Patent Application
20240317859
The present invention relates to a PD-1 agonist and a pharmaceutical composition comprising a PD-1 agonist for treating or preventing inflammatory diseases.
The immune response has an aspect that its inappropriate regulation could lead to diseases. An insufficient immune response to pathogens such as bacteria or viruses that have invaded the organism results in infectious diseases. Conversely, if adverse immune response to self-tissues occurs, autoimmune diseases develop. To prevent from falling into such pathological conditions and to utilize the system effectively, the immune system is provided with both mechanisms activating the function of immune cells to promote immune response and mechanisms suppressing immune cells to reduce immune response. Disruption of these endogenous regulatory mechanisms is considered to contribute to diseases resulting from insufficient or excessive immune response.
The best example demonstrating the importance of endogenous immunoregulation can be found in cancer immunotherapy that has rapidly become a reality in recent years. The molecules such as CTLA-4 and PD-1 appearing in this novel therapy, which is clearly distinct from conventional “immunotherapy”, are a type of endogenous immunosuppressive mechanisms also called immune checkpoints, and have a particularly remarkable impact among the same type of immunosuppressive mechanisms. The pathology of cancer can be regarded as a condition where cancer cells that should have been eliminated have proliferated because of insufficient immune response thereto. The cause of this insufficient immune response is related to the fact that the inside of cancer tissues is filled with various types of immunosuppressive mechanisms. This means that cancer tissues evade attacks from immune cells by actively using the immunosuppressive mechanisms intrinsic to the organism. By blocking CTLA-4 or PD-1, representative immunosuppressive mechanisms, the inhibition of anti-tumor immunity is reversed and the resulting immune activation brings about actual therapeutic effects. The fact shows that the above-mentioned endogenous immunosuppressive mechanisms have highly significant effects and can be important targets for the treatment of diseases by correcting the imbalance in immune response.
The critical importance of endogenous immunosuppressive mechanisms is also shown by the fact that the absence of any one of these mechanisms can cause much exaggerated inflammatory responses. For example, PD-1-deficient mice have been shown to spontaneously develop various inflammatory (Non-Patent Document No. 1: Okazaki et al. Nat. Immunol., 2013). Inflammatory adverse response resulting from excessive immune response is observed at a certain proportion of cancer patients undergoing immunotherapy with PD-1 inhibitors (Non-Patent Document No. 2: Young et al. Cancer Immunol. Res., 2018). These facts indicate that the immunosuppressive mechanisms that are actively utilized in cancer tissues are essentially physiological feedback mechanisms to protect healthy tissues in the body from excessive inflammatory responses. Importantly, the clinical efficacy of pharmaceutical drugs such as anti-PD-1 antibody in cancer therapy suggests that it is possible to modulate the intensity of immune response by regulating the function of PD-1.
As shown in the above-described background, drugs that inhibit immunosuppressive mechanisms were used in cancer therapy where the enhancement of immune response was desirable. However, to suppress overwhelming immune response in inflammatory diseases, the treatment demands an opposite approach that actively stimulates the immunosuppressive mechanisms. PD-1 is expressed on immune cells such as activated T cells and interact with ligand molecules such as PD-L1 and PD-L2 on the surface of the other cell. Upon antigen recognition, PD-1 stimulation with these ligand molecules interferes with the activating signal of the immune cells. Since PD-1 itself can initiate the immunosuppressive signaling, artificial stimulation of PD-1 might lead to the treatment of inflammatory diseases. While anti-PD-1 antibodies used in cancer therapy (eventually) enhance immune function by blocking PD-1 interaction with its ligand molecule, PD-1 agonists capable of inducing the function of PD-1 by binding to PD-1 are highly promising as drugs for positive immunosuppression.
The present invention aims at searching for conditions necessary for anti-human PD-1 antibody to have agonistic activity, establishing agonist antibody optimized based on such necessary conditions, and applying the agonist antibody as a therapeutic agent for human inflammatory diseases.
There are a number of inflammatory diseases caused by excessive immune response, and a novel and more effective treatment is demanded in many cases. Targeting an immune regulatory molecule PD-1, the present inventors have found a large number of anti-human PD-1 antibodies, which induce the immunosuppressive activity of PD-1 and are potentially capable of preventing or treating inflammatory diseases in human with the agonistic activity. The present inventors have examined binding domains of these antibodies bind to human PD-1 and found that the biological activity of these antibodies varies depending on the binding domain. Furthermore, the present inventors have found that binding to Fc receptors is necessary for anti-PD-1 antibodies to exert their agonistic activity and that an enhanced affinity to Fc receptors is necessary for immunosuppression in humans. In addition, the present inventors have found that the addition of ADCC-inducing capability to anti-PD-1 agonist antibodies enables depletion of PD-1-expressing activated immune cells, further improving desirable immunosuppressive activity.
The present invention has been achieved based on these findings, and a summary thereof is described as below.
(1) An agonist antibody to human PD-1 or a functional fragment thereof, wherein the antibody or a functional fragment thereof binds to domain #7 as shown in SEQ ID NO: 9, of human PD-1.
(2) The antibody or a functional fragment thereof of (1), wherein the antibody or a functional fragment thereof has any one of (A) to (D) below:
The present invention makes it possible to treat or prevent inflammatory diseases.
The present specification encompasses the contents disclosed in the specification and/or the drawings of Japanese Patent Application Nos. 2021-81913 and 2021-86534 based on which the present patent application claims priority.
(B) As a result of evaluation, about 30 clones having an immunosuppressive activity were obtained, including those with high activity and those with low activity. Among these anti-human PD-1 antibodies, only the clones having an immunosuppressive activity are shown in the graph.
FIG. 10CDEF Induction of acute GVHD in mouse and anti-inflammatory action of anti-human PD-1 agonist antibody. Splenocytes derived from C57BL/6 human PD-1 knock-in mice (donor) were transferred into BDF1 mice (recipient) to thereby induce acute GVHD. When HM266, HM242, HM297 and HM268 were respectively administered to this inflammation model in a preventive manner, donor cell ratios (C-D) and donor-derived CD8+ cell ratios (E-F) in the spleen and the liver 2 weeks after induction were significantly suppressed. These results suggested the applicability of anti-human PD-1 agonist antibody to a treatment of inflammatory diseases.
FIG. 11ABCD Induction of chronic GVHD/lupus-like symptoms in mouse and anti-inflammatory action of anti-human PD-1 agonist antibody. Splenocytes derived from C57BL/6 human PD-1 knock-in mice (donor) were freed of CD8+ cells and thereafter transferred into BDF1 mice (recipient) to thereby induce chronic GVHD/lupus-like symptoms. After the cell transfer, donor-derived Tfh cells (A), recipient-derived germinal center B cells (B), class switched B cells (C) and plasma cells (D) in the recipient's spleen were confirmed to increase remarkably. Prophylactic administration of HM266 to this inflammation model suppressed these increases. Mark “-” represents recipient mice into which no donor cells were transferred.
FIG. 27BCD Induction of acute GVHD in mouse and anti-inflammatory action of anti-human PD-1 agonist antibody. (B-D) AST (B), ALT (C) and IFN-γ (D) in mouse plasma 2 weeks after induction. Induction of GVHD increased the plasma levels of ALT, AST, and IFN-γ, but preventive administration of HM266, HM242, HM297 and HM268 suppressed these increases remarkably.
FIG. 28ABCD Induction of T cell-dependent antibody response in mouse and antibody production suppressive action of anti-human PD-1 agonist antibody. T cell-dependent antibody response was induced in C57BL/6 human PD-1 knock-in mice by intraperitoneal administration of 4-hydroxy-3-nitrophenylacetyl ovalbumin (NP-OVA) antigen together with alum adjuvant. Administration of HM266 suppressed the ratios of Tfh cells (A), germinal center B cells (B), class switched B cells (C) and plasma cells (D) in the mouse spleen 10 days after antigen administration.
FIG. 29ABCD Induction of chronic GVHD/lupus-like symptoms in mouse and anti-inflammatory action of anti-human PD-1 agonist antibody. Splenocytes derived from C57BL/6 human PD-1 knock-in mice (donor) were freed of CD8+ cells and then transferred into BDF1 mice (recipient) to thereby induce chronic GVHD/lupus-like symptoms. After cell transfer, donor-derived Tfh cells (A), recipient-derived germinal center B cells (B), class switched B cells (C) and plasma cells (D) were confirmed to increase remarkably in the recipient's spleen. When HM242 or CTLA-4-Ig was preventively administered to this inflammation model, these increases were suppressed. Mark “-” represents recipient mice into which no donor cells were transferred.
Hereinbelow, embodiments of the present invention will be described in detail.
The present invention provides an agonist antibody to human PD-1 (anti-human PD-1 agonist antibody) or a functional fragment thereof, wherein the antibody or a functional fragment thereof
As used herein, the “anti-human PD-1 agonist antibody” may be antibodies which bind to human PD-1 and trigger an intracellular signal transduction system similar to that triggered by human PD-L1 in the human body. Such antibodies may be selected in a co-culture system of T cells rendered to express human PD-1 and antigen-presenting cells rendered to express human FcγRIIB, using as an indicator the activity to suppress IL-2 which is produced from T cells in an antigen stimulation dependent manner (see Example described later).
