Patent Application
20250011447
The present application claims priority to Chinese Patent Application No. 202110704255.6, entitled “Bispecific antibody binding to BCMA and CD3, and preparation method therefor and use thereof”, and filed on Jun. 24, 2021, the entire disclosure of which is incorporated herein by reference in its entirety.
The disclosure provided herein relates to the technical field of genetic engineering antibody, and in particular to a bispecific antibody binding to BCMA and CD3 and preparation method therefor and use thereof.
Monoclonal antibodies can be used clinically for treatment of tumors, immune disease and infection, which are most widely used in tumor therapy. The Monoclonal antibody therapy against tumor is a kind of immunotherapy directed against specific targets of diseased cells to stimulate the immune system to kill target cells. To further improve drug efficacy and reduce side effects, new monoclonal antibodies-based medicines, including antibody drug conjugate (ADC), bispecific antibody (bsAb), chimeric antigen receptor T cells (CAR-T) and other new protein and cell-based drugs are developing rapidly. The cell-bridging bispecific antibody based on bispecificity has significantly improved cell killing activity and specificity due to its ability of connecting immune T cells and targeted cancer cells, which has groundbreaking clinical value, thereby attracting extra attention. For multiple myeloma, the cell-bridging bispecific antibody targeting BCMA with the above properties is one of the development directions to improve the effect of the antibody therapy and has become a hot spot in the field of antibody engineering research.
The CD3 molecule on the surface of T cells is composed of four subunits: δ, ε, γ, and ζ, with the molecular weight of 18.9 kDa, 23.1 kDa, 20.5 kDa, and 18.7 kDa, with the length of 171, 207, 182, and 164 amino acid residues, respectively. Together, they form 6 peptide chains, and form one active intact TCR complex, whose structural diagram is shown in graph A of
BCMA (B-cell maturation antigen, TNFRSF17, CD269) is a non-glycosylated type I transmembrane protein, which a member of the tumor necrosis (TNF) receptor superfamily that is preferentially expressed in differentiated plasma cells, involved in maturation, growth and survival of B cells. BCMA is originally reported as a Golgi integral membrane protein of human mature B lymphocytes, i.e., as an intracellular protein, suggesting that BCMA appears to have an important role in B-cell development and homeostasis.
BCMA expression is restricted to the B-cell lineage, which is predominantly expressed on plasma cells and plasma blasts (Graph B,
BCMA is an extremely important B-cell biomarker that is widely expressed on the surface of MM cells, which is a popular immunotherapy target for MM and other hematological malignancies. BCMA has always been the focus of three consecutive ASCO annual conferences. BCMA has become the second most popular target for anticancer therapy after CD19, according to a new analysis by the National Cancer Institute (CRI). MM is heterogeneous and is mostly caused by chromosomal translocation of t(11:14), t(4:14), t(8:14), del(13), del(17) (among others). Patients with MM may experience a variety of disease-related symptoms due to bone marrow infiltration, bone destruction, renal failure, immunodeficiency, and the psychological burden of a cancer diagnosis.
MM remains difficult to treat, since side effects always occur during treatment such as chemotherapy and stem cell transplantation, despite their efficacy in improving survival. Almost all patients may eventually experience relapse, despite significant advances in chemotherapy, proteasome inhibitors, immunomodulators thalidomide derivatives, and CD38-targeting antibodies. Therefore, there is still an urgent need for new medicines in this field. To date, the two most used treatment options for multiple myeloma patients include a combination of steroids, thalidomide, lenalidomide, bortezomib, or a variety of cytotoxic agents, as well as high-dose chemotherapy combined with autologous stem cell transplantation for younger patients. Most procedures are autologous transplantations, which use the patient's own cells. Although the transplantation is not curative, it has been shown to prolong the survival of selected patients, which can be given as the initial therapy in newly diagnosed patients or at the time of relapse. Sometimes, more than one kind of transplants may be recommended for selected patients for adequate disease control. The combination of chemotherapy agents including cyclophosphamide, doxorubicin, vincristine, and melphalan, and immunomodulators such as thalidomide, lenalidomide, bortezomib, and corticosteroids (e.g., dexamethasone) has become an important option for the treatment of myeloma in newly diagnosed patients and those with advanced disease for who experience treatment failure with chemotherapy or transplantation.
The current therapies for MM are generally non-curative. The stem cell transplantation may not be an option for most patients, because of advanced age, the presence of other serious medical conditions, or other physical limitations. Chemotherapy can only achieve partial control of multiple myeloma, rarely reaching complete remission. Therefore, there is an urgent need for new innovative therapies.
The antagonistic BCMA-specific antibodies prevent NF-κB activation associated with a potent prosurvival signal-transduction pathway in normal and malignant B-cells.
Other therapies against hematogenous neoplasms or autoimmune disorders focus on the interaction between BAFF, APRIL (i.e., ligands of the TNF ligand superfamily) and receptors TACI, BAFF-R, and BCMA activated by BAFF and/or APRIL. For example, atacicept (TACI-Ig) is created by combining the Fc-domain of human immunoglobulin to Zymogenetics' TACI to neutralize both ligands and prevent receptor activation. Atacicept is currently in clinical trials for the treatment of systemic lupus erythematosus (SLE, phase III), multiple sclerosis (MS, phase II) and rheumatoid arthritis (RA, phase II), as well as malignant B-cell chronic lymphocytic leukemia (CLL), non-Hodgkin's lymphoma (NHL) and MM in phase I clinical trials. In preclinical studies, atacicept reduced the growth and survival of primary MM cells and MM cell lines in vitro and in vivo, suggesting the relevance of TACI ligands to MM cells. Given that most MM cells and their derived cell lines express both BCMA and TACI, their receptors may favor ligand-mediated growth and survival. These data suggest that antagonistic effect against BCMA and TACI may be conducive to the treatment of plasma cell disorders. In addition, BCMA-specific antibodies that cross-react with TACI have been described (WO02/066516).
