameter can have a large impact on the results. By default, the function sets a "reasonable" perplexity proportional to the number of cells in x. However, it is often useful to manually try multiple values ​​to ensure that the conclusions are robust. This is a later step, usually after merging all the data sets (obtained from different sequencing plates) into one.

[0092] (iv) Integrating the RNAseq datasets from the different sequencing plates and clustering B cells in different subpopulations using integrative non-negative matrix factorization. In a preferred embodiment of the method for producing an antibody of the invention as described above, the cell, gene and dimensional data from different experiments (survey plates) are combined into one dataset. The integration approach is based on integrated non-negative matrix factorization (iNMF) using the rliger package (Welch 2019). Preferably, datasets of individual cells obtained from different spleen samples and experiments can be combined. Thereby, technical variability is counterbalanced. Since single cells are sorted into different locations on the plate, there may be some technical variability. After combining the data for all cells from the different sequencing plates, gene normalization, gene selection based on variance (>0.1) and scaling are again performed using the corresponding functions of the rliger package. Then, iNMF is performed with the "optimizeALS" function and the datasets are combined. iNMF results in the identification of factors (factor loadings - metagenes) for each cell of the new joint dataset. Each factor corresponds to a biological signal and can characterize a specific subcluster. Joint clustering is typically performed after iNMF. Based on the factor loadings, a label is assigned to each cell of the dataset using the "quantile_norm" function and a shared factor neiborbood graph is constructed where cells sharing similar factor loading patterns are connected. In this same step, quantile standardization is performed to standardize the corresponding clusters, factors and datasets. The Louvain algorithm was applied to merge smaller subclusters together.

[0093] (v) identifying individual memory B cells as cells from the subpopulation of cells identified by clustering. We use the function "runUMAP" on the standardized factors to visualize the data in two dimensions. UMAP (Uniform Manifold Approximation and Projection) is a new manifold learning technique for dimensionality reduction and is also included as a function in the rliger package. UMAP is built from a theoretical framework based on Riemannian geometry and algebraic topology. The result is a practical, scalable algorithm to apply to real-world data. The UMAP plot is a graph that displays the "Uniform Manifold Approximation and Projection", which visually indicates how separable the classes under consideration are with respect to a group of selected features. It is a 2D plot, representing each class as a cluster of points with a unique color. To identify genetic markers, we used the runWilcoxon function from the rLiger package. Alternatively, one can use the MAST package from Finak 2015, Genome Biology 16(278) https: / / doi.org / 10.1186 / s13059-015-0844-5, available through Seurat version 4.0.

[0094] In a preferred embodiment of the method of the present invention, the step of assembling nucleic acid sequences encoding antibody light and heavy variable chains from nucleic acid sequences of a plurality of expressed genes of the selected individual memory B cells is carried out by assembling VDJ contig sequences based on the determined nucleic acid sequences encoding antibody heavy and light chains contained in the plurality of expressed genes and a reference database containing previously collected variable heavy, constant heavy, variable light, and constant light chain sequences using a comparison algorithm for assembling contig sequences. More preferably, the comparison algorithm and reference database are the BASIC algorithm and the IMGT database.

[0095] The BASIC software uses a pre-compiled database of known variable (IGHV, IGKV, IGLV) and constant (IGHC, IGKC, IGLC) region sequences in humans from IMGT (http: / / www.imgt.org). The database was indexed with Bowtie2 (version 2.2.5) into four different files corresponding to different components (IGHV; variable heavy chain, IGHC; constant heavy chain, IGK / LV; variable light chain, IGK / LC; constant light chain), which is essentially a dedicated alignment tool optimized for the assembly of antibody sequences. BASIC calls Bowtie2 to align scRNA-seq reads from a single B cell to each of the four component index files. For each component, BASIC identifies a short sequence window that contains the maximum number of aligned reads. These four sequences served as anchors to guide the assembly step. BASIC performs a de novo assembly that stitches together the anchor sequences in the heavy and light chains. It can be assumed that a sequence may overlap with either the forward sequence or the reverse complement sequence of another read. Two reads overlap if the prefix of one sequence is equal to the suffix of the other sequence, or vice versa. BASIC repeatedly extends each anchor in the 3' direction (one read at a time) until there are no overlapping reads or no repeats are found. Each anchor is then extended in the 5' direction in the same manner. For each strand, BASIC reports a single sequence if the extended sequence from the variable region anchor is equal to the extended sequence from the constant region anchor.