The antibody of the present invention or a functional fragment thereof may bind to only domain #7 of human PD-1 (SEQ ID NO: 9; domain spanning from amino acid No. 38 to No. 48 in the amino acid sequence as shown in SEQ ID NO: 1) or may also bind to other domains of human PD-1 such as domain #6 (SEQ ID NO: 8; domain spanning from amino acid No. 109 to No. 120 in the amino acid sequence as shown in SEQ ID NO: 1) or domain #1 (SEQ ID NO: 3; domain spanning from amino acid No. 129 to No. 139 in the amino acid sequence as shown in SEQ ID NO: 1). Preferably, the antibody or a functional fragment thereof binds to only domain #7 of human PD-1. When the binding capacity of an anti-human PD-1 antibody is decreased by replacing domain #7 of human PD-1 as shown in SEQ ID NO: 1 with the amino acid sequence of the corresponding domain in the amino acid sequence of mouse PD-1 (SEQ ID NO: 2), the anti-human PD-1 antibody can be defined as binding to domain #7 of human PD-1 as shown in SEQ ID NO: 1. Here, the binding to domain #7 is judged independently from the binding to other domains of human PD-1. The binding to domains of human PD-1 other than #7 is defined in the same manner. The amino acid sequence of human PD-1 is registered at the NCBI database under accession number: NP_005009.2, and the amino acid sequence of mouse PD-1 at the NCBI database under accession number: NP_032824.1. Preferably, the antibody of the present invention or a functional fragment thereof also binds to substituted mouse PD-1 (Mouse PD-1 (hu38-48)) in which the amino acid sequence of domain #7 of mouse PD-1 is replaced with the amino acid sequence of domain #7 of human PD-1 and/or mutant human PD-1 (R143A point mutant) in which arginine at position 143 of human PD-1 is mutated to alanine.
The binding capacity (binding affinity) of the antibody of the present invention or a functional fragment thereof to human PD-1 may be 10−7 M or less, preferably 10−8 M or less, in terms of equilibrium dissociation constant (KD). Equilibrium dissociation constant is measured by the surface plasmon resonance (SPR) method. The binding capacity of the antibody of the present invention or a functional fragment thereof to human PD-1 can also be measured by flow cytometry using the concentration dependency of binding to PD-1 expressing cells.
Examples of the sequences of the heavy chain CDR1 to CDR3 and the light chain CDR1 to CDR3 of the antibodies of the present invention are shown in Table I below. CDRs may be identified by algorithms such as Kabat (J Exp Med. 1970; 132: 211-50), Chothia (J Mol Biol. 1987; 196: 901-17), IMGT (Dev Comp Immunol. 2003; 27: 55-77) and Paratome (PLOS Comput Biol. 2012; 8: e1002388), preferably by Kabat's method. In Table I, HM242, HM266, HM268 and HM297 are mouse anti-human PD-1 antibodies exhibiting PD-1 agonist activity that were prepared by immunizing mice with human PD-1. The antibody of the present invention may advantageously have a heavy chain variable region comprising heavy chain CDR1, CDR2 and CDR3 and a light chain variable region comprising light chain CDR1, CDR2 and CDR3, as shown in Table I below. In the antibody of the present invention, the heavy chain CDR1, CDR2 and CDR3 and the light chain CDR1, CDR2 and CDR3 may comprise sequences having at least 85%, 85-90%, 90-95%, 95-97% or 97% or more identity with the sequences as shown in Table I, in terms of CDR as a whole. Identity between two sequences can be determined using BLAST.
The variable region (Fv) of the antibody of the present invention may be the Fv region of an antibody derived from an animal other than human (for example, mouse, rabbit, rat, hamster, guinea pig, goat, sheep, donkey, llama, camel, chicken, ostrich, shark, etc.) or it may be a humanized Fv region of the heavy chain and/or the light chain of an antibody derived from a non-human animal. Humanization may be performed, for example, by transplanting the CDRs of VH and VL of non-human animal-derived antibodies into the frameworks of VH and VL of human antibodies (Nature, 332, 323-327, 1988). Humanized antibody may be one in which CDR sequences are retained. Alternatively, humanized antibody may also be one in which antigen binding is improved by, for example, identifying the amino acid residues directly involved in binding to antigen, the amino acid residues interacting with CDRs and the amino acid residues involved in retaining the three-dimensional structure of CDRs, and replacing these residues with amino acid residues of non-humanized antibodies (MABS, 8(7), 1302-1318, 2016).
Table II below shows exemplary sequences of the heavy chain variable region (VH region; position 1 to 117; according to EU numbering; hereinafter, the same shall apply) of mouse-derived antibodies exhibiting PD-1 agonist activity, as well as examples of their humanized VH region sequences.
Table III below shows exemplary sequences of the light chain variable region (VL region; position 1 to 107) of mouse-derived antibodies exhibiting PD-1 agonist activity, as well as examples of their humanized VH region sequences.
The antibody of the present invention may be one having the heavy chain variable region and/or the light chain variable region as shown in Tables II and/or Table III above. In the antibody of the present invention, the heavy chain variable region and the light chain variable region may individually comprise a sequence having at least 90%, 90-95%, 95-99% or 99% or more identity with the sequences as shown in Tables II and III above.
The antibody of the present invention may be one which has an improved affinity to human Fc receptors. The Fc receptor is preferably Fcγ receptor (FcγR), more preferably FcγRII, and still more preferably FcγRIIB. The binding affinity of the antibody of the present invention to human Fc receptors may be evaluated by a flow cytometer-based binding test for human Fc receptor-expressing cell line, or an human Fc receptor binding affinity test using surface plasmon resonance technology with Biacore 8K (Cytiva). The improvement in the affinity of the antibody of the present invention to human Fc receptors may be evaluated by either of the methods mentioned above. Preferably, the improvement is evaluated by flow cytometer, and more preferably, the improvement is evaluated by both methods. When measured by the human FcγRIIB binding affinity test using surface plasmon resonance technology, the affinity of the antibody of the present invention to human FcγRIIB can be expressed as an equilibrium dissociation constant (KD) ratio to an antibody having the Fc region of human IgG1-K322A (reference antibody). The antibody of the present invention has 1.5-fold or more affinity than the reference antibody, preferably 2-fold or more affinity than the reference antibody, and more preferably 2.5-fold or more affinity than the reference antibody (see Example described later). When measured by the flow cytometer-based binding test for human Fc receptor-expressing cell line, the affinity of the antibody of the present invention to human FcγRIIB can be expressed as a GMFI ratio to an antibody having the Fc region of human IgG1-K322A (reference antibody). Under measuring conditions where a GMFI value without antibody addition is about 40 and a GMFI value with addition of the reference antibody is 300-1500, the antibody of the present invention shows 2-fold or more GMFI, preferably 5-fold or more GMFI, and more preferably 20-fold or more GMFI as compared with the case where the reference antibody having the same Fv region is added.
The antibody of the present invention may also be one which has an improved affinity to human FcγRIIIA. When measured by the Fc receptor binding affinity test using surface plasmon resonance technology; the affinity of the antibody of the present invention to human FcγRIIIA(V158) can be expressed as an equilibrium dissociation constant (KD) ratio to an antibody having the Fc region of human IgG1-K322A (reference antibody). The antibody of the present invention has 1.5-fold or more affinity than the reference antibody, preferably 2-fold or more affinity than the reference antibody, more preferably 2.5-fold or more affinity than the reference antibody, and still more preferably 4-fold or more affinity than the reference antibody (see Example described later). When measured by the flow cytometer-based binding test for human Fc receptor-expressing cell line, the affinity of the antibody of the present invention to human FcγRIIIA(V158) can be expressed as a GMFI ratio to an antibody having the Fc region of human IgG1-K322A (reference antibody). Under measuring conditions where a GMFI value without antibody addition is about 10 and a GMFI value with addition of the reference antibody is 6000-15000, the antibody of the present invention shows 1.5-fold or more GMFI, preferably 2-fold or more GMFI, more preferably 4-fold or more GMFI, and still more preferably 5.0-fold or more GMFI as compared with the case where the reference antibody having the same Fv region is added.
The antibody of the present invention may be one having the Fc region of a human antibody (for example, IgG1, IgG4, etc.) and, preferably; the Fc region (position 216 to 447) of the human antibody is modified so that the affinity to human Fc receptors (at least one of human FcγRIIA, human FcγRIIB, human FcγRIIIA) is improved. By modifying the Fc region of an antibody, its affinity to human Fc receptors can be improved. Alternatively, the affinity of an antibody to human Fc receptors can also be improved by defucosylation treatment.
One example of the amino acid sequence of the Fc region of human IgG1 is shown in SEQ ID NO: 47, and one example of the amino acid sequence of the Fc region of human IgG4 is shown in SEQ ID NO: 64. As a means to improve the affinity of the antibody of the present invention to human Fc receptors, a mutation(s) may be introduced, for example, to at least one position, preferably a combination of positions, selected from the group of amino acid positions consisting of 233, 234, 236, 237, 238, 239, 267, 268, 271, 296, 323, 326, 328, 330 and 332 (according to EU numbering; hereinafter, the same shall apply) in the amino acid sequence of the Fc region of human IgG1 as shown in SEQ ID NO: 47. Further, such mutation(s) may be combined with defucosylation. When mutations are introduced into a combination of positions, for example, it may be selected from among 236/268, 239/268, 239/268/328/332, 239/267/268, 239/267/268/328, 239/267/268/332, 239/268/328, 239/268/332, 239/267/268/328/332, 233/237/238/268/271/330 and 267/328, preferably from among 239/268, 239/268/328/332, 239/267/268, 239/267/268/328332, 239/268/332, 233/237/238/268/271/330 and 267/328, and more preferably from among 239/268/328/332, 233/237/238/268/271/330 and 267/328. G at position 236 may be mutated to, for example, D, E, N, Q, F, H, I, K, L, M, P, R, S, T, V, W, Y or A, preferably to D, E, N or Q, and more preferably to D. H at position 258 may be mutated to, for example, D, E, N, Q, A or G, preferably to D, E, N or Q, and more preferably to D. S at position 239 may be mutated to, for example, D, E, N, Q, F, T, H, Y, G or L, preferably to D, E, N or Q, and more preferably to D. L at position 328 may be mutated to, for example, Y, E, F, H, I, Q, W, D, T, S, M, A, V. N or H, preferably to Y, F, D, E, N or Q, and more preferably to Y. I at position 332 may be mutated to, for example, E, D, F, L, R, S, T, D, N, Q, H, Y, A or M, preferably to E, T or M, and more preferably to E. E at position 233 may be mutated to, for example, D. G at position 237 may be mutated to, for example, W, F, A, D, E, L, M or Y, preferably to D or E, and more preferably to D. P at position 238 may be mutated to, for example, D. P at position 271 may be mutated to, for example, G. A at position 330 may be mutated to, for example, K, R or M, preferably to R. S at position 267 may be mutated to, for example, E, V, Q, A or L, preferably to E or G, and more preferably to G. K at position 326 may be mutated to, for example, L, Q, N, M, D, S, T or A. L at position 234 may be mutated to, for example, D, E, N, Q, T, H, I, V, F, W or Y. V at position 323 may be mutated to, for example, I, L or M. Y at position 296 mat be mutated to, for example, D.