Human Genome Sciences and GlaxoSmithKline have developed an antibody called belimumab that targets BAFF. Belimumab blocks the binding of soluble BAFF to its receptors BAFF-R, BCMA, and TACI on B cells. Belimumab does not directly bind B cells, but inhibits the survival of B cells, including autoreactive B cells, and reduces the differentiation of B cells into immunoglobulin-producing plasma cells through binding to BAFF.
Even though BCMA, BAFF-R and TACI (i.e., B cell receptors of the TNF receptor superfamily) and their ligands BAFF and APRIL are subject to treatment options against cancer and/or autoimmune disorders, there remains a need for other available options for the treatment of this medical condition.
The first objective of the present disclosure is to provide a bispecific antibody binding to BCMA and CD3 and its active fragments.
The second objective of the present disclosure is to provide nucleic acids encoding the above-mentioned antibody or its active fragments.
The third objective of the present disclosure is to provide the use of the aforementioned antibody or its active fragments.
Specifically, the present disclosure provides the following technical solutions:
Firstly, the present disclosure provides a bispecific antibody binding to BCMA and CD3, consisting of:
CDR1, CDR2 and CDR3 in the light chain variable regions of the first domain have amino acid sequences represented by SEQ ID NOs. 1-3 respectively or have amino acid sequences containing one or more of the following mutations when taking the amino acid sequences represented by SEQ ID NOs. 1-3 as reference sequences:
The present disclosure screens the genetically engineered single-chain antibody against BCMA from a fully synthetic single-chain human antibody library, obtains the variable region gene sequences of the antibody, constructs the mutation library through a point mutation kit, to obtain clones with high affinity. The DNA of these clones is mixed, and the combinatorial library of single-chain antibody is assembled through recombination, and the BCMA antibodies with high affinity to human BCMA is obtained after screening.
The amino acid sequence pattern of the antibody variable region provided by the present disclosure is FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. In the present application, the division of regions of FR and CDR is based on Kabat Numbering. FR1 to FR4 herein represent four framework regions, and CDR1 to CDR3 indicate three hypervariable regions. FR1 to FR4 can be isolated from the constant region sequence (such as the most used amino acids of human immunoglobulin light and heavy chain class, subclass or subfamily), or isolated from a separate human antibody framework region or from combination of different framework region genes.
After further screening, the present disclosure provides the following bispecific antibodies, with the first domain having higher affinity to BCMA:
In the light chain variable region, CDR1 has the amino acid sequence represented by any one of SEQ ID NOs. 1 and 12, CDR2 has the amino acid sequence represented by SEQ ID NO. 2, and CDR3 has the amino acid sequence represented by any one of SEQ ID NO. 3, 13, 14, 15 and 16.
Furthermore, among the above-mentioned antibodies that bind to BCMA, the following antibodies exhibit higher affinity to BCMA:
CDR region sequences in the heavy chain variable region are any of the following:
The CDR region sequences in the light chain variable region is any of the following:
Furthermore, among the above-mentioned antibodies binding to BCMA, the following antibodies exhibit higher affinity to BCMA:
CDRs in the heavy chain variable region and CDRs in the light chain variable region is any of the following:
Based on the sequences of the above-mentioned CDRs, antibodies (the first domain) with the following variable region sequences exhibit higher affinity to BCMA:
Based on the above heavy chain variable region and light chain variable region, the amino acid sequences having at least one of the following two features are also within the scope of the present disclosure: a) binding to the same antigenic epitope; and b) having a sequence identity greater than 70%, 80%, 85%, 90%, 97%, 98% or 99%.
Furthermore, antibodies (the first domain) with the following variable region sequences exhibit higher affinity to BCMA:
In the bispecific antibodies of the present disclosure, the first domain is an antibody capable of binding to BCMA, consisting of two complete heavy chain-light chain pairs, wherein each pair of light chain and heavy chain is connected by disulfide bonds.
The Fc fragment of the heavy chain of the first domain may be Fc fragment of human or humanized antibodies (IgG1, IgG2, IgA, IgE, IgM, IgG4 or IgD).
As one embodiment of the present disclosure, the Fc fragment of the heavy chain of the first domain is Fc fragment of human or humanized IgG4 antibodies.
As a preferred embodiment of the present disclosure, the heavy chain of the first domain has the amino acid sequence represented by SEQ ID NO. 32, and the light chain has the amino acid sequence represented by SEQ ID NO. 33.
Or, the heavy chain has the amino acid sequence represented by SEQ ID NO. 49, and the light chain has the amino acid sequence represented by SEQ ID NO. 42.
Or, the heavy chain has the amino acid sequence represented by SEQ ID NO. 50, and the light chain has the amino acid sequence represented by SEQ ID NO. 44.
Or, the heavy chain has the amino acid sequence represented by SEQ ID NO. 51, and the light chain has the amino acid sequence represented by SEQ ID NO. 46.
Or, the heavy chain has the amino acid sequence represented by SEQ ID NO. 52, and the light chain has the amino acid sequence represented by SEQ ID NO. 48.
These antibodies binding to BCMA have high affinity to BCMA, and can ensure that the functions of both antibodies can be fully exerted when they form bispecific antibodies with antibodies binding to CD3.
On the basis of the above heavy chain and light chain sequences, amino acid sequences having at least one of the following features are also within the scope of the present disclosure: a) binding to the same antigenic epitope; and b) having a sequence identity greater than 70%, 80%, 85%, 90%, 97%, 98% or 99%. The abovementioned human anti-human BCMA antibodies provided by the present disclosure binds to human BCMA with an affinity of 0.2 nM to 10 nM, which inhibit BCMA ligand binding to human BCMA, but can bind to cells expressing BCMA including human multiple myeloma cells or lymphoma cells.
For the second domain binding to CD3 antigen, the heavy chain variable region has the amino acid sequence represented by SEQ ID NO. 28, and the light chain variable region has the amino acid sequence represented by SEQ ID NO. 29.
