WO2025248815A1 - Antibody or fragment thereof, and anti-influenza vaccine - Google Patents

Abstract

The present invention addresses the problem of providing an anti-idiotype antibody against a broadly neutralizing antibody and an antigen that is produced using the anti-idiotype antibody and can induce reactivity to a wide range of subtypes of influenza viruses.　The present invention provides an antibody or a fragment thereof which specifically binds to a "MEDI8852 antibody" or a "FluA-20 antibody" that is a specific anti-influenza HA broadly neutralizing antibody.

Description

Antibody or fragment thereof, and anti-influenza vaccine

The present invention relates to an antibody or a fragment thereof, and an anti-influenza vaccine.

Many pathogenic viruses, such as influenza viruses, can evade the host immune system by altering the immunogenicity of proteins present on their surface.
As a result of these changes, many different subtypes of pathogenic viruses arise.

Subtypes of pathogenic viruses can reduce the effectiveness of vaccines. Therefore, there is a need for the development of a vaccine (sometimes called a "universal vaccine") that can protect against infection with a single dose, regardless of subtype.

A known method for developing a universal vaccine is the use of antibodies (called broadly neutralizing antibodies, or "bnAbs") that can broadly cross-react and neutralize viral antigens with different amino acid sequences (see, for example, Non-Patent Document 1).

Front Immunol 2021 Oct 19:12:708227. doi: 10.3389/fimmu. 2021.708227.

However, antigens and immunization methods capable of inducing broadly neutralizing antibodies that protect against a wide range of influenza virus subtypes have not yet been fully established.
Under these circumstances, the present inventors aimed to develop an anti-idiotype antibody (hereinafter also referred to as "anti-Id antibody") against a broadly neutralizing antibody. Such an anti-idiotype antibody is expected to function as an effective vaccine against influenza viruses.

The present invention was made in light of the above circumstances, and aims to provide anti-idiotype antibodies against broadly neutralizing antibodies, and antigens that can elicit reactivity against a wide range of influenza virus subtypes using such anti-idiotype antibodies.

As a result of investigations, the inventors have newly discovered that the above-mentioned problems can be solved by using an antibody or a fragment thereof composed of a specific sequence, and have thus completed the present invention. More specifically, the present invention provides the following:

(1) An antibody or fragment thereof that specifically binds to the "MEDI8852 antibody," an anti-influenza HA broadly neutralizing antibody, and that satisfies all of the following requirements A:
[Requirement A]
- Heavy chain CDR1 is the amino acid sequence shown in SEQ ID NO:2.
- Heavy chain CDR2 is the amino acid sequence shown in SEQ ID NO:3.
The 27th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Tyr, Phe, Leu, Ile, and Glu.
The 29th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Phe, Leu, Ile, Val, and Ala.
The 73rd amino acid in the heavy chain FR3 based on the Kabat definition is Lys.
The 76th amino acid of the heavy chain FR3 based on the Kabat definition is Ser.

(2) The antibody or fragment thereof described in (1), wherein the 27th amino acid of the heavy chain FR1 is Tyr and the 29th amino acid of the heavy chain FR1 is Phe.

(3) The antibody or fragment thereof according to (1) or (2), wherein the amino acid sequences of CDR1 to CDR3 of the heavy and light chains further satisfy all of the following [Requirement B]:
[Requirement B]
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:4.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:6.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:7.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:8.

(4) The antibody or fragment thereof according to any one of (1) to (3), which further satisfies all of the following [Requirement C]:
[Requirement C]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:1.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:5.

(5) The antibody or fragment thereof according to any one of (1) to (3), which further satisfies all of the following [Requirement D]:
[Requirement D]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO: 17 or 18.
The light chain variable region has the amino acid sequence shown in SEQ ID NO: 19 or 20.