When mutation(s) is/are to be introduced into the Fc region of human IgG1, they may be combined with other mutations depending on the purpose, as exemplified by K322A (a mutation known to suppress complement-dependent cytotoxicity (CDC) activity by decreasing the binding of complement C1q) or E293A (a mutation known to suppress ADCC activity by decreasing the binding to FcγRIIIA).
The combination of mutations to be introduced into the Fc region of human IgG1 may be any one of the following combinations: G236D/H268D, S239D/H268D, S239D/H268D/L328Y/I332E, G236D/H268D/K322A, S239D/H268D/K322A, S239D/S267G/H268D/K322A, G236D/H268D/E293A/K322A, S239D/H268D/K322A/L328Y/I332E, S239D/H268D/E293A/K322A, S239D/S267G/H268D/K322A/L328Y, S239D/S267G/H268D/K322A/I332E, S239D/H268D/K322A/L328Y, S239D/H268D/K322A/I332E, S239D/S267G/H268D/K322A/L328Y/I332E, E233D/G237D/P238D/H268D/P271G/A330R, or S267E/L328F. Preferably, the combination may be any one of the following combinations: S239D/H268D/K322A, S239D/S267G/H268D/K322A, G236D/H268D/E293A/K322A, S239D/H268D/K322A/L328Y/I332E, S239D/H268D/E293A/K322A, S239D/S267G/H268D/K322A/I332E, S239D/H268D/K322A/I332E, E233D/G237D/P238D/H268D/P271G/A330R, or S267E/L328F.
When mutations are to be combined with defucosylation in the Fc region of human IgG, the combination of mutations in the Fc region should be such that defucosylation improves affinity for human Fc receptor, especially human FcγRIIIA, 1.5-fold or more affinity than the fucosylated IgG having same mutations, preferably 2-fold or more affinity than the fucosylated IgG having same mutations, more preferably 4-fold or more affinity than the fucosylated IgG having same mutations. Preferred mutation and defucosylation combinations include, but are not limited to, S239D/S267G/H268D/K322A and defucosylation combinations.
As an exemplary means to improve the affinity to human Fc receptors, a mutation may be introduced at any one or more of the following positions: 236, 239, 268, 328, or 332 in the amino acid sequence of the Fc region of human IgG4 as shown in SEQ ID NO: 64, for example. Preferably, a mutation may be introduced at a combination of those positions. Further, such mutations may be combined with defucosylation.
When a mutation is to be introduced into the Fc region of human IgG4, it may be combined with other mutations such as S228P that is known to be effective for improving antibody stability.
When mutations are to be combined, the combination may be selected from among S228P/G236D/Q268D, S228P/S239D/Q268D and S228P/S239D/Q268D/L328Y/I332E, more preferably from between S228P/S239D/Q268D and S228P/S239D/Q268D/L328Y/I332E.
Table IV shows exemplary sequences of Fc regions of human IgG1 (WT and K322A) and IgG4 (WT and S228P), as well as examples of the sequences of Fc regions in their Fc region modified forms with mutations introduced.
indicates data missing or illegible when filed
The antibody of the present invention may advantageously be one which has a high ratio of monomers having a four-chain structure consisting of two light chains and two heavy chains. Monomer ratio may be measured by SEC-HPLC (see Example described later). Monomer ratio is preferably at least 80%, more preferably at least 90%.
The antibody of the present invention may further undergo functional modifications, such as introduction of amino acid mutations other than those described above, subclass substitution, sugar chain modification, etc.
The antibody of the present invention preferably has an ADCC activity higher than that of IgG1 or IgG1-K322A in which the same Fv region is employed. Compared to antibodies with low ADCC activity, antibodies with high ADCC activity exhibit a high suppressive action on antigen-stimulated T cells. ADCC activity may be examined as follows: for example, in a system using Fc receptor expressing gene-modified Jurkat T cells instead of effector cells, signals from the Fc receptors activate the luciferase gene integrated downstream of NFAT response sequence to thereby generate luciferase, which is then quantified with a luminometer (see Example described later).
The antibody of the present invention preferably comprises a variable region in which each of the Fab regions of a heavy chain and a light chain of an antibody (showing PD-1 agonist activity) derived from a non-human animal (e.g., mouse) has been humanized, and the Fc region of an Fc region-modified form of human IgG1 or IgG4. Exemplary amino acid sequences of the heavy chain and the light chain of preferable antibodies of the present invention and examples of nucleic acid sequences encoding those amino acid sequences are shown in Table V below. The antibody of the present invention may be one which has the heavy chain sequence and/or the light chain amino acid sequence as shown in Table V below. In the antibody of the present invention, the heavy chain and the light chain thereof may comprise an amino acid sequence having at least 90%, 90-95%, 95-99% or more than 99% identity with the amino acid sequence as shown in Table V below. Further, nucleotides encoding the antibody of the present invention may also be nucleotides encoding these heavy chain and light chain amino acid sequences, and such nucleotides may be altered from the nucleotide sequences as shown in Table V depending on the amino acid sequence of interest. The designing of nucleotide sequences in accordance with the amino acid sequences of interest is known to one of ordinary skill in the art, and it is desirable that nucleotide sequences (codons) encoding individual amino acids be optimized in accordance with the amino acid sequences of interest.
Examples of the antibody of the present invention include HM242-C3-X3-K, HM242-C3-X3-KE, HM242-C3-X3-KI, HM242-C3-X3-KSI, HM242-C3-X4-K, HM242-D3-X3-K, HM242-D3-X3-KE, HM242-D3-X3-KI, HM242-D3-X3-KS, HM242-D3-X3-KSI, HM242-D3-X4-K, HM268-K8-X3-K, HM268-K8-X3-KE, HM268-K8-X3-KI, HM268-K8-X3-KS, HM268-K8-X3-KSI and HM268-K8-X4-K. HM242-C3-X3-KI, HM242-C3-X3-KSI, HM242-D3-X3-KI, HM242-D3-X3-KS, HM242-D3-X3-KSI, HM268-K8-X3-KI, HM268-K8-X3-KS and HM268-K8-X3-KSI are preferred, with HM242-D3-X3-KS, HM268-K8-X3-KI, HM268-K8-X3-KS and HM268-K8-X3-KSI being the most preferred.
As used herein, the “functional fragment” typically means a molecule which has one or plurality of the scFv, Fv, F(ab′)2, Fab′ or Fab of the antibody of the present invention, and such molecule preferably binds to Fc receptors. Specific examples of functional fragment include, but are not limited to, an antibody-drug complex composed of the antibody of the present invention and a drug; a polypeptide having the scFv of the antibody of the present invention and the scFv of an anti-Fc receptor antibody; and a fusion protein having the scFv of the antibody of the present invention and an Fc region. When a functional fragment has an Fc region (for example, when a functional fragment is a fusion protein of the scFv of the antibody of the present invention and an Fc region), it is desirable that the Fc region is the Fc region of the Fc region-modified forms described in the present specification. The protein improvement and codon optimization techniques mentioned above may also be applied to functional fragments.
The antibody of the present invention and a functional fragment thereof may be prepared as a recombinant antibody or a functional fragment thereof by genetic engineering techniques. Briefly, a DNA encoding a heavy chain gene and a light chain gene of the antibody of the present invention or a DNA encoding a functional fragment of the antibody of the present invention is synthesized, inserted into an expression vector (e.g., plasmid, bacteriophage, or virus), and then introduced into host cells (e.g., CHO cells, HEK cells, etc.). By culturing the host cells, recombinant antibodies or functional fragments thereof can be obtained from the culture. Codon optimization is preferable in synthesizing a DNA encoding a heavy chain gene and a light chain gene of an antibody or a DNA encoding a functional fragment of an antibody. The heavy chain gene and the light chain gene of the antibody may be inserted into either the same expression vector or different expression vectors. The expression vector may comprise promoters, enhancers, polyadenylation signals, replication origins, selectable marker genes, and so forth. Further, the expression vector may include secretory signal peptide sequences.
The ratio of the heavy chain gene and the light chain gene of the antibody to be inserted into the expression vector is preferably 1:1 but may be appropriately changed in order to improve the expression level of the antibody. Introduction of the recombinant expression vector into host cells may be performed by known methods such as the calcium phosphate method, the DEAE-dextran method, microinjection, lipofection, electroporation, transduction, scrape loading, the shotgun method, etc. When host cells into which a recombinant expression vector has been introduced are cultured, a recombinant antibody or a functional fragment thereof is produced in the culture. Appropriate culture conditions (medium, culture time, culture temperature, CO2 concentration, etc.) can be appropriately selected by one of ordinary skill in the art. The thus produced antibody or a functional fragment thereof may be recovered by known methods of protein separation/purification using isoelectric point, size, solubility (in water, organic solvent, etc.), affinity to a specific substance (substrate, coenzyme, etc. in the case of an enzyme), and so forth. In order to obtain a defucosylated form of a recombinant antibody or a functional fragment thereof, 2-deoxy-2-fluoro-L-fucose is added to the medium and host cells are cultured therein, or alternatively, fucosyltransferase 8 (FUT8)-deficient cells are used as host cells. Other methods may also be used. The defucosylated form may be separated and purified by the same method as described above.
The antibody of the present invention or a functional fragment thereof exerts a more potent T cell-suppressive action than CTLA-4-Ig in human T cell/B cell mixed lymphocyte reaction (MLR), and is capable of exhibiting an anti-inflammatory action mediated by suppression of the function of human T cells (see Example described later).