Preferably, the light chain variable region and the heavy chain variable region of the second domain are connected by linker peptides to form a single-chain antibody, which has the amino acid sequence represented by SEQ ID NO. 31.
The bispecific antibody of the present disclosure is preferably designed with the following structure: the first domain contains two complete light chain-heavy chain pairs, the second domain contains two single-chain antibodies, which are connected in any of the following manners to form the symmetrical structure:
The present disclosure finds that for the sequences of the above-mentioned first domain and the second domain, bispecific antibodies with the above-mentioned symmetrical structure can better retain the specific antigen-binding ability of the original antibody of the first domain and the second domain than those with other structures, with excellent biological function of binding to BCMA and CD3, and obvious advantages in manufacturing technique and pharmaceutical properties. The present disclosure has developed a BCMA- and CD3-binding bispecific antibody with the abovementioned molecular structure, which has a specific targeting effect and can efficiently stimulate a directed immune response to kill tumor cells.
The amino acid sequences of the linker peptides for connecting the first domain and the second domain are preferably (GGGGX)n, wherein X is Gly or Ser, and n is a natural number of 1 to 4.
As a preferred embodiment of the present disclosure, the amino acid sequence of the linker peptide is represented by SEQ ID NO. 30.
As an example of bispecific antibodies with the aforementioned structure, the present disclosure provides a bispecific antibody against human CD3 and BCMA. In the structures and sequences of the heavy chain variable region and the light chain variable region of the above-mentioned single-chain antibodies, and the heavy chain and the light chain of monoclonal antibodies, the present disclosure has constructed two bispecific antibodies that can simultaneously bind to CD3 and BMCA, with structures and sequences listed as follows:
The sequences represented by SEQ ID NOs. 1-52 disclosed and claimed above comprise “conservative sequence modifications”, that is, nucleotide and amino acid sequence modifications that do not significantly affect and change the binding features of antibodies or those containing the indicated amino acid sequence. The conservative sequence modifications include substitutions, additions or deletions of nucleotide or amino acid. The modifications can be introduced into SEQ ID NOs. 1-52 by standard techniques in the field, such as site-directed mutagenesis and PCR-mediated mutagenesis. The conservative amino acid substitutions include substitution of an amino acid residue with an amino acid residue having similar side chains or other amino acid residues. In the field, the families of amino acid residues with similar side chains have been defined, consisting of amino acids with basic side chains (such as lysine, arginine, histidine), amino acids with acidic side chains (such as aspartic acid, glutamic acid), and amino acids with uncharged polar side chains (such as glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine, tryptophan), amino acids with non-polar side chains (such as alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine), amino acids with β-branched side chains (such as threonine, valine, isoleucine), and amino acids with aromatic side chains (eg, tyrosine, phenylalanine, tryptophan, histidine). Therefore, it is preferred to replace a non-essential amino acid residue in a human anti-BCMA antibody with another amino acid residue from the same side chain family.
Therefore, antibodies having the amino acid sequence disclosed above and/or those containing the amino acid sequence disclosed above, including antibodies substantially encoded by similar sequences of conservative sequence modifications, or those containing similar sequences of conservative sequence modifications, are considered to be within the scope of the present disclosure.
The present disclosure also provides nucleic acids encoding the bispecific antibody binding to BCMA and CD3.
Considering the degeneracy of codons, the gene encoding the indicated antibody in the present disclosure can be modified in its coding region without changing the amino acid sequence, to obtain the gene encoding the same antibody. A person skilled in the art can artificially synthesize and modify the gene according to the codon preference of the host expressing the antibody, to improve the expression efficiency of the antibody.
The present disclosure also provides biological materials containing the nucleic acids encoding the bispecific antibodies binding to BCMA and CD3, including recombinant DNA, expression cassette, vector, host cell, engineering bacteria or cell line.
The indicated vectors include but are not limited to cloning vectors and expression vectors and may be plasmid vectors, viral vectors, and transposons.
The indicated host cell or cell line may be a cell or cell line derived from microorganisms or animals.
The present disclosure provides the preparation method of the bispecific antibody binding to BCMA and CD3, which consists of constructing expression vectors containing encoding genes of the first domain and the second domain; introducing the expression vectors into host cells, to obtain host cells expressing the bispecific antibody: culturing the host cell, and obtaining the bispecific antibody through separation and purification.
When preparing the bispecific antibody, a person skilled in the art can select conventional host cells, expression vectors, ways for introducing expression vectors into host cells, and antibody isolation and purification methods as required.
Based on the function of the bispecific antibody binding to BCMA and CD3 provided by the present disclosure, the present disclosure lists any one of the following uses of the bispecific antibody binding to BCMA and CD3 or its encoding nucleic acid or biological materials containing the nucleic acid:
The above-mentioned BCMA-expressing B cells-related diseases are preferably malignant tumors and autoimmune diseases with high expression/overexpression of BCMA.
The bispecific antibody provided by the present disclosure can be used for treatment and diagnosis. Antibodies can be labeled, cross-linked or conjugated and fused with other proteins or polypeptide molecules to form complexes (such as cytotoxic substances, radioactive toxins and/or chemical molecules, etc.) for diagnosis and treatment.
The present disclosure provides multispecific antibodies that contain the indicated bispecific antibody binding to BCMA and CD3, fusion proteins, immunotoxins, medicines or detection reagents.
The above-mentioned immunotoxins comprise the bispecific antibodies connected to cytotoxic agents in various forms.
Various forms of connection refer to antibody being labeled, cross-linked in vitro or conjugated molecularly. The cytotoxic agents include chemical molecules, radioactive isotopes, polypeptides, toxins and other substances that can kill or induce cell death.
The above-mentioned fusion proteins comprise the above-mentioned bispecific antibody provided by the present disclosure and a complex consisting of other proteins or polypeptide molecules with certain functions.
Specifically, the fusion proteins can be recombinant fusion protein molecules obtained with recombinant expression vectors constructed by linking antibody genes with immunotoxin or cytokine genes through mammalian cells or other expression systems.