[0096] At this point, the VDJ contigs of the variable antibody regions have been assembled. IgBLAST (v1.16.0) can be used to annotate the assembled VDJ contigs. Using Blast on the sorted, trimmed fastq files (output of step 1.1), constant regions identifying the Ig isotype (IgG, IgM, IgA, etc.) can be identified. Sequenced antibodies may be incomplete in the context of recombinant expression. For example, the identity of the V, D, and J segments of the majority of sequences can be easily identified even when there are several "missing" segments in the sequence. To remove non-functional antibodies, a filtration step was used that aims to remove antibody sequences due to, for example, a frameshift mutation somewhere in the sequence or a truncation of one end of the VDJ. Furthermore, only antibodies in which the VDJ of both the heavy and light chains have been identified and are functional are of interest. Cells with sequence data for only one chain can be removed as well. If the tool identifies two contigs in the same cell with the same confidence percentage, we can choose the contig whose identification is based on a longer sequence alignment. (For example: the antibody gene family is so expanded that a 20bp fragment is likely to perfectly align with many different gene IDs. However, in this example, the alignment to gene A may only be 17bp worth, while the alignment to gene B may be 19bp long. Thus, we can assume that the fragment aligns to gene B.) We only considered VDJ sequences belonging to cells that showed good quality scores during the transcriptome analysis.

[0097] In a preferred embodiment of the method of the present invention, at least steps c), d) and / or e) are performed by a data processing device.

[0098] Preferably, the data processing device referred to herein is configured by at least partially tangibly embedding in said device a computer program-based algorithm for performing the method of the present invention. The processing device may comprise at least one integrated circuit configured to perform logical operations. Typically, the processor may also include at least one application specific integrated circuit (ASIC) and / or at least one field programmable gate array (FPGA). Furthermore, the data processing device may comprise a memory for storing data operatively connected thereto. The memory may be a physical data storage device, connected permanently or temporarily. Preferably, the data processing device is a computer.

[0099] In a preferred embodiment of the method of the invention, the step of expressing nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) in a host cell to produce an antibody comprises: (i) generating expression plasmids for antibody light and heavy chains; and (ii) introducing the expression plasmid into a host cell and expressing the antibody; Includes.

[0100] In a preferred embodiment of the method of the invention, said method further comprises the step of isolating said antibody from said host cell.

[0101] Isolation of the antibody from the host cell can be achieved by purifying or partially purifying the antibody from the host cell or the culture of the host cell. For protein purification, various techniques can be used, including precipitation, filtration, ultrafiltration, extraction, chromatographic techniques such as ion exchange, affinity and / or size exclusion chromatography, PLC or electrophoresis. Those skilled in the art are familiar with how antibodies can be purified to provide them in isolated form. Preferred techniques are those described in the accompanying examples below.

[0102] In a preferred embodiment of the method of the present invention, said host cell is a bacterial cell, a fungal cell or a eukaryotic cell. Preferably, said eukaryotic cell is selected from the group consisting of HEK293T cells, CHO cells, BHK cells, NSO-GS cells and cell lines derived from any of the aforementioned cells. Cell culture conditions of the aforementioned host cells that allow the expression of antibodies are well known to those skilled in the art.

[0103] In a preferred embodiment of the method of the present invention, said antigen is a hapten selected from the group consisting of fentanyl or a derivative thereof, N6-methyladenosine (m6A), and inosine.

[0104] The following embodiments are specifically contemplated in accordance with the present invention.