The antibody of the present invention or a functional fragment thereof may be used for treatment and/or prevention of inflammatory diseases. As used herein, inflammatory diseases refer to diseases caused by excessive inflammation resulting, for example, from wounds, chemical substances, infections, or recognition of self-tissue by immune cells. Inflammatory diseases include autoimmune diseases and diseases classified as type I-IV allergies. Specific examples of such inflammatory diseases include, but are not limited to, Behcet's disease, systemic lupus erythematosus, lupus nephritis, cutaneous lupus erythematosus, chronic discoid lupus erythematosus, serum sickness, systemic sclerosis (systemic scleroderma, progressive systemic sclerosis), multiple sclerosis, scleroderma, polymyositis, dermatomyositis, periarteritis nodosa (polyarteritis nodosa), aortitis syndrome (Takayasu arteritis), malignant rheumatoid arthritis, rheumatoid arthritis, arthritis, juvenile idiopathic arthritis, spondylarthrites, mixed connective tissue disease, Sjogren's syndrome, adult-onset Still's disease, vasculitis, allergic granulomatous angiitis, hypersensitivity angiitis, rheumatoid vasculitis, large-vessel vasculitis, ANCA-associated vasculitis (e.g., microscopic polyangiitis, granulomatosis with polyangiitis, and eosinophilic granulomatosis with polyangiitis), Cogan's syndrome, RS3PE, temporal arteritis, polymyalgia rheumatica, fibromyalgia, antiphospholipid antibody syndrome, eosinophilic fasciitis, IgG4-related diseases (e.g., primary sclerosing cholangitis, autoimmune pancreatitis, etc.), Guillain-Barre syndrome, myasthenia gravis, chronic atrophic gastritis, hepatitis, autoimmune hepatitis, nonalcoholic steatohepatitis, primary biliary cirrhosis, Goodpasture's syndrome, glomerulonephritis, rapidly progressive glomerulonephritis, megaloblastic anemia, hemolytic anemia, autoimmune hemolytic anemia, Coombs positive autoimmune hemolytic anemia, pernicious anemia, autoimmune neutropenia, idiopathic thrombocytopenia purpura, Basedow's disease (Graves disease (hyperthyroidism)), Hashimoto's thyroiditis, autoimmune adrenal insufficiency, primary hypothyroidism, Addison's disease (chronic adrenal insufficiency), idiopathic Addison's disease, type I diabetes mellitus, slowly progressive insulin-dependent diabetes mellitus (latent adult onset autoimmune diabetes), localized scleroderma, psoriasis, psoriatic arthritis, bullous pemphigoid, pemphigus, pemphigoid, herpes gestationis, linear IgA bullous skin disease, epidermolysis bullosa acquisita, alopecia areata, vitiligo, vitiligo vulgaris, Stevens-Johnson syndrome, fixed drug eruption, neuromyelitis optica, chronic inflammatory demyelinating polyneuropathy, multifocal motor neuropathy, sarcoidosis, giant cell arteritis, amyotrophic lateral sclerosis, Harada disease, autoimmune optic neuropathy, idiopathic azoospermia, spontaneous abortion, recurrent fetal loss, infertility, infertility related to lack of fetal-maternal tolerance, inflammatory bowel diseases (e.g., ulcerative colitis, Crohn's disease), celiac disease, ankylosing spondylitis, asthma, severe asthma, infections (viral, bacterial, fungal, parasitic), chronic obstructive pulmonary disease, urticaria, chronic urticaria, transplantation immunity, familial Mediterranean fever, eosinophilic sinusitis, eosinophilic gastrointestinal disease, dilated cardiomyopathy, systemic mastocytosis, inclusion body myositis, pelvic inflammatory disease, Alzheimer's disease, Peyronie's disease, gallbladder disease, pilonidal disease, peritonitis, surgical adhesions, stroke, Lyme disease, meningoencephalitis, autoimmune uveitis, non-infectious uveitis, hypersensitivity, chronic allergic diseases (atopic dermatitis, fever, allergic rhinitis), food allergy, conjunctivitis, anaphylactic shock, contact dermatitis, fibrosing alveolitis, hypersensitivity pneumonitis, allergic bronchopulmonary aspergillosis, allergic encephalitis, IgA nephropathy, Meniere's disease, cholangitis, pancreatitis, Trauma (surgery), acute and chronic graft-vs-host disease, transplant rejection, heart diseases (ischemic diseases such as myocardial infarction, atherosclerosis), intravascular coagulation, osteoporosis, osteoarthritis, periodontitis, hypoxia, gestational herpes, acquired hypogonadism, hypoparathyroidism, and cytokine storm.
The antibody of the present invention or a functional fragment thereof may be used for treatment and/or prevention of inflammatory diseases, especially those diseases in which T cells and autoantibodies are involved, such as systemic lupus erythematosus, lupus nephritis, multiple sclerosis, systemic sclerosis (systemic scleroderma, progressive systemic sclerosis), multiple sclerosis, scleroderma, polymyositis, dermatomyositis, rheumatoid arthritis, Sjogren's syndrome, allergic granulomatous angiitis, large-vessel vasculitis, ANCA-associated vasculitis (e.g., microscopic polyangiitis, granulomatosis with polyangiitis, and eosinophilic granulomatosis with polyangiitis), antiphospholipid antibody syndrome, IgG4-related diseases (e.g., primary sclerosing cholangitis, autoimmune pancreatitis, etc.), idiopathic thrombocytopenia purpura, Basedow's disease (Graves' disease (hyperthyroidism)), Hashimoto's throiditis, psoriasis, psoriatic arthritis, bullous pemphigoid, pemphigus, pemphigoid, alopecia areata, vitiligo, vitiligo vulgaris, Stevens-Johnson syndrome, fixed drug eruption, neuromyelitis optica, inflammatory bowel diseases (e.g., ulcerative colitis, Crohn's disease), asthma, and acute and chronic graft-vs-host disease.
The present invention provides a pharmaceutical composition comprising an agonist antibody to human PD-1 (anti-human PD-1 agonist antibody) or a functional fragment thereof, wherein the antibody or a functional fragment thereof binds to domain #7 of human PD-1 as shown in SEQ ID NO: 9, and a pharmaceutically acceptable carrier. The pharmaceutical composition of the present invention may be used as a pharmaceutical drug.
The pharmaceutical composition of the present invention may be administered to subjects (human or non-human animal) systemically or locally by an oral or parenteral route.
The pharmaceutical composition of the present invention may comprise an effective amount of the anti-human PD-1 agonist antibody or a fragment thereof, and may be formulated into a preparation by mixing, dissolving, emulsifying, encapsulating, lyophilizing, etc. with a pharmaceutically acceptable carrier.
The pharmaceutical composition of the present invention is suitable for either oral or parenteral administration. Preferable preparations for oral administration include, but are not limited to, liquids which have an effective amount of the anti-human PD-1 agonist antibody or a functional fragment thereof dissolved in a diluent such as water or physiological saline; capsules, granules, powders or tablets containing an effective amount of said antibody or fragment as a solid or granules; suspensions which have an effective amount of said antibody or fragment is suspended in an appropriate dispersion medium; and emulsions prepared by first dissolving an effective amount of said antibody or fragment in a solution which is then dispersed and emulsified in an appropriate dispersion medium.
For parenteral administration, the anti-human PD-1 agonist antibody or a functional fragment thereof may be formulated into injections (including lyophilized formulation), suspensions, emulsions, creams, ointments, inhalants, suppositories or the like, together with pharmaceutically acceptable solvents, excipients, binders, stabilizers, dispersants and the like. In the formulation of injections, the anti-human PD-1 agonist antibody or a functional fragment thereof may be dissolved in an aqueous solution (preferably Hanks solution or Ringer solution) or a physiologically compatible buffer such as physiological saline buffer. Further, the pharmaceutical composition of the present invention may take the form of suspension, solution or emulsion, etc. in an oily or aqueous vehicle. Alternatively, the anti-human PD-1 agonist antibody or a functional fragment thereof may be produced in the form of a powder, which may be prepared into an aqueous solution or suspension with sterile water or the like before use. For administration by inhalation, the anti-human PD-1 agonist antibody or a functional fragment thereof may be pulverized to prepare a powder mixture with an appropriate base such as lactose or starch. Suppository formulations may be prepared by mixing the anti-human PD-1 agonist antibody or a functional fragment thereof with a routinely used suppository base such as cocoa butter. Further, the pharmaceutical composition may be formulated as a sustained release preparation by encapsulating in a polymer matrix or the like. Parenteral administration includes, but are not limited to, intravenous, intramuscular, subcutaneous, rectal, nasal, intraoral, and transdermal administration.
The dose may be about 0.1 to 100 mg/kg (body weight) per human adult, and this dose may be administered once or multiple times at intervals of about 1 day to 6 months.
The pharmaceutical composition of the present invention may be used alone. Alternatively, the composition may be used in combination with other therapeutics such as antihistamines, antiallergics, vasoconstrictive nasal sprays, steroids, small molecule immunosuppressants (cyclosporine, tachlorimus, etc.), antibody preparations (e.g., anti-IgE antibody, anti-IL-4 antibody, anti-IL-5 antibody, anti-IL-6 antibody, anti-IL-13 antibody, anti-IL-4 receptor antibody, anti-IL-5 receptor antibody, anti-IL-22 antibody, anti-IL-25 antibody, anti-IL-33 antibody, anti-TNF-α antibody, anti-TSLP antibody, anti-BAFF antibody, etc.), and recombinant soluble fusion proteins (CTLA-4-Ig, etc.). From such a combined use, synergism of drug efficacies can be expected.
Hereinbelow, the present invention will be described more specifically with reference to the following Example.