The above-mentioned medicines and detection reagents may also contain other active ingredients or excipients allowed in the fields of pharmacy and detection reagents (e.g., antibody components and pharmacologically acceptable delivery molecules or solutions. The component for treatment is sterile and can be freeze-dried at low temperature).
The bispecific antibody in the present disclosure can inhibit one or more biological activities induced by BCMA, which acts by blocking the binding of BCMA to its ligands, or by killing cells with high expression of BCMA, or by internalizing the complex formed after binding to BCMA to deplete BCMA on the cell surface. All interfering functions of BCMA antagonists are equally considered as objectives of the present disclosure.
The beneficial effects of the present disclosure are as follows.
The present disclosure uses genetic engineering and phage surface display library technology to screen the specific antibody against human B cell surface antigen BCMA from the library of single-chain antibody with natural fully human sequence. Multiple clones with significantly improved binding ability are obtained through affinity maturation engineering, with affinity to human BCMA between 0.26 nM to 0.31 nM. The apparent affinity measured by flow cytometry is 61 to 97 times higher than that of the parental antibody, confirming that the anti-human BCMA antibody and the optimized mutants in the present disclosure have good binding ability to BCMA.
On this basis, the present disclosure provides a bispecific antibody, which can simultaneously bind to BCMA and CD3, has specific targeting effect, and can efficiently stimulate directed immune responses. The bispecific antibody better retains the biological functions of anti-CD3 antibody and anti-BCMA antibody, which has excellent biological function of simultaneously binding to BCMA and CD3 as a bispecific antibody, forming a bridge between tumor cells and immune effector cells, effectively activating immune effector cells and directed immune responses, significantly enhancing the effects of immune cells in killing tumor cells, and reducing the ADCC effect to the largest extent, with high safety.
The bispecific antibody provided in the present disclosure has a good prospect for treatment, mainly manifested as specific binding to human BCMA and CD3. ELISA and flow cytometry show specific binding to BCMA on the surface of H929 and RPMI8226 cells and T cells, indicating good target specificity. The bispecific antibody has a high target cell-killing efficiency through the in vitro cell-killing efficiency test.
In addition, the bispecific antibody provided in the present disclosure has a completely symmetrical structure, which will not produce protein isomers with other structures when expressed by the host, thereby greatly reducing the difficulty of extraction and purification processes and having advantages of simple preparation and high yield, which shows broad application prospects in tumor immunotherapy.
The present disclosure provides bispecific antibody candidate for the development of anti-tumor antibody medicines targeting BCMA, CAR-T reagents, and the prevention and treatment of other diseases such as B cell-related inflammation and autoimmune disorders. The bispecific antibody in the present disclosure can simultaneously bind to immune cells and tumor cells, guide T-cell immune response, and specifically and effectively kill tumor cells, which can be developed as antibody drugs for multiple myeloma.
The following Examples are provided to illustrate the present disclosure but are not intended to limit the scope of the present disclosure.
Antibody library technology is to clone all antibody variable region genes of a certain animal (including humans) into plasmids or phages, which then infect Escherichia coli, leading to expression of antibody fragments on the surface of phage particles, or in the periplasm, and cytoplasm of Escherichia coli, then the target antigen is used to screen the clones carrying specific antibody genes from the antibody library, so as to obtain the corresponding specific antibody. Various antibodies needed in basic research and clinical development such as those against tumor-associated membrane protein antigens, autoantigens related to autoimmune diseases, and viral antigens of viral diseases have been screened from the antibody library. These results show the great potential of antibody library technology in basic research and antibody medicine development. In particular, fully human monoclonal antibodies obtained from the human antibody library overcomes the obstacle of getting human monoclonal antibody with mouse hybridoma technology: due to the high conservation of human antibody sequences, these antibodies have much lower immunogenicity to human immune system than those derived from animals, which are accordingly safer.
The non-immune human natural antibody library uses healthy people who are called donors as the source of antibody genes. The immunogenicity of antibodies obtained by using natural human antibody sequences is very low to human immune system. All RNAs were extracted from the donor's peripheral blood mononuclear cells (PBMC) with an RNA extraction kit (such as QIAGEN's RNeasy Mini Kit, catalog number 74104) by using molecular biology and DNA manipulation techniques, and then the same amount of RNAs were extracted from each RNA sample a pooled, to synthesize cDNA of light and heavy chain gene with a reverse transcription kit (such as cDNA synthesis kit of ThermoFisher Scientific, SuperScript™ IV Reverse Transcriptase). With these antibody cDNAs as templates, in the first round of amplification, all member genes of each subgroup were obtained by PCR amplification with subgroup-specific upstream primers and constant region primers. In the second round of PCR, the 3′-terminal primer set of the heavy chain variable region (VH) with bridging and the 5′-terminal primer set of the light chain variable region (VL) with bridging were paired with their respective primers at the other end (containing specific restriction endonuclease sites), allowing the obtained VH and VL to be linked together to form a single-chain antibody (scFv) gene. These scFv genes were cloned into lysogenic phagemid (such as M13 phagemid) vectors to construct an antibody library by using the specific endonuclease sites at the 5′-terminal and 3′-terminal of these single-chain antibody genes. The size of the library reached 5 billion colony forming units (cfu).