[0105] Embodiment 1. A method for producing an antibody that specifically binds to an antigen, preferably fentanyl or a derivative thereof, comprising: a) contacting a B cell sample from an animal, preferably a mouse, immunized with an antigen with a labeled antigen; b) from that sample - cells that are CD19 positive, - cells that are CD138 negative, - Cells that bind the labeled antigen individually isolating c) determining the nucleic acid sequence of a plurality of expressed genes, preferably the entire transcriptome, for each of said isolated individual cells; d) selecting individual memory B cells from among the individual isolated cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of BhIhe41, Parm1, CD80, CobI, IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA, IgE, Sspn, Ackr2, Nt5e, and Mki67 within the nucleic acid sequence of the plurality of expressed genes; e) assembling nucleic acid sequences encoding antibody light and heavy variable chains from the nucleic acid sequences of the expressed genes of the selected individual memory B cells; and f) expressing in a host cell the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) to produce the antibody. The method includes:

[0106] Embodiment 2. The animal is: (i) administering at least one immunogenic particle displaying a plurality of antigen molecules as a priming step; (ii) administering at least one carrier molecule displaying a single antigen molecule as a boosting step; 2. The method of embodiment 1, wherein the subject is immunized with the antigen using an immunization method comprising:

[0107] Embodiment 3. The method of embodiment 1 or 2, wherein said individually isolating cells in step b) is carried out by using a single cell sorting technique.

[0108] Embodiment 4. The method of any one of embodiments 1 to 3, wherein said individually isolating cells in step b) further comprises isolating individual cells that are viable cells.

[0109] Embodiment 5. The method of embodiment 4, wherein the viable cells are negative for 7-aminoactinomycin (7AAD) staining.

[0110] Embodiment 6. The method according to any one of embodiments 1 to 5, further comprising the step of determining whether the isolated cells in step b) express a B cell receptor of the IgG isotype.

[0111] Embodiment 7. The method according to any one of embodiments 1 to 6, wherein determining the nucleic acid sequences of the plurality of expressed genes in step c) is carried out by using single-cell sequencing techniques.

[0112] Embodiment 8. The method of any one of embodiments 1 to 7, wherein determining the nucleic acid sequences of said plurality of expressed genes in step c) comprises bioinformatics evaluation of the determined nucleic acid sequences.

[0113] Embodiment 9. The method of embodiment 8, wherein said bioinformatics evaluation comprises generating data for each individual cell that contains data reflective of an in vivo expression profile.

[0114]

[0023] Embodiment 10. The step of generating a dataset for each individual cell containing data reflecting said expression profile comprises: (i) removing low quality sequence reads; (ii) removing adapter sequences from the single cell sequencing; (iii) aligning the nucleic acid sequence to an indexed reference genome; (iv) compiling BAM files for the multiple nucleic acid sequences of expressed genes aligned against the reference genome for each individual cell and assigning each BAM file a quality score for the quality of the alignment of the sequences against the indexed reference genome; (v) removing low quality sequence reads based on assigned quality scores, preferably Phred scores; (vi) annotating the aligned sequences of the BAM file to chromosomal loci; (vii) collecting a count matrix dataset comprising identifiers for individual cells and identifiers for expressed genes; (viii) removing low quality matrix datasets (representing individual cells) based on the following criteria: a high percentage of mitochondrial genes in individual cells, the total number of nucleic acid sequence reads and the number of expressed genes; and (ix) Removing significantly under-expressed individual genes 10. The method of embodiment 9, comprising:

[0115]

[0023] Embodiment 11. The method of any one of embodiments 8-10, wherein said bioinformatics evaluation comprises a cluster analysis of individual cells based on a dataset for each individual cell containing data reflecting an expression profile.

[0116] Embodiment 12. The cluster analysis comprises: (i) normalizing expression levels for each gene between individual cell data sets by deconvolution; (ii) identifying reference genes to be clustered by modeling gene expression variability of expressed genes of cells; (iii) performing a dimensionality reduction analysis on the dataset of individual cells, preferably a principal component analysis on the dataset of individual cells which results in clustering of individual cells in different subpopulations based on features derived from said dataset; (iv) merging RNAseq datasets from different sequencing plates; clustering B cells in different subpopulations using integrated non-negative matrix factorization; and (v) identifying individual memory B cells as cells from the subpopulation of cells identified by clustering. 12. The method of embodiment 11, comprising:

[0117] Embodiment 13. The method according to any one of embodiments 1 to 12, wherein the step of assembling nucleic acid sequences encoding antibody light and heavy variable chains from the nucleic acid sequences of a plurality of expressed genes of the selected individual memory B cells is carried out by assembling VDJ contig sequences based on the determined nucleic acid sequences encoding antibody heavy and light chains contained in the plurality of expressed genes, and a reference database containing previously collected variable heavy, constant heavy, variable light, and constant light chain sequences, using a comparison algorithm for assembling contig sequences.