(6) An antibody or fragment thereof that specifically binds to the "FluA-20 antibody," an anti-influenza HA broadly neutralizing antibody, in which the amino acid sequences of CDR1 to CDR3 of the heavy and light chains satisfy all of the following [Requirement E]:
[Requirement E]
The heavy chain CDR1 has the amino acid sequence shown in SEQ ID NO:10.
The heavy chain CDR2 has the amino acid sequence shown in SEQ ID NO:11.
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:12.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:14.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:15.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:16.

(7) The antibody or fragment thereof according to (6), which further satisfies all of the following [Requirement F]:
[Requirement F]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:9.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:13.

(8) An anti-influenza vaccine comprising the antibody or fragment thereof described in any one of (1) to (7), or a nucleic acid encoding the antibody or fragment thereof.

The present invention provides anti-idiotype antibodies against broadly neutralizing antibodies, and antigens that can elicit reactivity against a wide range of influenza virus subtypes using such anti-idiotype antibodies.

FIG. 1 shows the reactivity of serum from mice immunized with "#2-911" to HA in an example. FIG. 1 is a conceptual diagram showing the relationship between a template antibody and an anti-Id antibody induced from the template antibody, and the mechanism by which the anti-Id antibody induces an antibody similar to the template antibody. FIG. 1 shows the competitiveness between the serum of a mouse immunized with "#2-911" and "MEDI8852 antibody" in an example. FIG. 1 shows the reactivity of serum from mice immunized with "#2-911 Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from mice immunized with "#2-911 Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from mice immunized with "#2-911" in an example. FIG. 1 shows the reactivity of serum from rats immunized with "#2-911 Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from rabbits immunized with "#2-911 Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from rats immunized with "#2-911 Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from rabbits immunized with "#2-911 Fab nanoparticles" to HA in an example. This is a diagram showing the reactivity of "Hu #2-911 mouse chimera" to "MEDI8852 antibody" in an example. FIG. 1 shows the reactivity of serum from mice immunized with "R2-150Fab nanoparticles" to HA in an example. FIG. 1 shows the reactivity of serum from mice immunized with "R2-150Fab nanoparticles" to HA in an example. FIG. 1 shows the results of analyzing the interaction between "#2-911" and "MEDI8852 antibody." FIG. 1 shows the results of analyzing the interaction between "#2-911" and "MEDI8852 antibody." FIG. 1 shows the results of analyzing the interaction between "#2-911" and "MEDI8852 antibody." FIG. 1 shows the binding of alanine-substituted mutants of “#2-911” to “MEDI8852 antibody.” FIG. 1 shows the reactivity of monoclonal antibodies obtained from mice administered with "#2-911" as an antigen against various HAs.

The following describes an embodiment of the present invention, but the present invention is not limited to this.

<Antibody of the present invention or fragment thereof>
The antibody or fragment thereof of the present invention specifically binds to a predetermined anti-influenza HA broadly neutralizing antibody, namely, the "MEDI8852 antibody" or "FluA-20 antibody" described below.
The antibody or fragment thereof of the present invention corresponds to the anti-idiotype antibody of the "MEDI8852 antibody" or "FluA-20 antibody."

Influenza viruses are classified into several types (types A, B, etc.) based on differences in the antigenicity of their surface proteins.
Furthermore, 18 subtypes of influenza A are known due to sequence polymorphism of hemagglutinin (HA).
Such polymorphisms make it difficult to generate antibodies that are broadly cross-reactive with influenza virus proteins, and hinder the development of vaccines against influenza virus infections.

In light of these circumstances, the inventors focused on developing anti-idiotype antibodies against the known broadly neutralizing anti-influenza HA antibodies "MEDI8852 antibody" and "FluA-20 antibody." "Anti-idiotype antibody" refers to an antibody that specifically binds to the idiotope formed in the antigen-binding region of a specific antibody molecule (in this invention, "MEDI8852 antibody" or "FluA-20 antibody").