Biopanning refers to a process of screening antibodies with specific targets to obtain specific clones. To obtain human BCMA antigen-specific human antibodies, liquid phase screening was used for panning. The general steps of the liquid phase panning are as follows: firstly, the commercialized BCMA antigen (BCMA-Fc fusion protein, Acrobiosystems, catalog number BC7-H5254) was modified with biotin, allowing each molecule to cross-link 3 to 5 biotin molecules, and the free biotin was removed. An appropriate amount of biotin-labeled BCMA antigen was bound to streptavidin-coupled magnetic beads (such as Dynabeads™ M-280 Streptavidin, ThermoFisher Scientific, catalog number 11205D). A human antibody library containing 10 billion phage particles expressing different antibodies was thawed. Both magnetic beads and antibody library were blocked with 4% non-fat dry milk solution (4% MPBS), to eliminate non-specific action sites. Since there may be streptavidin on the magnetic beads that has not been bound by avidin, and the fact that BCMA antigen molecules are fusion proteins composed of human Fc, it is necessary to add a sufficient amount (50-100 times more than the amount of target protein used) of streptavidin and human antibody Fc to the thawed antibody library at the same time, to remove antibody clones that bind them. The blocked antibody library solution and magnetic beads (total volume <1 ml) were added to a 1.5 ml Eppendorf tube and incubated for 2 hours at room temperature with rotational mixing, to allow BCMA-specific phage antibodies to bind. After incubation, the Eppendorf tube was placed on the magnetic rack for 1 minute, to separate the magnetic beads and the solution. A pipette was applied to remove the solution as much as possible, then a new tip was used to replace the used tip, to add 1 ml of washing solution PBST (phosphate buffer solution (PBS) with Tween 20 at a final concentration of 0.05%) to the Eppendorf tube which was kept away from the magnetic rack and then placed on the rack to separate the magnetic beads and the solution after the pipette was applied to gently suspend the magnetic beads. The solution was removed and the process was repeated three times. Then the washing solution was changed to PBS and the magnetic beads were washed 3 times. After washing, most of non-specific and low-affinity phage antibodies were removed, while specific phage antibodies remained on the magnetic beads. Eluent (10 mM Glycine, pH2.0) was added to the magnetic beads, which was then resuspend left at room temperature for 10 minutes, placed on the magnetic rack to separate the magnetic beads and the solution. The solution was drawn into a clean Eppendorf tube, and 1/10 1 M Tris solution (pH8.0) was added to neutralize the solution, obtaining the eluate containing thousands of different phage antibodies that bind to the BCMA antigen, thus the first round of panning was completed.
To further harvest specific antibody clones with high affinity, more rounds of panning are required. To this end, the phage antibody solution eluted from the first round of panning was used to infect E. coli (such as TGI strain) in logarithmic phase that can be infected by M13 phage, to obtain an infection solution. A small amount of the solution was taken for serial 10-fold gradient dilutions (usually diluted to one millionth of the stock solution, and the last three gradients were taken for coating) to determine the titer of the output produced in the first round. The titer is also called the maximum diversity of the first round, usually, the output titer after the first round of panning was below 10 E6 cfu. The rest of the infection solution was coated on the bacterial culture plate containing the corresponding antibiotics and cultured overnight, to obtain colonies: the colony layer was scraped off and resuspended with culture medium. Sufficient amount of resuspension liquid containing the diversity of the first round output was taken and added into a shake flask containing sufficient liquid culture medium (2YT-CG, wherein carbenicillin and glucose were added into 2YT culture medium, at final concentration of 100 μg/ml and 2%, respectively). The resuspension liquid was diluted to below 0.1 OD600 and cultured until the logarithmic phase, when OD600 reached about 0.5. To make these antibodies obtained in the first round of panning reappear on the surface of phage particles, 10 ml of bacterial solution was taken and added with helper phage M13K07 to make the multiplicity of infection 20:1. The mixture was kept at 37° C. for 30 minutes (the stage was called phage rescue), then centrifuged, resuspended with 50 ml expression medium (2YT-AK, wherein carbenicillin and kanamycin were added into 2YT culture medium, at final concentrations of 100 μg/ml and 30 μg/ml, respectively), and cultured overnight at 30° C. and 200 rpm. On the following day, the culture supernatant was harvested by centrifugation, and was added with 1/5 volume of PEG8000/NaCl (PEG-8000 20%, NaCl 2.5 M), thoroughly mixed, and incubated on ice for 1 hour. The phage antibody particles were harvested by high-speed centrifugation (11,500×g) for 30 minutes. The precipitates were resuspended with 1 ml PBS solution and centrifuged at high speed to remove bacterial debris. The supernatant was the amplification solution after the first round of panning, wherein each contained antibody clone had been amplified more than ten thousand times. The amplification solution can be used for the second round of panning. The operation of the second round of panning was the same as that of the first round, except that the washing with PBST/PBS was increased to 6 times (6/6). In the third round, the washing frequency can be further increased to 10/10. Multiple rounds of panning usually effectively enrich specific clones, with significantly reduced diversity but high affinity, which is convenient for subsequent monoclonal screening.
To obtain specific monoclonal antibodies, a monoclonal phage enzyme-linked assay (Monophage ELISA) is required. For this purpose, single colonies that were well-separated in the second and/or third round of gradient dilutions were individually inoculated into 96-well culture plates containing 2YT-AG (93 colonies per plate, with three wells as negative control), and cultured overnight, which was the master plate. The bacterial solution in each well of the master plate was inoculated into a new culture plate to grow to the logarithmic phase, and the above-mentioned phage rescue was performed to allow expression of the antibody of each clone on the surface of the phage. The BCMA antigen (1 μg/ml) was coated on a regular 96-well enzyme-linked plate, while human Fc with the same concentration was coated on another enzyme-linked plate. The bacterial solution of each individually expressed monoclonal phage antibody was added to the corresponding wells of the BCMA plate and the Fc plate respectively, followed by appropriate second antibody and horseradish peroxidase (HRP)-conjugated third antibody. The substrate was colored, and the absorbance value (450 nM) was read. The judgment method of BCMA-positive clones was as follows: The clone was negative on the Fc plate (not more than 1.5 times the absorption value of the negative well on the plate where the clone was located), but positive on the BCMA plate (more than 3 times the absorption value of the negative well on the plate where the clone was located), with absorption value of the wells higher than that of the corresponding wells on the Fc plate. Analysis showed that the clones corresponding to 10 wells were only positive for BCMA antigen, and negative for Fc, and the clones were collectively called hit.