Regarding commercially available antibodies and recombinant protein: Anti-human PD-1 antibody (clone: EH12.2H7, BioLegend), anti-human PD-1 antibody (clone: J116, Invitrogen), anti-human PD-1 antibody (clone: MIH4, Invitrogen), control mouse IgG1 (clone: MOPC-21, BioLegend), control human IgG1 (clone: QA16A12, BioLegend), control human IgG4 (clone: QA16A15, BioLegend), and recombinant human CTLA-4-Fc region chimeric protein (CTLA-4-Ig, Cat #: 591908, BioLegend) were used. Regarding known antibodies: 949 (anti-human PD-1 antibody described in WO2011/110621, its sequence of Fc region is human IgG4-S228P), PD1AB-6 (PD1AB-6-IgG1-K322A described in WO2017/058859), PD1-17 (anti-human PD-1 antibody described in WO2004/056875, its sequence of Fc region is human IgG1), 3.7C6 (anti-human PD-1 antibody described in WO2020/247648, its sequence of Fc region is human IgG1) and Antibody 1 (anti-human PD-1 antibody described in WO2019/168745, its sequence of Fc region is human IgG1) were prepared by the method described in below.
Techniques for producing anti-human PD-1 mouse monoclonal antibody are known in the art. For example, the method described in Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986) was used. As immunization host, A/J and BALB/c mice were used. As immunogens, a plasmid vector expressing the full length or a mutant of human PD-1 protein (NCBI accession number: NP_005009.2) (15-50 μg), a fusion protein of recombinant human PD-1 extracellular domain and human IgG1-Fc region (25 μg), and 293T cells so engineered to transiently express human PD-1 or a mutant thereof (5×107 cells) were used. One of these immunogens was intramuscularly, intradermally, intraperitoneally or intravenously injected at an interval of 10 to 50 days. When adjuvant was necessary; Sigma adjuvant system (S6322-1VL; Sigma-Aldrich) was used. Three days after the final immunization, splenocytes of the immunization host and P3U1 mouse myeloma cells were fused to prepare hybridomas. Screening for hybridomas producing anti-human PD-1 antibody was performed as follows. Briefly, the culture supernatant of each hybridoma was added to HEK293 cells which had been rendered to express human PD-1. After staining with R-phycoerythrin-labeled goat anti-mouse IgG(H+L) F(ab′)2 fragment (115-116-146; Jackson ImmunoResearch) as secondary antibody; flow cytometry was performed for analysis. Finally selected hybridomas were cloned by limiting dilution analysis, and cultured at high density in a cell line bioreactor (Wheaton). From the resultant culture supernatants, anti-human PD-1 antibodies were purified using Ab-Capture ExTra (P-003-10; Protenova).
In order to specify the variable region sequence of anti-human PD-1 antibodies, frozen hybridomas were sent to BizComJapan, Inc. to use contract analysis services provided by Fusion Antibodies, Plc.
To perform humanization, the complementarity determining region (CDR) of mouse antibody and the residues necessary for retaining CDR conformation of mouse antibody were transplanted into the framework of human antibody.
A heavy chain expression vector was constructed using a DNA encoding a heavy chain sequence and an expression vector (pcDNA3.4, Thermo Fisher Scientific). Likewise, a light chain expression vector was constructed using a DNA encoding a light chain sequence and an expression vector (pcDNA3.4, Thermo Fisher Scientific). The two vectors were mixed in such a manner that the ratio of heavy chain and light chain would be 1:1, and transfected into CHO cells at 0.8 μg of DNA per ml of culture medium using ExpiFectamine CHO Reagent (Thermo Fisher Scientific). One day after the transfection, ExpiCHO Feed and ExpiFectamine CHO Enhancer were added to the cells, which were then cultured at 32° C. for 10 to 12 days. The culture medium was subjected to centrifugation and filtration to remove cells and the supernatant was collected. The amount of antibody in the culture supernatant was measured with Cedex Bio (Roche Diagnostic). The culture supernatant was applied to Protein A column pre-equilibrated with phosphate-buffered saline (PBS), and then the column was washed with PBS. Antibody was eluted with citrate buffer (pH 3.4), and the eluent was neutralized by adding 160 μl of 1M Tris-HCl (pH 9.0) per ml of eluent. After concentration by ultrafiltration, antibody was purified by gel filtration chromatography using Superdex 200 Increase 10/300 GL (Cytiva). Defucosylated antibody was prepared by adding 2-deoxy-2-fluoro-L-fucose (2-DFF) to the culture medium to give a final concentration of 10 to 1000 μM. Comparable levels of FcγRIIIA binding affinity were observed when the final concentration of 2-DFF was within the range of 10 to 1000 μM. Purification of the defucosylated antibody was performed in the same manner as described above.
The purified antibody was suspended in a sample buffer (with reductant and without reductant) and heated at 98° C. for 3 minutes. The resultant sample was applied to 4-15% Mini-PROTEAN TGX polyacrylamide gel (Bio-Rad) and electrophoresed at a constant voltage of 200 V for 30 minutes. Staining was performed with SYPRO Ruby Protein Gel Stain (Thermo Fisher Scientific), followed by analysis with LAS500 (Cytiva).
The ratio of monomers contained in samples was measured by SEC-HPLC. As a column, TSKgel G3000SWXL, 5 μm, 7.8 mm×300 mm (TOSOH) was used. As regards a mobile phase, 84.4 g of potassium dihydrogen phosphate and 66.2 g of dipotassium hydrogen phosphate were dissolved in water to give a 1000 ml solution, which was diluted 10-fold for use. For the ratio of monomers, each corresponding peak area of a sample solution was measured by the automatic integration method, and the proportions of the respective areas were determined by the area normalization method.
Evaluation of Thermal Stability with Differential Scanning Calorimeter
For evaluation of the thermal stability of anti-human PD-1 antibody, denaturation midpoint temperature (Tm) and denaturation onset temperature (Tonset), each used as a predictive indicator, were measured. For 400 μl aliquots of anti-human PD-1 antibody sample (diluted to 0.1 mg/ml), calorimetric change that occurred when temperature was raised from 25° C. to 100° C. at a rate of 200° C./hr was measured with a differential scanning calorimeter (Microcal PEAQ-DSC; Malvern Panalytical). As a reference for measurement, PBS was used. With respect to the resultant data, baseline correction based on the result of buffer measurement and fitting analysis with a non-two-state model were conducted. Then, Tm and Tonset for Fab, CH2 and CH3 regions were calculated.
The binding affinity of each antibody to human PD-1 was measured with Biacore 8K (Cytiva). Briefly, anti-His Tag antibody was immobilized by amine coupling using Amine Coupling Kit and His Capture Kit, onto 8 channels of a sensor chip (provided with 16 flow cells; one channel consisting of two flow cells) which had carboxymethyl dextran immobilized on a gold film. The immobilization was performed by feeding an activation solution (NHS and EDC mixed in equal amounts) at a flow rate of 10 μl/min for 7 minutes, an immobilization solution at a flow rate of 5 μl/min for 7 minutes, and 1 M ethanolamine hydrochloride-NaOH (pH 8.5) at a flow rate of 10 μl/min for 7 minutes. The buffer used at the time of immobilization was 1×HBS-EP+ Buffer. After the immobilization and before applying a sample, Human PD-1/PDCD1 Protein (PD-1) adjusted to 1 μg/ml with 1×HBS-EP+ Buffer was fed to one of the two flow cells of each channel in an amount of 20 μl at a flow rate of 10 μl/min to thereby capture PD-1 on the sensor chip. As regards the other flow cell, 1×HBS-EP+ Buffer was fed in the same manner to prepare a control flow cell. Each antibody was adjusted to 16 nM with 1×HBS-EP+ Buffer, and then subjected to 2-fold 7-step serial dilution using 1×HBS-EP+ Buffer. The resultant sample solution (30 μl) was fed to each flow cell at a flow rate of 10 μl/min and the response was measured. After subtracting the response of control flow cell from the response of the flow cell which was allowed to capture the ligand, KD value was calculated by curve fitting using Biacore 8K Evaluation Software.
The binding affinity of each antibody to Fc receptors [FcγRIIB/C (CD32b/c) (FcγRIIB) (R&D Systems) and Fc gamma RIIIA/CD16a, CF (FcγRIIIA(V158)) (R&D Systems)] was measured with Biacore 8K. Briefly, anti-His Tag antibody was immobilized by amine coupling using Amine Coupling Kit and His Capture Kit onto 8 channels of a sensor chip (provided with 16 flow cells; one channel consisting of two flow cells) which had carboxymethyl dextran immobilized on a gold film. The immobilization was performed by feeding an activation solution (NHS and EDC mixed in equal amounts) at a flow rate of 10 μl/min for 7 minutes, an immobilization solution at a flow rate of 5 μl/min for 7 minutes, and 1 M ethanolamine hydrochloride-NaOH (pH 8.5) at a flow rate of 10 μl/min for 7 minutes. The buffer used at the time of immobilization was 1×HBS-EP+ Buffer. After the immobilization and before applying a sample, Fc receptor adjusted to 1 μg/ml with 1×HBS-EP+ Buffer was fed to one of the two flow cells of each channel in an amount of 20 μl at a flow rate of 10 μl/min to thereby capture Fc receptor on the sensor chip. As regards the other flow cell, 1×HBS-EP+ Buffer was fed in the same manner to prepare a control flow cell. Each antibody was adjusted to 1 mg/ml with 1×HBS-EP+ Buffer, and then subjected to 2-fold 7-step serial dilution using 1×HBS-EP Buffer (when measuring FcγRIIIA(V158), 4-fold 7-step serial dilution was performed). The resultant sample solution (15 μl) was fed to each flow cell at a flow rate of 30 μl/min and the response was measured. After subtracting the response of control flow cell from the response of the flow cell which was allowed to capture the ligand, KD value was calculated by analysis of equilibrium constants using Biacore8 K Evaluation Software.