Furthermore, the present inventors believed that the following results could be expected by inoculating any subject with the anti-idiotype antibody of "MEDI8852 antibody" or "FluA-20 antibody" as an antigen.
- Antibodies and the like having the same or similar functions or cross-reactivity as the target antibody (in this invention, "MEDI8852 antibody" or "FluA-20 antibody") can be produced in the body of the inoculated subject.
- Antibodies with expanded functions (especially broader cross-reactivity) than the target antibody can be produced in the body of the vaccinated subject.
- Antibodies reactive to a wide range of influenza HA subtypes can be produced in the body of the vaccinated person.
Furthermore, the antibodies etc. produced above can be cloned.

The inventors therefore conducted extensive research and discovered specific anti-idiotype antibodies (the "MEDI8852 antibody-anti-Id antibody" and "FluA-20 antibody-anti-Id antibody" described below), leading to the completion of the present invention.

The inventors confirmed that the serum obtained by inoculating a living body with "MEDI8852 antibody-anti-Id antibody" and "FluA-20 antibody-anti-Id antibody" reacts with a wide range of influenza A subtypes.

Furthermore, the present inventors unexpectedly confirmed that the serum obtained by inoculating a living body with "MEDI8852 antibody-anti-Id antibody" and "FluA-20 antibody-anti-Id antibody" also reacts with influenza B virus HA.
Neither the "MEDI8852 antibody" nor the "FluA-20 antibody" binds to influenza virus type B. Therefore, it was a highly unexpected result to obtain an anti-idiotype antibody that induces the production of an antibody that binds to an antigen that is not bound by the template antibodies (the "MEDI8852 antibody" and the "FluA-20 antibody").
Therefore, the antibody or fragment thereof of the present invention functions as an antigen capable of inducing the production of broadly neutralizing antibody-like antibodies (or sera containing such antibodies) that bind to a wide range of known subtypes of influenza A and B viruses.

In the present invention, "influenza virus" includes any influenza virus, and particularly includes influenza A virus and influenza B virus.

By administering the antibody or fragment thereof of the present invention, antibodies or serum reactive to many subtypes of influenza A virus and influenza B virus can be produced.

In a preferred embodiment of the present invention, the antibody or fragment thereof of the present invention comprises:
- A substance capable of inducing antibodies (or serum) reactive to influenza A virus HA subtypes (preferably H1 to H8, H10 to H18).
- A substance capable of inducing antibodies (or serum) reactive to influenza B virus (particularly Victoria strain and Yamagata strain) HA.
- A substance that can induce antibodies (or serum) reactive to all of the above HAs.

In the present invention, "antibody" includes proteins with a Y-shaped four-arm structure, also known as immunoglobulins.

In the present invention, the term "antibody fragment" includes proteins or peptides having the structure of the antibody complementarity determining region (hereinafter also referred to as "CDR") or the structure of the variable region of the antibody of the present invention.
Examples of such proteins or peptides include Fab, Fab', F(ab') ₂ , Fv (variable fragment of antibody), single-chain antibodies including scFv (H chain, L chain, H chain V region, L chain V region, etc.), diabodies (scFv dimers), dsFv (disulfide-stabilized V region), and VHH (variable domain of heavy chain of heavy chain antibody), as well as proteins or peptides comprising at least a portion of CDR.

"Fab" is an antibody fragment obtained by treating an antibody molecule with the protease papain, and has a molecular weight of approximately 50,000 and antigen-binding activity, in which approximately the N-terminal half of the H chain and the entire L chain are linked by a disulfide bond.
The method for producing Fab is not particularly limited, but examples thereof include a method in which DNA encoding the Fab of an antibody is inserted into an expression vector for prokaryotes or eukaryotes, and the vector is introduced into a prokaryote or eukaryote to express the Fab.