[0118] Embodiment 14. The method of embodiment 13, wherein said comparison algorithm and reference database is the BASIC algorithm and database.

[0119] Embodiment 15. The method of any one of embodiments 1 to 14, wherein at least steps c), d) and / or e) are performed by a data processing device.

[0120] Embodiment 16. The step of expressing in a host cell the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) to produce an antibody comprises: (i) generating expression plasmids for the antibody light and heavy chains; (ii) introducing the expression plasmid into a host cell to allow expression of the antibody; 16. The method according to any one of embodiments 1 to 15, comprising:

[0121] Embodiment 17. The method of embodiment 16, wherein said method further comprises isolating said antibody from said host cell.

[0122] Embodiment 18. The method of embodiment 16 or 17, wherein the host cell is a bacterial cell, a fungal cell or a eukaryotic cell.

[0123] Embodiment 19. The method of embodiment 18, wherein the eukaryotic cell is selected from the group consisting of HEK293T cells, CHO cells, BHK cells, NSO-GS cells, and cell lines derived from any of the foregoing cells.

[0124] Embodiment 20. The method of any one of embodiments 1-19, wherein the antigen is a hapten selected from the group consisting of fentanyl or a derivative thereof, N6-methyladenosine (m6A), and inosine.

[0125] All references cited throughout this specification are incorporated herein by reference in their entirety as well as with respect to the disclosure content specifically mentioned. EXAMPLES

[0126] The examples are illustrative of the invention and should not be construed as limiting the scope of the invention.

[0127] [Example 1] How to generate antibodies Below is a list of examples, below which we describe the output of an optimized bioinformatics pipeline designed to identify the highest quality memory B cells in a given cell population to speed up antibody hit identification. Memory B lymphocytes typically produce the highest affinity IgG class antibodies for any given hapten. This single cell sequencing approach is designed to identify these memory cells by transcriptome level expression of "memory markers". To do this, the transcriptome is analyzed using a series of bioinformatics tools described in this patent. Below, we present data generated from two mice vaccinated with haptenized fentanyl (via the vaccination platform described in International Publication No. WO2021 / 214043), followed by a summarized data set generated from mice vaccinated with haptenized m6A and inosine to support the functionality of this novel strategy.

[0128] After vaccinating mice with several doses of the hapten-coated carrier, the spleens of the immunized animals were harvested. The spleens were homogenized by grinding in ice-cold PBS or RPMI medium, passed through a cell strainer, and pelleted. The pelleted splenocytes were resuspended in FCS containing 10% DMSO for freezing. When the splenocytes were later thawed, they were subjected to a fluorescent staining protocol that allows the identification of hapten-binding B lymphocytes (characterized by surface expression and detection of CD19, the absence of CD138, and the ability to bind to fluorescently labeled haptens). The cells were then subjected to flow cytometry and single-cell sorting to isolate individual antigen-binding B cells in preparation for single-cell sequencing. A scheme of this entire process is described in Figure 1. cDNA libraries were prepared using the SMART platform library preparation protocol, and importantly, a template switching oligo (TSO) was incorporated with an oligo-dT primer during the reverse transcription step to ensure that the 5' end of the original RNA molecule was fully captured, which is particularly relevant for the highly variable VDJ regions of the antibody locus. The derived cDNA libraries were then subjected to pre-amplification and multiple clean-up steps prior to sequencing. Next-generation sequencing (NGS) of 75 base pair reads was performed on the Illumina platform using a NextSeq 550 system to generate raw transcriptome data.

[0129] It was subsequently confirmed that the splenocyte harvesting protocol was capable of preparing high-quality cells amenable to this sequencing strategy. To do this, raw sequencing reads were first aligned to the mouse (Mus musculus) reference genome (release version M25) and indexed using Samtools. After the reads were aligned, various quality scores were assessed to identify low-quality cells (i.e., cells that were dead or in the process of dying at the time of RNA isolation). Low-quality cells typically have low overall library read depth / diversity (presumably as a result of cell death or poor handling during library preparation), high expression of mitochondrial genes, and low numbers of total features (genes represented). Figures 2-4 reveal that the majority of harvested and sequenced cells were of good quality, while cells that did not meet the quality cutoff for each of these metrics were eliminated from subsequent analysis.