The activity of anti-human PD-1 antibody was evaluated as an effect on cytokine production caused by interaction between human PD-1-expressing T cells and antigen-presenting cells. Prepared as T cells were: a cell line obtained by treating DO11.10 T cell hybridoma cell line (kindly provided by Department of Immunology and Genomic Medicine, Graduate School of Medicine, Kyoto University) with Cas9 (Invitrogen) to knock-out mouse PD-1; and the knock-out cell line which was rendered to express human PD-1. Prepared as antigen-presenting cells were: a cell line obtained from IIA1.6 B cell line (kindly provided by Department of Immunology and Genomic Medicine, Graduate School of Medicine, Kyoto University) by knocking out mouse PD-L1; and the knock-out cell line which was rendered to express human PD-L1 or mouse FcγRIIB. For evaluating human PD-1 agonist activity, mouse or human FcγRIIB-expressing IIA1.6 cells were used; and for evaluating antagonist activity, human PD-L1-expressing IIA1.6 cells were used. Further, for evaluating dependency on each Fc receptor, IIA1.6 cells expressing various human Fc receptors were used. Human PD-1-expressing DO11.10 T cell hybridomas and respective IIA1.6 cells were suspended in a medium (10% fetal bovine serum-containing RPMI1640 medium). DO11.10 T cell hybridomas and IIA1.6 cells were seeded in round bottom 96-well plates at 5×104 and 1×104 cells/well/50 μl, respectively. Anti-human PD-1 antibody was added to the plates to give a final concentration of 5, 0.5, 0.05 or 0.005 μg/ml at 50 μl/well. Subsequently, OVA323-339 peptide (Eurofins) was added as antigen to give a final concentration of 2 μg/ml at 50 μl/well. Eighteen hours later, IL-2 concentration in the culture supernatant was measured using mouse IL-2 DuoSet ELISA (R&D Systems).
In the evaluation of human PD-1 agonist activity, the cytokine suppression obtained by using human PD-L1-expressing IIA1.6 cells was taken as positive control; and the result obtained by using DO11.10 T cell hybridomas not rendered to express human PD-1 was taken as negative control. In the evaluation of human PD-1 antagonist activity, anti-human PD-1 antibody (clone: EH12.2H7) with known antagonist activity was used.
Measurement of Affinity to Human Fc Receptor-Expressing Cell Line with Flow Cytometer
Various antibodies (5 μg/ml, 50 μl each) were added to IIA1.6 cells which had been rendered to express human FcγRIIB or human FcγRIIIA. The cells were incubated at 4° C. for 15 minutes. Then, cells were washed, and antibodies bound to cell surfaces were stained by incubating with APC-labeled anti-human IgG Fab antibody at 4° C. for 15 minutes. After washing cells again, antibodies bound to cell surfaces were detected by FACS analysis and GMFI was calculated as affinity to FcγRIIB or FcγRIIIA.
The gene of human PD-1 (wild-type) or substituted human PD-1 in which regions 1 to 8 as shown in
Preparation of Human PD-1 Knock-In Mouse using Genome Editing
gRNA targeting 5′-GCCAGGGGCTCTGGGCATGT-3′ and a donor vector containing human PD-1 gene was microinjected into C57BL/6N mouse-derived pronuclear stage fertilized egg together with Cas9 protein (Invitrogen). The resultant egg was transplanted into the oviduct of foster mouse. With respect to the resultant mice (F0), indel mice were selected and mated with wild-type mice to obtain F1 mice. Those F1 mice confirmed for gene transfer were further mated to thereby obtain homozygous mice. These homozygous mice were used in experiments as human PD-1 knock-in mice.
Evaluation of PD-1 Agonist using House Dust Mite Extract (HDM)-Induced Allergy Model
In experiments under preventive regimen, HDM (D. Pteronyssinus; Greer) (total protein: 400 ng; 10 ng in terms of Derp1) was administered intraperitoneally to human PD-1 knock-in mouse. Simultaneously, 500 μg of anti-human PD-1 antibody was administered intraperitoneally (Day 0). Further, the same dose of anti-human PD-1 antibody was administered intraperitoneally after 3 days, 7 days and 10 days. In experiments under therapeutic regimen, anti-human PD-1 antibody was administered only once after 10 days. From 7 days after intraperitoneal administration of HDM, HDM (total protein 25 μg/25 μl) was administered intranasally under anesthesia. This intranasal administration was performed for 8 consecutive days. Four hours after final intranasal administration, mice were euthanized after collecting blood samples, and bronchoalveolar lavage fluid and the lung were collected. Mononuclear cells infiltrating into the alveoli were isolated from the bronchoalveolar lavage fluid. The lung was subjected to enzyme treatment and density gradient centrifugation to isolate mononuclear cells and the numbers of CD4+ T cells, eosinophils (CD11c″ SiglecF+) and the like were determined by FACS analysis. Intracellular cytokines in bronchoalveolar lavage fluid-derived mononuclear cells and lung-derived mononuclear cells were analyzed by FACS. Blood concentration of HDM specific IgE was measured by ELISA (mouse serum anti-HDM IgE antibody assay kit; Chondrex).
Evaluation of PD-1 Agonist using Acute Graft-Versus-Host Disease (GVHD) Model
Cyclophosphamide (Fuji Film Wako Pure Chemical) dissolved in PBS was intraperitoneally administered to BDF1 mice (recipient, H-2Kb+H-2Kd+) at a dose of 100 mg/kg (Day-1). One day later, splenocytes were collected from human PD-1 knock-in mice (donor, H-2Kb+H-2Kd−), washed with PBS, and suspended in PBS at 5×107 cells/200 μl. This cell suspension was transferred into BDF1 mice in 200 μl via the tail vein (Day 0). After Day 0, anti-human PD-1 agonist antibody was administered intraperitoneally once every 3 to 4 days at a dose of 500 μg/mouse. For evaluating GVHD symptoms, body weights of mice were measured after 4, 7, 11 and 14 days and the percent weight loss from Day 0 was calculated. Further, blood samples were taken after 7 and 14 days. Plasma was collected by centrifugation, and then PBMC was isolated by density gradient centrifugation. After 14 days, mice were euthanized. The spleen and the liver were collected and cell suspensions were prepared. These suspensions were analyzed with a flow cytometer and cell counting was performed. From the results obtained, rates of PBMC and donor cells (H-2Kb+H-2Kd−) in the spleen or the liver and respective cell counts were calculated. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in plasma were measured with Transaminase CII-Test Wako (Fuji Film), and IFN-γ concentration in plasma with Mouse IFN-γ ELISA MAX Deluxe (BioLegend).
Wild-type mice were immunized at two subcutaneous sites in the back with 100 μg each of MOG35-55 peptide (Eurofins) emulsified by mixing with Freund's complete adjuvant. Pertussis toxin (Fuji Film Wako Pure Chemical, 400 ng) was administered to the mice via the tail vein. After 1 day, pertussis toxin (400 ng) was administered again. After 13 days, cells were isolated from the spleen and the axillary lymph node. CD8+ cells and Gr-1+ cells were removed with AutoMACS. The remaining cells were seeded in flat bottom 12-well plates at 1.05×107 cells/well and cultured in the presence of 20 μg/ml MOG peptide and 10 ng/ml IL-23 for 5 days. On the first and second days of culture, cells were rendered to express human PD-1 using retrovirus. After 5 days of culture, cells were harvested and their concentration was adjusted with PBS so that the number of activated CD4+ T cells in lymphocytes would be 1×106 in 200 μl. The resultant cell sample was transferred into wild-type mice via the tail vein at 200 μl/mouse (Day 0). After Day 0 of cell transfer, anti-human PD-1 agonist antibody was administered intraperitoneally at 500 μg/mouse once every 3 days. Measurement of body weight and scoring were performed at the timing indicated in
Cells were prepared from the spleen and the lymph nodes of human PD-1 knock-in mice, and CD25−CD4+ cells were isolated with a cell sorter. These cells were suspended in PBS at 5×105 cells/200 μl and transferred into RAG2 KO mice in 200 μl via the tail vein (Day 0). From Day 0, anti-PD-1 agonist antibody was administered to the mice intraperitoneally once every 3 to 4 days at 500 μg/mouse. Body weight was measured twice a week and the percent weight loss from Day 0 was calculated. At week 8, mice were euthanized, the colon was collected and its length was measured. Further, CD4+ T cells infiltrating inherently into the lamina propria of the colon were collected and analyzed with a flow cytometer after staining intracellular cytokines.
Cyclophosphamide (Fuji Film Wako Pure Chemical) dissolved in PBS was intraperitoneally administered to BDF1 mice (recipient, H-2Kb+H-2Kd+) at a dose of 100 mg/kg (Day-1). After 1 day, splenocytes were collected from human PD-1 knock-in mice (donor, H-2Kb+H-2Kd−) and freed of CD8+ cells using AutoMACS. Subsequently, the resultant cells were washed with PBS, and suspended in PBS at 2×107 cells/200 μl. This cell suspension was transferred into BDF1 mice in 200 μl via the tail vein (Day 0). From Day 0, anti-human PD-1 agonist antibody was administered intraperitoneally once every 3 to 4 days at a dose of 500 μg/mouse. For evaluating lupus-like symptoms, blood samples were weekly taken for 11 weeks, and plasma concentration of IL-21 or anti-dsDNA antibody was measured by ELISA. After 7 or 14 days, mice were euthanized and the spleen was collected to prepare cell suspensions. These donor-derived Tfh cells (CXCR5+ICOS+H-2Kd−CD4+) or recipient-derived germinal center B cells (CD95+GL7+H-2Kd+B220+), class-switched B cells (IgD−IgM−H-2Kd+B220+), and plasma cells (CD138+H-2Kd+CD19+) were analyzed with a flow cytometer.