The hit bacterial solution was inoculated into 3 ml 2YT-CG from corresponding wells of the master plate and cultured overnight at 37° C. and 200 rpm. On the following day, the phagemid DNA was extracted, and the sequence of the single-chain antibody region containing each hit was determined with specific primers. DNA sequences of the coding regions were translated into amino acid sequences, and multiple sequence alignments (CLUSTALW, website link https://www.genome.jp/tools-bin/clustalw) were performed to determine the clone specificity. Analysis revealed that the ten hits belonged to two different clones in sequence, one of which was named B10, thus obtaining the variable region sequence of a fully human antibody against human BCMA antigen. The amino acid sequences of CDR1, CDR2 and CDR3 in the heavy chain variable region of B10 were represented by SEQ ID NO. 4-6 respectively, and those of CDR1, CDR2 and CDR3 in the light chain variable region of B10 were represented by SEQ ID NOs. 1-3 respectively: the amino acid sequence of the heavy chain variable region was represented by SEQ ID NO. 17, and that of the light chain variable region was represented by SEQ ID NO. 18. The amino acid sequence of the single-chain antibody composed of the light chain and heavy chain variable regions of B10 was represented by SEQ ID NO. 36.
To verify whether the obtained BCMA antibody clones bind to the purified BCMA antigen and BCMA expressed on the cell membrane surface, the gene of the B10 single-chain antibody was cloned into the eukaryotic expression vector pFH (prepared by EXCYTE LLC), to obtain the plasmid pFH-B10. In the vector, scFv gene was fused with the Fc gene of human IgG4 to express scFv-Fc protein, which can be affinity purified with protein-A or detected by (HRP or fluorescein) labeled anti-human Fc antibody.
After obtaining the scFv-Fc protein of B10, binding of the antibody to BCMA was tested by Octet RED, confirming the specific binding of B10 to BCMA.
Binding of B10 to BCMA-expressing cell line H929 (NCI-H929, ATCC® CRL-9068™) was detected by flow cytometry, indicating the specific binding of B10 to the BCMA antigen on the cell membrane surface (see
The affinity maturation in vitro is a rapid directed molecular evolution operation: mutations were introduced at the DNA level and screening was performed at the protein binding level. For B10 antibodies, in order to avoid introducing mutations that may enhance immunogenicity or structures that may significantly change the antibody, mutation sites were limited to the sequence of the CDR regions, and each position was subjected to saturation mutagenesis separately, with each mutation separately expressed and tested.
The operation process of affinity maturation was as follows: firstly, genes of light and heavy chain variable regions of B10 were respectively cloned into pUFL vectors (pUFL is a plasmid capable of expressing antibody Fab in E. coli, prepared by EXCYTE LLC), to obtain Fab form of B10. A full set of primers of single-site random mutations in CDR region was designed, and PCR mutations were performed on each CDR position with site mutation kit (QuikChange Lightning Multi Site-Directed Mutagenesis Kit, Agilent catalog number 210515), BL21 (DE3) competent bacteria were transformed respectively, and monoclonal colonies plates were prepared separately. These mutants were collectively referred to as single site mutation library. Meanwhile, the Fab vector of the B10 parental antibody was also transformed. >30 colonies were selected to prepare DNA for sequence analysis of mutation regions, to evaluate the mutation efficiency. When the mutation efficiency was appropriate (greater than 80%), 92 colonies were selected according to the mutation at each CDR position and added to a 96-well culture plate containing the corresponding culture medium (2YT containing 100 μg/ml Carbenicillin+0.1% glucose), with additional 3 parental colonies chosen, and 3 wells left as blank controls. The plates were placed at 37° C., and cultured at 300 rpm for 6 hours, added with IPTG to a final concentration of 1 mM, transferred to 30° C., and cultured at 300 rpm overnight. Fab fragments were expressed and secreted into the medium.
The conditions of the enzyme-linked reaction were optimized through preliminary experiments, to allow the A450 absorption value of the binding of B10 parental antibody Fab to the antigen BCMA-Fc 3 times higher than the background signal. The BCMA-Fc antigen was coated at 0.25 μg/ml on the enzyme-linked plate one day in advance, and the secreted and expressed mutant Fab was added to the wells on the following day, with HRP-conjugated anti-human Fab as the second antibody, the substrate was colored, and A450 was read, to obtain clones corresponding to the wells with signals 1.5 times higher than the maximum value of the parental wells, and these clones were called hits. All hits of the light and heavy chain variable regions were collected, inducted to express and tested by ELISA to verify their affinity higher than the parental antibody. Sequence analysis of mutant regions of the plasmid of the re-positive clone was performed, to obtain the mutation information of CDR amino acids. The process was called primary screening. All mutation information in the CDR region that was beneficial to the improvement of affinity was obtained: mutation sites that improved affinity in the heavy chain variable region CDR are shown in Table 1, and those in the light chain variable region CDR are shown in Table 2.
After screening and obtaining the information of all single mutations beneficial to the improvement of affinity in the entire CDR region from the primary library through ELISA, new multi-site mutation primers were redesigned to include major beneficial mutations, and mutation library was constructed again with the point mutation kit, which was called the combinatorial library. In order to screen all mutation combinations in the light and heavy chain variable regions as much as possible, combinatorial libraries containing multiple mutations in the light and heavy chain variable regions were respectively constructed and screened. The top 5 clones with the highest affinity in the combinatorial library of the light chain variable region were obtained, whose amino acid sequences of the light chain variable region contained five mutant sequences that are represented by SEQ ID NOs. 22-26, and CDR1 in the light chain variable region had the amino acid sequence represented by any one of SEQ ID NOs. 1 and 12, CDR2 in the light chain variable region had the amino acid sequence represented by SEQ ID NO. 2, and CDR3 in the light chain variable region had the amino acid sequence represented by any one of SEQ ID NOs. 3, 13, 14, 15, or 16.