"F(ab') ₂ " is an antibody fragment with antigen-binding activity and a molecular weight of approximately 100,000, which is slightly larger than Fab fragments bound via disulfide bonds in the hinge region, and is obtained by treating an antibody molecule with the protease pepsin.
The method for producing F(ab') ₂ is not particularly limited, but examples include a method in which Fab is produced by bonding with a thioether bond or a disulfide bond.

"Fab'" is an antibody fragment with a molecular weight of approximately 50,000 that has antigen-binding activity and is obtained by cleaving the disulfide bond in the hinge region of "F(ab') ₂ ".
The method for producing Fab' is not particularly limited, but examples thereof include inserting DNA encoding the Fab' fragment of an antibody into an expression vector for prokaryotes or eukaryotes, and then introducing the vector into a prokaryote or eukaryote to express Fab'.

"Fv" is the minimum unit of an antibody fragment having antigen-binding activity, and is composed of one heavy chain variable region (VH) and one light chain variable region (VL).
Methods for producing Fv include, but are not limited to, constructing cDNA encoding VH and VL, inserting the DNA into a prokaryotic or eukaryotic expression vector, and introducing the expression vector into a prokaryote or eukaryote to express the Fv. However, since it is generally difficult to obtain Fv, scFv, which will be described later, is widely used.

An "scFv" is an antibody fragment that has antigen-binding activity and is a VH-P-VL or VL-P-VH polypeptide in which one heavy chain variable region (VH) and one light chain variable region (VL) are linked using an appropriate peptide linker (P).
Methods for producing scFv are not particularly limited, but include methods in which cDNA encoding the VH and VL of an antibody is obtained, DNA encoding the scFv is constructed, the DNA is inserted into an expression vector for prokaryotes or eukaryotes, and the expression vector is introduced into a prokaryote or eukaryote to express the scFv.

A "diabody" is an antibody fragment formed by dimerization of scFvs and has bivalent antigen-binding activity. The bivalent antigen-binding activities may be the same, or one of the two may be a different antigen-binding activity.
Methods for producing diabodies are not particularly limited, but include obtaining cDNA encoding the VH and VL of an antibody, constructing DNA encoding an scFv such that the amino acid sequence of the peptide linker (P) is 8 residues or less in length, inserting the DNA into an expression vector for prokaryotes or eukaryotes, and introducing the expression vector into a prokaryote or eukaryote to express the diabody.

A "dsFv" is a polypeptide in which one amino acid residue in each of VH and VL is substituted with a cysteine residue, and the cysteine residues are linked via a disulfide bond. The amino acid residue to be substituted with a cysteine residue can be selected based on the prediction of the three-dimensional structure of an antibody according to the method described by Reiter et al. (Protein Engineering, 7, 697-704, 1994).
Methods for producing dsFv are not particularly limited, but include methods in which cDNA encoding the VH and VL of an antibody is obtained, DNA encoding a dsFv is constructed, the DNA is inserted into an expression vector for prokaryotes or eukaryotes, and the expression vector is introduced into a prokaryote or eukaryote to express the dsFv.

"VHH" refers to a protein or peptide that consists of only an antibody heavy chain region and is capable of specifically binding to an antigen.
Methods for producing VHHs are not particularly limited, but include constructing DNA encoding the VHH, inserting the DNA into an expression vector for prokaryotes or eukaryotes, and introducing the expression vector into a prokaryote or eukaryote to express the VHH.

A CDR-containing peptide is composed of at least two regions of the three heavy chain CDRs (referred to as heavy chain CDR1 to CDR3 or HCDR1 to HCDR3) and the three light chain CDRs (referred to as light chain CDR1 to CDR3 or LCDR1 to LCDR3).
In a preferred embodiment of the present invention, the CDR-containing peptide comprises two CDRs of a heavy chain (heavy chain CDR1 and CDR2).

In one aspect of the present invention, the antibody of the present invention may have specific amino acids not only in the heavy chain CDRs and/or light chain CDRs, but also in locations outside the CDRs of the heavy chain variable region and/or light chain variable region.