[0130] Next, the good quality cell transcriptomes were further analyzed to identify memory B cell populations. Briefly, genes that are differentially expressed by each individual cell at statistically relevant levels are identified. However, there may be differentially expressed genes that are biologically meaningless in the context of B cell subclass identification (e.g., genes that play a housekeeping role whose levels oscillate depending on cell cycle progression or nutritional status) and there may be biologically meaningful genes that just do not present a significant level of variance between cell types. Furthermore, there may be differentially expressed genes that define a subpopulation in one individual animal that are found to be insignificant in another animal, in turn indicating that the gene(s) in question are not reproducibly reliable subclass markers. To account for these issues, the average variance between genes was calculated (Figure 5). Markers were selected from genes that scored above the average variance in both animals and were shown to be statistically overexpressed within definable clusters after a clustering algorithm run simultaneously (the output of which is presented in Figure 6). The algorithm can be manually modified to assume the presence of a certain number of distinct clusters based on both visual and statistical confidence in the output. Here, we have taken the prediction that six clusters are present in their B-lymphocyte dataset with the greatest confidence. For each of these six clusters, the top 20 statistically relevant gene markers are presented in the heatmap of FIG. 7. Some of these markers have been previously defined in the literature, while others were identified during these studies. Many of these genes are also presented in the violin plot of FIG. 8, revealing the extent to which certain markers are over-represented in (and therefore define) the memory population. Finally, FIG. 9 shows a combined representation of the over-expression of three known memory cell markers in a UMAP plot, which can be easily visualized when compared to FIG. 6.

[0131] As a first validation of the hypothesis that memory B cells code for the highest affinity antibodies, the FACS data obtained during the workflow in Figure 1 were replotted to highlight the variance of the various cell populations within the data. Indeed, the majority of cells with the highest fentanyl binding capacity, which also generally produced class-switched IgG as expected (as determined by this method), were from the memory population (Figure 10). Class switching was also further validated bioinformatically after assembling antibody gene sequences in these cells (using BASIC, a tool that combines known antibody sequences deposited in the IMGT database with the raw reads present in a given dataset to de novo assemble novel antibody sequences). The memory population presented, at the sequence level, a notable increase in the representation of IgG1 and IgG2b (here, it should be noted that the specific switched subclass(es) may ultimately vary depending on the hapten used for immunization) antibodies compared to all other populations (Figure 11). We then plotted the sequenced antibodies in a "circos plot" to highlight the heavy and light chain variable region genes that were paired together in individual B cells. Notably, the same heavy and light chain pairs were present and overexpressed in memory cells of both mice (Figure 12), further suggesting that this pipeline can reproducibly identify the best antibody candidates (in this respect, overexpressed pairings are presumably enriched due to in vivo affinity-based selection; verified below).

[0132] The utility of the selection method discussed herein is best illustrated in FIG. 19. The combination of repertoire analysis and transcriptomics reveals a small subset of B cells that encode the best quality antibodies. And while transcriptomics alone can separate memory B cells from the rest (FIGS. 6 and 7), there is still a significant number of memory B cells that encode seemingly random, poor quality antibody sequences (FIG. 12). And when repertoire analysis is performed on the entire B cell dataset, the enrichment of the minor population goes unnoticed, given the vast number of poor quality candidates that mask it. Indeed, the combination of transcriptomics (FIG. 19A) and repertoire analysis reveals a small population of high quality antibodies (FIG. 19B). This cumulative selection method therefore enriches in three separate steps: 1. capturing only cells capable of interacting with the antigen of interest by FACS, 2. identifying memory populations by transcriptomics, and 3. identifying a minority population of high quality candidates by antibody sequence analysis (repertoire analysis), thus preventing unnecessary and costly screening of hundreds to thousands of low quality candidates (Figure 19B - unshaded areas, Figure 19C).