The amino acid sequences of the top three clones with the highest affinity in the combinatorial library of the heavy chain variable region were obtained with ELISA screening of the combinatorial library of the heavy chain variable region, whose amino acid sequences of the heavy chain variable region contained three mutant sequences that were represented by the sequences SEQ ID NOs. 19-21 respectively, the CDR1 in the heavy chain variable region had the amino acid sequence represented by any one of SEQ ID NOs. 7, 8, and 9, and CDR2 in the heavy chain variable region had the amino acid sequence represented by SEQ ID NO. 10, and CDR3 in the heavy chain variable region had the amino acid sequence represented by SEQ ID NO. 11.
DNAs of the top 5 clones with the highest affinity of the light and heavy chain variable regions were mixed respectively and assembled into the final single-chain antibody combinatorial library by enzyme digestion and linking recombination. The library was screened to select the top 10 with the highest signals, and their DNA sequences were analyzed, obtaining four different single-chain antibody clones, with the corresponding amino acid sequences represented by SEQ ID NOs. 37-40.
To check whether the affinity of obtained mutants had been improved, the B10 parental antibody and fusion proteins of the finally obtained 4 mutants B10.3, B10.4, B10.5, B10.9 (amino acid sequences represented by SEQ ID NO. 37-40) with Fc (i.e., scFv-Fc) were respectively expressed and purified. The purified products were detected by flow cytometry on two BCMA-positive cell lines H929 and RPMI8226 with different expression levels. The percentages of binding to H929 cells were: 75.31% for B10 parental antibody, increasing to 97.96% for mutant B10.3, improving to 97.89% for mutant B10.4, increasing to 97.17% for mutant B10.5, and improving to 91.61% for mutant B10.9. The percentages of binding to RPMI8226 cells were: 5.54% for B10 parental antibody, increasing to 54.89% for mutant B10.3, improving to 57.51% for mutant B10.4, increasing to 58.18% for mutant B10.5, and improving to 42.89% for mutant B10.9. Results showed that all four mutants had significantly improved binding to BCMA compared to B10 (see Figure. 3).
In order to understand the affinity of B10 and its derived antibodies, the B10 parental antibody and the 4 mutants finally obtained were compared with the BCMA antibody J6 which had known affinity (i.e., clone J6 M0, whose affinity was about 0.2 nM according to the record of patent [WO2012163805A1]) by ELISA respectively. After antibodies with different concentrations were coated on enzyme-linked plates, the biotinylated BCMA antigen, streptavidin-conjugated horseradish peroxidase (HRP) and chromogenic substrate were added step by step. The obtained signal-antibody gradient curve is shown in Graph A of
Results showed that in terms of intrinsic affinity, the S-curve was not obtained since the binding of B10 parental antibody to human BCMA was significantly weaker than that of J6, and its affinity value could not be generated. In addition, the affinity of B10.3 was about 0.26 nM, B10.4 about 0.31 nM, B10.5 0.27 nM, and B10.9 about 0.29 nM.
In order to understand the apparent affinity of these antibodies to BCMA on the cell surface, the binding of B10 parental antibody and the finally obtained 4 mutants to BCMA-expressing cell line RPMI8226 were compared with J6 by flow cytometry (see Graph B of
In the present Example, the bispecific antibody was designed with the tumor cell surface antigen BCMA and the immune cell surface antigen CD3 as targets.
Combining protein structure design software and many manual experimental screening, the present disclosure screened and determined the bispecific antibodies with symmetrical structures including single-chain antibody units and monoclonal antibody units from a variety of bispecific antibodies binding to BCMA and CD3. The technical platform is referred to as FIST (fusion of IgG and scFv technology) in the present disclosure. For example, anti-BCMA monoclonal antibody units can be used as IgG antibodies, including 2 complete light chain-heavy chain pairs (i.e., containing complete Fab and Fc domains, and the heavy and light chains were connected by disulfide bonds); and anti-CD3 can be used as single-chain antibody units, including 2 single-chain antibodies (ScFvs). A linker peptide (SEQ ID NO. 30) was used to connect the single-chain antibody and the monoclonal antibody. The single-chain antibody and the monoclonal antibody related to the following method to obtain bispecific antibodies with a symmetrical structure (
The C-terminus of the single-chain antibody was linked to the N-terminus of the heavy chain variable region (VH) of the monoclonal antibody, to obtain nFIST (Graph A of
In the present Example, the variable region sequence of the anti-CD3 single-chain antibody UCHT1 was derived from literature (Beverley, P. C. & Callard, R. E. Distinctive functional characteristics of human “T” lymphocytes defined by E rosetting or a monoclonal anti-T cell antibody (1981) Eur. J. Immunol. 11, 329-334) and was humanized (Shalaby et. al., Development of humanized bispecific antibodies reactive with cytotoxic lymphocytes and tumor cells overexpressing the HER2 protooncogene. (1992) J Exp Med. January 1; 175 (1): 217-25). The constructed single-chain antibody was named K3 (SEQ ID NO. 31), containing two cysteines (Cys) inserted in the heavy chain and light chain variable regions respectively, which forms a pair of interchain disulfide bond after folding, with the amino acid sequence of the heavy chain variable region represented by SEQ ID NO. 28, and that of the light chain variable region represented by SEQ ID NO. 29.
The bispecific antibody coding genes in the form of nFIST or cFIST were designed according to Example 4, and cloned into the expression vector pG4HK, to obtain two bispecific antibodies K3B10.3 and B10.3K3. Their amino acid sequences of the heavy chains were represented by SEQ ID NO. 34 and 35, respectively (fusion peptide formed by connecting the heavy chain of the monoclonal antibody with the single-chain antibody through a linker peptide): with the amino acid sequence of the light chain represented by SEQ ID NO. 33. The expression plasmids corresponding to K3B10.3 and B10.3K3 were stably transfected into CHO-K1, respectively, to prepare bispecific antibody proteins.