In mixed lymphocyte reaction of human CD4+ T cells (LONZA) and human CD19+ B cells (Precision for Medicine) both isolated from PBMC of healthy donors by negative selection, the activity of anti-human PD-1 agonist antibody against T cell's reaction to alloantigen was evaluated. Specifically, human T cells (2×105 cells) stained with CFSE were mixed with B cells (1×105 cells), and the mixture was seeded in round bottom 96-well plates at 200 μl/well. After culturing for 6 to 7 days, cells were harvested. The harvested cells were freed of B cells by negative selection with CD19 antibody. Then, B cells from the same donor were newly added to re-stimulate the T cells. At the time of this re-stimulation, anti-human PD-1 antibodies or CTLA-4-Ig were added simultaneously to give a final concentration of 5, 0.5, 0.05 or 0.005 μg/ml. After 12 hours, the culture supernatant was collected, and IFN-γ concentration was measured with Human IFN-γ ELISA MAX Deluxe (BioLegend).
As a target cell, Raji-hPD-1 cells (InvivoGen) were suspended in a medium at 1.1×106 cells/ml and seeded in round bottom 96-well plates at 90 μl/well. Simultaneously, anti-human PD-1 antibody was added to the plates at 20 μl/well to give a final concentration of 1000, 300, 100, 30, 10, 3, 1, 0.3, 0.1, 0.03 or 0.01 ng/ml. After one hour culture, Jurkat-Lucia NFAT-CD16 Cells (InvivoGen) were suspended as effector cells in a medium at 2.2×106 cells/ml and added thereto at 90 μl/well. After 6-hour culture, 50 μl of QUANTI-Luc solution (InvivoGen) was added to 20 μl of the culture supernatant. Luminescence intensity was measured with a microplate reader.
Frozen PBMC from healthy donor (Precision for Medicine) were thawed, suspended in a medium at 1×107 cells/2 ml, and seeded in 12-well plates at 2 μl/well. After 24-hour preculture, cells were collected, stained with CFSE, suspended in a medium at 3×106 cells/ml, and seeded in round bottom 96-well plates at 100 μl/well. As an antigen to this cell, tetanus toxoid (Merck) was added at 50 μl/well to give a final concentration of 1 μg/ml. Further, anti-human PD-1 antibodies or CTLA-4-Ig were added at 50 μl/well to give a final concentration of 5, 0.5, 0.05 or 0.005 μg/ml. After 4-5 days, proliferation of CD4+ T cells linked to CFSE as an indicator was evaluated with a flow cytometer. Further, IFN-γ concentration in the culture supernatant was measured with Human IFN-γ ELISA MAX Deluxe (BioLegend).
Evaluation of PD-1 Agonist with Respect to ERK Phosphorylation
Human PD-1-expressing DO11.10 T cell hybridomas and mouse FcγRIIB-expressing IIA1.6 cells were suspended in a medium (10% fetal bovine serum-containing RPMI1640 medium). DO11.10 T cell hybridomas and IIA1.6 cells were seeded in round bottom 96-well plates at 5×104 and 1×104 cells/well/50 μl, respectively. Control mouse antibody (MOPC-21) or anti-human PD-1 antibody HM266 was added to the plates at 50 μl/well to give a final concentration of 5 μg/ml. Subsequently, OVA323-339 peptide (Eurofins) was added as antigen at 50 μl/well to give a final concentration of 2 μg/ml. Immediately after the addition of OVA (0 hour) or after incubating at a 37° C., 5% CO2 for 2, 4 or 6 hours, cells were harvested. The resultant cells were fixed with 4% paraformaldehyde for 15 minutes, washed, and permeabilized with 0.1% Triton X-100 for 15 minutes. After washing, the cells were suspended in 50% methanol-PBS and left standing overnight at −20° C. After washing, the cells were stained with Alexa488-anti-phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) mAb, PE-anti-DO11.10 TCR mAb and APC-anti-B220 mAb. Stained cells were analyzed by FACS, and phosphorylated ERK positive ratio in DO11.10 (DO11.10 TCR+B220−) was calculated.
Co-precipitate of NP-OVA (total protein: 100 μg) and alum adjuvant were administered intraperitoneally to human PD-1 knock-in mice; simultaneously, 500 μg of anti-human PD-1 antibody was administered intraperitoneally (Day 0). The same dose of anti-human PD-1 antibody was further administered intraperitoneally 3 days and 7 days after the first administration. Seven and 10 days after NP-OVA administration, blood samples were collected. Antigen-specific antibody titer and the amount of IL-21 in blood were measured by ELISA. Antibodies (anti-NP2 and anti-NP25) binding to a complex of 2 or 25 molecules of NP binding to one molecule of bovine serum albumin (BSA) were measured by ELISA, and were defined either as high affinity antigen-specific antibody or as low affinity antigen-specific antibody. Further, mice were euthanized 10 days after immunization with antigen, and the spleen was collected to prepare cell suspensions. These Tfh cells (CXCR5+ICOS+CD4+), germinal center B cells (CD95+GL7+B220+), class-switched B cells (IgD−IgM−B220+) and plasma cells (CD138+B220−) were analyzed with a flow cytometer.
Activated human T cells were prepared as described below. Briefly, human CD4+ T cells isolated from PBMC of healthy donors by negative selection (LONZA) were adjusted to 1×106 cells/ml, seeded in anti-CD3 antibody (3 μg/ml) immobilizing 24-well plates at 1 ml/well, and stimulated for 3 days. Human Tfh-like cells were prepared as described below. Briefly, naïve T cell (CD45RA+CXCR5−CD11c−) fraction was sorted from human CD4+ T cells with a cell sorter and adjusted to 1×106 cells/ml. 25 μl of pre-washed Dynabeads Human T-activator CD3/28 (Thermo Fisher) per 1 ml of the sorted cells was added. The resultant mixture of cells and beads was seeded at 1 ml/well in 24 well plates. On the following day, cells which had been stimulated with the beads were harvested. After removal of the beads, cells were adjusted to 5×105 cells/ml, and seeded at 100 μl/well in anti-CD3 antibody (5 μg/ml) immobilized flat bottom 96-well plates. Subsequently, Human IL-23 (final concentration: 25 ng/ml), Human TGFβ (final concentration: 5 ng/ml), Human IL-12 (final concentration: 1 ng/ml) and anti-human CD28 (final concentration: 1 μg/ml) were added to make 200 μl/well. Then, plates were incubated at a 37° C., 5% CO2 for 48 to 72 hours. The thus cultured cells were analyzed with FACS, and it was confirmed that more than 50% of the cells became Tfh-like cells (CXCR5+ICOS+PD-1+). THP-1 cell line (parental cell line or a cell line overexpressing human FcgγRIIB) was suspended in dilute solution of mitomycin C (500 μg/ml) at 1×107 cells/ml, incubated at 37° C. for 2 hours and washed 3 times with a culture medium. The thus prepared activated human T cells or human Tfh-like cells were seeded in round bottom 96-well plates at 5×104 cells/well and the THP-1 cells at 2.5×104 cells/well. Further, as a stimulant, CytoStim (Miltenyi Biotec) was added at 0.2 μl/well and anti-human PD-1 antibodies or CTLA-4-Ig were added to give a final concentration of 5-5000 ng/ml. Then, plates were cultured at a 37° C., 5% CO2 for 18 hours. After the culture, the supernatant was collected, and cytokine concentrations were measured using IFN-γ ELISA MAX Deluxe (BioLegend) and Human IL-21 Duoset ELISA (R&D).
Evaluation of PD-1 Agonist in the Autoantigen Response of PBMC from Autoimmune Disease Patients
RA or SLE patient-derived PBMC (Precision for Medicine, STEMCELL Technologies) was thawed, suspended in a medium at 1×107 cells/2 ml and seeded in 12-well plates at 2 ml/well. After 16-24 hour preculture, cells were collected, stained with CFSE, suspended in a medium at 3×106 cells/ml and seeded in round bottom 96-well plates at 100 μl/well. As antigens to these cells, citrullinated fibrinogen (Cayman) for RA patient-derived PBMC and Sm antigen (AROTEC DIAGNOSTICS) for SLE patient-derived PBMC were added at 50 μl/well to give final concentrations of 10 μg/ml and 5 μg/ml, respectively. Further, anti-human PD-1 antibodies or CTLA-4-Ig were added at 50 μl/well to give a final concentration of 5, 0.5, 0.05 or 0.005 μg/ml. After 6 to 7 days, proliferation of CD4+ T cells linked to CFSE as an indicator was evaluated with a flow cytometer. Further, IFN-γ concentration in the culture supernatant was measured with Human IFN-γ ELISA MAX Deluxe (BioLegend).
There are anti-PD-1 antibodies which are claimed to suppress T cell activity by PD-1 agonist activity (Ref. 1). However, it has never been demonstrated if any anti-PD-1 antibodies can really stimulate PD-1. Preceding researches discuss T cell inhibition relying on an artificial system based on the effect of immobilized anti-PD-1 antibody on T cell stimulation with immobilized anti-CD38 antibody. However, such an artificial condition does not faithfully reproduce T cell activation in vivo. It is a different question whether the administration of anti-PD-1 antibody in a free form can suppress in vivo T cell activation in the interaction with antigen-presenting cells.