[0133] [Example 2] Antibody cloning and function confirmation To further test the hypothesis that memory compartments generate high affinity antibodies, a series of monoclonal antibodies were cloned, expressed, and biochemically purified following the general workflow in Figure 13. The purified antibodies were subjected to a series of characterizations, including affinity measurements by Octet, revealing that all 11 memory cell-derived mAbs bind haptenized fentanyl with sub-nanomolar affinity (Table 1). Next, to better understand the fentanyl binding mechanism, a selection of these mAbs were crystallized while bound to fentanyl or fentanyl hapten. This revealed that the fentanyl molecule is captured by the antibody with a binding mode in the form of a recessed pocket. Figures 14A and B reveal that the binding pocket is about 15 angstroms deep, which promotes high affinity by completely surrounding the fentanyl molecule.

[0134] [Table 1]

[0135] Antibodies were subjected to octet assays using biotinylated fentanyl hapten as bait. The affinities reported here are well supported by the raw data, except for those represented by the "<70pM" values. These affinities were too high for the instrument and its algorithms to call reliably, so we conservatively called them higher than the highest value recorded reliably (that of mAb 440; heavy and light chains see SEQ ID NOs: 1 and 2).

[0136] Considering that a series of sub-nanomolar affinity antibodies against fentanyl could be successfully identified with a 100% success rate, we next proceeded to validate the functionality of the sequencing and bioinformatics workflow with a different set of antigens from another molecular class. Spleens of mice vaccinated with m6A and inosine (belonging to the molecular class of nucleosides) were harvested and processed as in Figure 1. This transcriptomic dataset was subjected to and evaluated with the same filters and quality metrics as in Figures 2-5. Then, to evaluate the functionality of the pipeline, we applied a B-cell clustering algorithm, again obtaining a reliable definition of six clusters per dataset (Figure 15). Generally speaking, the same markers identified during the fentanyl-based studies were identified in these studies as well, some of them represented in the plot shown in Figure 16.

[0137] To test the hypothesis that these memory cells again possess the highest affinity B cell receptors, the original FACS data was replotted with the cluster identification of each individual cell superimposed on the data. In both cases (m6A and inosine vaccinated splenocytes), the cells with the highest m6A or inosine binding capacity and the highest levels of surface IgG1 (confirming that class switching had occurred) were from the memory compartment (Figure 17).

[0138] Circos plots were reassembled based on the variable region genes that are paired together in individual memory cells. Although there is less obvious overrepresentation of any particular gene (compared to the fentanyl study), there are a few genes that appear to be more common from a visual point of view in both data sets (e.g., heavy chain IHGV10-1, FIG. 18). This association is supported by the fact that the antibody repertoires presented in these data are similar to each other, considering that these two different antigens are also similar to each other (FIG. 18). Therefore, it is reasonable to assume that similar antibodies are selected in mice vaccinated with either m6A or inosine, which is indeed revealed by these data.

[0139] References WO2020 / 247584 WO2020 / 018596 Baehr 2020, Journal of Pharmacology and Experimental Therapeutics 375(3): 469-477 Smith 2019, J Am Chem Soc. 141(26): 10489-10503 Triller 2017, Immunity 47(6): 1197-1209 Cho 2018, Nat. Commun. 9(1): 2757 Cho 2018, Nat. Comun. 16(9): 2757 Picelli 2014, Nature Protocols 9, 171-181 WO2021 / 214043 Lun 2016, Genome Biol. 17:75 Finak 2015, Genome Biology 16(278) https: / / doi.org / 10.1186 / s13059-015-0844-5 Eisenberg, 1984, J. Mol. Biol. 179, 125-142 Canzar 2017, BioinformaticsBioinformatics, 33 (3): 425-227 Welch 2019, Cell 177(7): 1873-1887

Claims

1. 1. A method for producing an antibody that specifically binds to an antigen, preferably fentanyl or a derivative thereof, comprising the steps of: a) contacting a B cell sample from an animal, preferably a mouse, immunized with an antigen with a labeled antigen; b) from that sample - cells that are CD19 positive, - cells that are CD138 negative, - Cells that bind the labeled antigen individually isolating c) determining the nucleic acid sequence of a plurality of expressed genes, preferably the entire transcriptome, for each of said isolated individual cells; d) selecting individual memory B cells from among the individual isolated cells by identifying the presence of a nucleic acid sequence of one or more expressed genes selected from the group consisting of BhIhe41, Parm1, CD80, CobI, IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA, IgE, Sspn, Ackr2, Nt5e, and Mki67 within the nucleic acid sequence of the plurality of expressed genes; e) assembling nucleic acid sequences encoding antibody light and heavy variable chains from the nucleic acid sequences of the expressed genes of the selected individual memory B cells; and f) expressing in a host cell the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) to produce the antibody. The method includes:

2. The animal is (i) administering at least one immunogenic particle displaying a plurality of antigen molecules as a priming step; and (ii) administering at least one carrier molecule displaying a single antigen molecule as a boosting step; The method of claim 1, wherein the subject is immunized with the antigen using an immunization method comprising:

3. 3. The method according to claim 1 or 2, wherein the step of isolating individual cells in step b) is carried out by using a single cell sorting technique.