Hereinafter, when a number relating to the position of a specific amino acid is mentioned (for example, "the 27th amino acid in the heavy chain FR1 is one amino acid selected from the group consisting of Tyr, Phe, Leu, Ile, and Glu"), the number means the number defined by Kabat.
For the Kabat definition, see "Sequences of Proteins of Immunological Interest, Fifth Edition, NIH Publication No. 91-3242, U.S. Department of Health and Human Services, 1991." According to the Kabat definition, particularly for amino acids constituting a framework region, amino acids appearing at the same position in the three-dimensional structure of an antibody molecule can be uniquely designated.

In the case of "#2-911" (an example of an antibody that satisfies requirements A, B, and C described below; see Table 6), the above-mentioned specific amino acids present in locations other than the CDRs of the heavy chain variable region and/or light chain variable region include the amino acids Tyr27 (tyrosine at position 27) and Phe29 (phenylalanine at position 29) located in the N-terminal framework of heavy chain CDR1 (also referred to as FR1), and further include the amino acids Lys73 (lysine at position 73) and Ser76 (serine at position 76) located in the framework between heavy chain CDR2 and heavy chain CDR3 of "#2-911" (also referred to as FR3).
In addition, even if Tyr27 is substituted with Phe (phenylalanine), Leu (leucine), Ile (isoleucine), or Glu (glutamic acid) instead of tyrosine, the antibody function is not lost.
The above Phe29 may be substituted with Leu (leucine), Ile (isoleucine), Ala (alanine), or Val (valine) instead of phenylalanine without impairing the function of the antibody.

In one aspect of the present invention, the antibody of the present invention may have a region containing specific amino acids instead of one or more heavy chain CDRs defined in requirement A below.
In the case of "#2-911", such amino acids include the following:
His35 (the 35th histidine) contained in the heavy chain CDR1 sequence region, and Phe53 (the 53rd phenylalanine) contained in the heavy chain CDR2 sequence region
- Amino acids Tyr27 (27th tyrosine) and Phe29 (29th phenylalanine) on the N-terminal framework of heavy chain CDR1
- Amino acids Lys73 (73rd lysine) and Ser76 (76th serine) on the framework (also called FR3) located between heavy chain CDR2 and heavy chain CDR3.
Furthermore, even if Tyr27 is substituted with Phe, Leu, Ile, or Glu instead of tyrosine, the antibody function is not lost.
Furthermore, even if Phe29 is substituted with Leu, Ile, Ala, or Val instead of phenylalanine, the antibody function is not lost.

The antibodies or fragments thereof of the present invention are described in detail below.

(1) Antibodies or fragments thereof that specifically bind to the "MEDI8852 antibody" The present invention encompasses antibodies or fragments thereof that specifically bind to the "MEDI8852 antibody" (hereinafter, these antibodies or fragments thereof are also collectively referred to as "MEDI8852 antibody-anti-Id antibodies").

The "MEDI8852 antibody" is a broadly neutralizing antibody whose epitope is the stalk region of the HA of influenza A virus, as reported in WO2017/123685 and Kallewaard et al. (2016), Cell, 166, 596-608.

In the present invention, "specifically binds to MEDI8852 antibody" includes the case where the antibody of the present invention or a fragment thereof binds to "MEDI8852 antibody" but hardly binds or does not bind at all to other proteins (antibodies, etc.).
Whether an antibody or a fragment thereof specifically binds to the "MEDI8852 antibody" can be determined by a method such as enzyme-linked immunosorbent assay (ELISA).

(1-1) Aspect A
In one embodiment of the present invention, the "MEDI8852 antibody-anti-Id antibody" satisfies all of the following requirement A:
[Requirement A]
- Heavy chain CDR1 is the amino acid sequence shown in SEQ ID NO:2.
- Heavy chain CDR2 is the amino acid sequence shown in SEQ ID NO:3.
The 27th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Tyr, Phe, Leu, Ile, and Glu.
The 29th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Phe, Leu, Ile, Val, and Ala.
The 73rd amino acid in the heavy chain FR3 based on the Kabat definition is Lys.
The 76th amino acid of the heavy chain FR3 based on the Kabat definition is Ser.