To find out anti-human PD-1 agonist antibody that functions under physiological conditions, an ideal evaluation system will be the one reproducing antigen-specific T cell activation by the interaction between T cells and antigen-presenting cells. In addition, a test system that permits easy modification of experimental conditions is useful for the analysis of the mechanism of action. A combination of established cell lines, DO11.10 T cell hybridoma and IIA1.6 cells as antigen-presenting cells, provides a useful platform, which enables antigen-specific T cell activation and the analysis of mechanism of action in an efficient and reproducible manner (
DO11.10 T cell hybridoma cells expressing human PD-1 were confirmed to clearly reduce IL-2 production upon antigen stimulation by PD-L1-expressing IIA1.6 cells in a PD-1-dependent mechanism (
To find and characterize biologically functional anti-human PD-1 antibodies, a number of monoclonal antibodies binding to various domains of human PD-1 were prepared. Hybridomas producing these antibodies were generated from human PD-1-immunized mice. The antibodies were classified by the binding specificity to the eight putative domains constituting the molecular surface of PD-1 that were selected based on the structure of human PD-1 (
Next, the obtained antibodies were tested for the agonistic activity. The detection of immunosuppressive activity by anti-PD-1 agonist antibodies should be conducted in the absence of PD-L1-dependent immunosuppression. The screening system for agonistic antibodies is different from that for blocking antibody. This system involves a combination of human PD-1-expressing DO11.10 T cell hybridoma and IIA1.6 cells which were made to be lacking PD-L1 but expressing FcγRIIB. The present inventors evaluated the anti-human PD-1 antibody panel with a variety of binding epitopes as well as commercially available anti-human PD-1 monoclonal antibodies for biological activities using both screening systems for blocking antibodies and agonist antibodies. As a result, agonist antibodies and blocking antibodies with various degrees of activity were obtained (
The biological activities were correlated with the binding site of antibodies, and agonist antibodies and blocking antibodies found clearly distinct binding sites (
Since the anti-human PD-1 agonist antibodies were selected in the experimental system involving antigen-specific T cell activation in the interaction with antigen-presenting cells, the immunosuppressive activity by the addition of free anti-human PD-1 antibody indicates its promise to function under physiological conditions. The current agonist antibodies did not affect PD-1-deficient T cells, clearly showing the PD-1-dependent immunosuppressive action. When anti-CD3 antibody was co-immobilized with anti-PD-1 as in the other studies, the immobilized amount of anti-CD3 antibody may substantially decrease due to some distraction by anti-PD-1 antibody used in combination. On such an occasion, apparent activation of T cells is suppressed, but this is caused by attenuation of stimulation from anti-CD3 antibody and what was observed is a non-specific phenomenon irrelevant to the function of anti-PD-1 antibody. Actually, the present inventors have also experienced such cases. To our experience, when this type of experimental system is to be used in PD-1 studies, it will be very important to pay attention to the immobilized amount of anti-CD3 antibody, and PD-1-dependent immunosuppression should be confirmed using PD-1 deficient T cells.
The anti-human PD-1 agonist antibodies shown by the present inventors suppress activated T cells in vitro, implying efficacy on a wide variety of inflammatory diseases. Activated T cells have been reported to be involved in various diseases in different organs and allergic diseases by type 1 hypersensitivity (Ref. 2). For example, it is known that the neutralization of activated T cell-derived cytokines decreases inflammatory responses in these disease (Ref. 3). Cancer patients under treatment with anti-PD-1 blocking antibody often experience symptoms similar to diseases with strong T cell involvement as adverse events (Ref. 4). In animal experiments, PD-1 induction is observed in T cell-dependent inflammation (Ref. 5), and PD-1-deficiency or PD-1 blockade result in exaggerated inflammation (Ref. 6), showing that PD-1 is very important for the regulation of activated T cells in pathological conditions. Therefore, therapeutic intervention targeting PD-1 is expected to suppress activated T cells involved in pathological conditions and has a potential to attenuate a wide spectrum of proinflammatory activities in various inflammatory diseases. Next, the present inventors examined the anti-inflammatory action of PD-1 agonist antibodies in multiple mouse inflammation models.
Anti-inflammatory property of PD-1 agonist antibody in vivo was tested using an animal experimental model of allergic asthma induced by inhalation of dust mite antigen. Preventive administration of anti-human PD-1 agonist antibody HM266 to human PD-1 knock-in mice reduced infiltration of eosinophils and CD4+ T cells into alveolar space and IL-4-, IL-5- and IL-13-producing CD4+ T cells (
The anti-inflammatory efficacy of anti-human PD-1 agonist antibodies was further examined in other T cell-mediated disease models such as multiple sclerosis model (EAE model), colitis model, acute GVHD and chronic GVHD (lupus-like) model. These disease models were induced by T cell transfer, and treatment with anti-human PD-1 agonist antibody started at the same time as T cell transfer. As a result, Encephalomyelitis score in EAE model was significantly reduced by HM266 treatment (
The consistent anti-inflammatory action of anti-human PD-1 agonist antibodies in multiple T cell-mediated disease models suggested that these antibodies in present invention are useful as a therapeutic strategy for a wide range of inflammatory diseases. Neutralizing antibodies against T cell-derived cytokines that have already been in clinical use may require continuous treatment in order to sustain anti-inflammatory effects; however, anti-human PD-1 agonist antibody may be able to achieve long-term suppression of T cells and to cure of diseases. Specifically, the remarkable suppression of CD8+ T cells in acute GVHD model suggests that anti-human PD-1 agonist antibodies will be effective as therapeutics for GVHD and alopecia areata in which CD8+ T cells is mainly involved. Since these antibodies could suppress CD4+ T cells in asthma, colitis and EAE models, anti-human PD-1 agonist antibodies will be applicable as therapeutics for allergic disorders, colitis, multiple sclerosis, as well as psoriasis, in all of which CD4+ T cells play an important role. The blockade of autoantibodies in the lupus-like model suggests that anti-PD-1 agonist antibodies are applicable as therapeutics for diseases involving autoantibody, as exemplified by lupus nephritis and rheumatoid arthritis.
The present inventors converted the anti-human PD-1 agonist antibodies into humanized antibodies in order to enable application in humans. The seven anti-human PD-1 agonist antibodies that are originally mouse IgG were converted to chimeric antibodies by replacing the sequences of Fc regions to those of human IgG1-K322A. The agonistic activity of the resultant chimeric antibodies was evaluated using human PD-1-expressing cell lines and human FcγRIIB-expressing cell lines, and all the chimeric antibodies showed an activity comparable to that of the original mouse IgG (
These humanized antibodies were tested for the immunosuppressive activity in primary human lymphocytes. Human peripheral blood mononuclear cell-derived CD4+ T cells were mixed with B cells from other individual, and the extent of allogenic T cell activation was compared. While CTLA-4-Ig inhibited T cell response, none of these four anti-human PD-1 agonist antibodies showed a significant suppressive activity (
To solve this problem, the present inventors sought to modify the humanized anti-human PD-1 agonist antibodies. Some agonist antibodies require antibody crosslinking to exert their agonistic activity, and Fc receptors play an important role as a means of antibody crosslinking. However, no one has addressed whether Fc receptor-dependent crosslinking is required for the activity of anti-human PD-1 agonist antibodies. In the validation by the present inventors, anti-human PD-1 agonist antibodies failed to exert agonist activity in the absence of Fc receptors. In contrast, in the presence of Fc receptors, anti-human PD-1 agonist antibodies exerted agonistic activity, and FcγRIIB was the most effective for the activity exerted by the anti-human PD-1 agonist antibody among several existing human Fc receptors (
The agonistic activity of humanized anti-human PD-1 agonist antibody increases dependent on the human FcγRIIB levels on the antigen-presenting cells (
Then, the present inventors sought to increase the affinity of agonist antibodies to human FcγRIIB. Briefly, by introducing amino acid modifications into the Fc region and so on, the inventors have found combinations of modifications which can selectively improve the affinity to FcγRIIB. These modified Fc regions were attached to anti-human PD-1 agonist antibodies to produce chimeric antibodies. The resultant chimeric antibodies demonstrated the enhanced agonistic activities even under conditions of low expression of FcγRIIB, and they showed an improved immunosuppressive activity in primary human cells (
Some of the Fc variants not only promoted PD-1 agonist activity via improved FcγRIIB affinity but also increased affinity to FcγRIIIA, which is responsible for the FcγRIIIA-mediated ADCC activity (
So far, for cell surface molecules such as TNF receptor superfamily, some antibodies targeting these cell surface molecules were known to exert agonistic activity in an FcγRIIB-dependent manner (Ref. 7). These agonist antibodies trigger the signal transduction through TRAFs (TNF receptor-associated factors), it is downstream of target molecules, and so on as a result of the binding to FcγRIIB and the crosslinking of target molecules (Ref. 8). Conversely, PD-1 has an immunoinhibitory domain (ITIM/ITSM) in its intracellular domain, recruits phosphatase SHP-1/2, and dephosphorylates various immune activation signaling molecules, thereby causing immunosuppression (Ref. 9). The agonistic signal transduction by the anti-human PD-1 antibody related to the present invention is completely different from signal transduction by known antibodies in terms of the molecular species and the intracellular signaling mode. Indeed, the FcγRIIB-dependency of the agonist antibodies to the molecules of the immunoglobulin superfamily; to which the anti-PD-1 agonist antibody belongs has never been reported before, and it is unpredictable that anti-human PD-1 agonist antibodies in the present invention show the same characters as known antibodies as above. Furthermore, it is also unpredictable that the agonistic activity of anti-human PD-1 antibodies emerges only when the antibody has bound to the specific domain of human PD-1.
Anti-CD40 antibody, one of the agonist antibodies to the above-described TNF receptor family, has proceeded to clinical trial; however, the therapeutic outcome was not as remarkable as the efficacy shown in mouse models (Ref. 10). In this report, they speculate that Fc modification to increase the binding of the antibodies to FcγRIIB may improve the clinical response in humans. However, with respect to the anti-PD-1 agonist antibody, the dependence on FcγRIIB has not previously reported in the first place and the necessity of high affinity of anti-PD-1 agonist antibody to FcγRIIB has never been known at all. The novel ideas by the present inventors of enhancing the affinity of anti-human PD-1 agonist antibodies for FcγRIIB enables the antibodies in the present invention to exert adequate PD-1 agonistic activity and achieve immunosuppressive treatment of inflammatory diseases even in the human physiological condition with low FcγRIIB expression level. Further, it has not known before that the Fc modification to enhance ADCC activity further improves the immunosuppressive effect of anti-human PD-1 agonist antibodies. The enhancement of PD-1 agonist activity, together with ADCC activity, by the modification of Fc region produce potent immunosuppressive anti-human PD-1 antibodies compared to pre-existing anti-human PD-1 antibodies, thus showing the inventive step of the antibody of the present invention.
All publications, patents and patent applications cited herein are incorporated herein by reference in their entirety.
The present invention is applicable to treatment and/or prevention of inflammatory diseases.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2021-081913 | May 2021 | JP | national |
| 2021-086534 | May 2021 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2022/020011 | 5/12/2022 | WO |