More anti-BCMA×CD3 bispecific antibodies were constructed using the form of cFIST. The heavy chain variable region of the BCMA-binding IgG unit was derived from SEQ ID NOs. 17-21; while the light chain variable region of the BCMA-binding IgG unit was derived from SEQ ID NOs. 18, 22, or 25-27. Plasmids of these antibodies were transfected into 293F cells, to prepare by transient expression and one-step purification of protein-A, to obtain B10K3 (the amino acid sequence of the heavy chain represented by SEQ ID NO. 41; that of the light chain represented by SEQ ID NO. 42), B10.3K3, B10.4K3 (the amino acid sequence of the heavy chain represented by SEQ ID NO. 43; that of the light chain is represented by SEQ ID NO. 44), B10.5K3 (the amino acid sequence of the heavy chain represented by SEQ ID NO. 45; that of the light chain represented by SEQ ID NO. 46), and B10.9K3 (the amino acid sequence of the heavy chain represented by SEQ ID NO. 47; that of the light chain represented by SEQ ID NO. 48). SDS-PAGE purity analysis showed monomer purity above 90%. Tests of their binding to BCMA antigen is shown in Graph A of
The above-mentioned bispecific antibodies were processed with the following steps, to obtain high-purity bispecific antibodies. (1) Material liquid pretreatment: The supernatant of fermentation culture was centrifuged at 2,000 rpm for 10 min, and then filtered with a 0.22 μM filter membrane. (2) Affinity chromatography: MabselectSuRe affinity chromatography column (purchased from GE Company, Cat. No. 18-5438-02) was utilized to capture antibodies in the pretreated fermentation broth, which were passed through the affinity chromatography column after fully balancing the column with the equilibration buffer (10 mM PB, 0.1 M NaCl, pH 7.0), and eluted with elution buffer (0.1 M citric acid, pH 3.0). (3) Cation exchange chromatography: Samples prepared by affinity chromatography was further purified by molecular sieve exchange chromatography, with buffer solution (50 mM PBS, 0.2 Man Na2SO4, pH 6.7).
K3B10.3 and B10.3K3 were tested by SDS-PAGE and HPLC-SEC, and results of SDS-PAGE are shown in
PBMC cells of healthy human donors were collected, T cells were purified, resuspended in PBS at 1×106 cells/reaction tube, rinsed once with 1 ml binding buffer (PBS containing 0.5% w/v BSA+2 mM EDTA), centrifuged (350×g, 4° C., 5 min), and then resuspended in 200 μl of binding buffer. The bispecific antibodies K3B10.3 and B10.3K3 were added to 5 μg/ml respectively and incubated on ice for 45 minutes. Cells were rinsed once as above and resuspended in 100 μl of binding buffer. 5 μl fluorescence labeled secondary antibody (PE anti-human IgG Fc Antibody, Biolegend, 409304) was added to the sample tube, the isotype control (PE Mouse IgG2a, K Isotype Ctrl (FC) Antibody, Biolegend, 400213) was added to the isotype control tube, both were incubated on ice in the dark for 15 min. Cells were rinsed once as above. Cells were resuspended in 200 μl of PBS, and then tested on the machine (Beckman).
Results of binding of K3B10.3 and B10.3K3 to human T cells tested by flow cytometry are shown in
To test the binding of anti-BCMA unit to BCMA, T cells in the experiment were replaced with RPMI8226 (ATCC® CCL-155) to perform a similar FACS experiment. Results of the control are shown in Graph D of
In the present Example, H929-luc and RPMI8226-Luc were used as targeted cells, and PBMC was utilized as immune effector cell to test the targeted cell-killing effect mediated by bispecific antibodies B10K3, B10.3K3, B10.4K3, B10.5K3, B10.9K3 and K3B10.3, with specific experimental steps as follows:
With Steady-GLO (Promega company) as the substrate, the buffer in the kit was thawed and added to the substrate powder, mixed well, and divided into 5 ml or 10 ml each tube to complete the reconstruction of Steady-GLO substrate. After the co-cultured cells were evenly mixed by pipetting up and down, 100 μl of the cells were transferred to an opaque white plate, then 100 μl of the reconstructed Steady-GLO substrate was added, and mixed by patting gently, left for 5 minutes before reading the plate. The instrument used for detection was Synergy HT.
Target cell killing ratio (percentage)=100×(the number of wells with only targeted cells−the number of wells tested)/the number of wells with only targeted cells
The antibody concentration corresponding to the target cell killing ratio in all tested wells was converted to log 10, which was used as the horizontal axis with the cell killing ratio as the vertical axis to plot a curve. The killing concentration-gradient curves of the bispecific antibodies B10K3, B10.3K3, B10.4K3, B10.5K3, B10.9K3 in the form of cFIST against the sensitive cell line RPMI8226 are shown in Graph B of
Results of K3B10.3 and B10.3K3 are shown in
In Tables 5, 6 and 7 above, the two columns of values for each bispecific antibody were the results of two duplicate experiments.
Although the present disclosure has been specified with general descriptions and specific embodiments above, it is apparent to those skilled in the art that some modifications or improvements can be made based on the present disclosure. Therefore, any modifications or improvements without departing from the spirit of the present disclosure are within the scope of the present disclosure.
The present disclosure relates to a bispecific antibody binding to BCMA and CD3 and a preparation method therefore and the use thereof. The present disclosure provides a bispecific antibody, which can simultaneously bind to BCMA and CD3, has a specific targeting effect and can efficiently stimulate directed immune responses. The bispecific antibody retains the biological functions of an anti-CD3 antibody and an anti-BCMA antibody, enables the bispecific antibody molecule to achieve excellent biological functions of simultaneously binding to BCMA and CD3 in a targeted manner, forms a bridge between tumor cells and immune effector cells, effectively activates the immune effect or cells and the directed immune response, significantly enhances the effects of immune cells in killing tumor cells, has a high targeted cell killing efficiency, and reduces the ADCC effect to the largest extent, has high safety, and has good application prospects in tumor immunotherapy.
| Number | Date | Country | Kind |
|---|---|---|---|
| 202110704255.6 | Jun 2021 | CN | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CN2022/090931 | 5/5/2022 | WO |