4. 4. The method of any one of claims 1 to 3, wherein isolating individual cells in step b) further comprises isolating individual cells that are viable cells.

5. The method of claim 4, wherein the viable cells are negative for 7-aminoactinomycin (7AAD) staining.

6. The method according to any one of claims 1 to 5, further comprising the step of determining whether the isolated cells in step b) express a B cell receptor of the IgG isotype.

7. The method according to any one of claims 1 to 6, wherein determining the nucleic acid sequences of the plurality of expressed genes in step c) is carried out by using single-cell sequencing techniques.

8. The method according to any one of claims 1 to 7, wherein determining the nucleic acid sequences of the plurality of expressed genes in step c) comprises bioinformatics evaluation of the determined nucleic acid sequences.

9. 9. The method of claim 8, wherein the bioinformatics evaluation comprises generating a dataset for each individual cell containing data reflecting an in vivo expression profile.

10. generating a dataset for each individual cell containing data reflecting said expression profile, (i) removing low quality sequence reads; (ii) removing adapter sequences from the single cell sequencing; (iii) aligning the nucleic acid sequence to an indexed reference genome; (iv) compiling BAM files for the multiple nucleic acid sequences of expressed genes aligned to the reference genome for each individual cell and assigning to each BAM file a quality score for the quality of the alignment of the sequences to the indexed reference genome; (v) removing low quality sequence reads based on assigned quality scores, preferably Phred scores; (vi) annotating the aligned sequences in the BAM file to chromosomal loci; (vii) compiling a count matrix dataset comprising identifiers for individual cells and identifiers for expressed genes; (viii) removing low quality matrix datasets (representing individual cells) based on the following criteria: a high percentage of mitochondrial genes in individual cells, the total number of nucleic acid sequence reads and the number of expressed genes; and (ix) removing data for individual genes that are significantly under-expressed; 10. The method of claim 9, comprising:

11. 11. The method of any one of claims 8 to 10, wherein the bioinformatics evaluation comprises a cluster analysis of individual cells based on a dataset for each individual cell containing data reflecting an expression profile.

12. The cluster analysis (i) normalizing expression levels for each gene between individual cell data sets by deconvolution; (ii) identifying reference genes to be clustered by modeling gene expression variability of expressed genes of cells; (iii) performing a dimensionality reduction analysis on the dataset of individual cells, preferably a principal component analysis on the dataset of individual cells which results in clustering of individual cells in different subpopulations based on features derived from said dataset; (iv) integrating the RNAseq datasets from the different sequencing plates and clustering the B cells in different subpopulations using integrative non-negative matrix factorization; and (v) identifying individual memory B cells as cells from the subpopulation of cells identified by clustering.

12. The method of claim 11, comprising:

13. The method according to any one of claims 1 to 12, wherein assembling nucleic acid sequences encoding antibody light and heavy variable chains from nucleic acid sequences of a plurality of expressed genes of the selected individual memory B cells is carried out by assembling VDJ contig sequences based on the determined nucleic acid sequences encoding antibody heavy and light chains contained in a plurality of expressed genes and a reference database containing previously collected variable heavy chain, constant heavy chain, variable light chain, and constant light chain sequences using a comparison algorithm for assembling contig sequences.

14. The method of claim 13 , wherein the comparison algorithm and reference database is a BASIC algorithm and database.

15. expressing the nucleic acid sequences encoding the antibody light and heavy chains assembled in step e) in a host cell to produce an antibody; (i) generating expression plasmids for the antibody light and heavy chains; and (ii) introducing the expression plasmid into a host cell to allow expression of the antibody; The method according to any one of claims 1 to 14, comprising:

Images