In the above embodiment, heavy chain CDR1 and CDR2, and four amino acids in the FRs (i.e., the 27th amino acid of heavy chain FR1, the 29th amino acid of heavy chain FR1, the 73rd amino acid of heavy chain FR3, and the 76th amino acid of heavy chain FR3 based on the above-mentioned Kabat definition) directly interact with the "MEDI8852 antibody" and contribute to the binding between the "MEDI8852 antibody-anti-Id antibody" and the "MEDI8852 antibody."
The four amino acids in the FR are exposed to the "MEDI8852 antibody" during binding with the "MEDI8852 antibody," and are thereby involved in antibody-antibody binding.

From the viewpoint of making it easier to obtain antibodies equivalent to the antibodies that were observed to have good reactivity in the Examples described below, in a preferred embodiment of the present invention, the 27th amino acid in heavy chain FR1 based on the Kabat definition is Tyr, and the 29th amino acid in heavy chain FR1 based on the Kabat definition is Phe.

(1-2) Mode B
In one embodiment of the present invention, the amino acid sequences of CDR1 to CDR3 of the heavy and light chains of the "MEDI8852 antibody - anti-Id antibody" satisfy all of the following [Requirement B] in addition to the requirements of "Aspect A" above.
[Requirement B]
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:4.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:6.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:7.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:8.

However, the present invention also encompasses antibodies or fragments thereof that specifically bind to the "MEDI8852 antibody," an anti-influenza HA broadly neutralizing antibody, which has the two CDRs of [Requirement A].
For example, the present invention encompasses antibodies or fragments thereof that specifically bind to the "MEDI8852 antibody" in which heavy chain CDR1 has the amino acid sequence represented by SEQ ID NO: 2 and heavy chain CDR2 has the amino acid sequence represented by SEQ ID NO: 3.
For example, the present invention encompasses antibodies or fragments thereof that specifically bind to the "MEDI8852 antibody" in which heavy chain CDR1 has the amino acid sequence represented by SEQ ID NO: 2, heavy chain CDR2 has the amino acid sequence represented by SEQ ID NO: 3, and heavy chain CDR3 has the amino acid sequence represented by SEQ ID NO: 4.

From the viewpoint of more easily achieving the effects of the present invention, the "MEDI8852 antibody-anti-Id antibody" preferably satisfies all of the following [Requirement C].
[Requirement C]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:1.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:5.

From the viewpoint of more easily achieving the effects of the present invention, the "MEDI8852 antibody-anti-Id antibody" preferably satisfies all of the following [Requirement D].
[Requirement D]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO: 17 or 18.
The light chain variable region has the amino acid sequence shown in SEQ ID NO: 19 or 20.

The "MEDI8852 antibody-anti-Id antibody" satisfies at least requirement A.
Furthermore, the "MEDI8852 antibody-anti-Id antibody" preferably satisfies requirement B, and more preferably satisfies requirement C or requirement D.

The "MEDI8852 antibody-anti-Id antibody" according to a preferred embodiment of the present invention includes the antibodies referred to in the examples as "#2-911 (#2-911 antibody)" and "Hu #2-911" (a humanized version of "#2-911," humanized #2-911 antibody).

(2) Antibodies or fragments thereof that specifically bind to "FluA-20 antibody" The present invention encompasses anti-idiotype antibodies or fragments thereof that specifically bind to "FluA-20 antibody" (hereinafter, these antibodies or fragments thereof are also collectively referred to as "FluA-20 antibody-anti-Id antibodies").

The "FluA-20 antibody" is a broadly neutralizing antibody whose epitope is the trimer-forming interface of the head region of HA of influenza A virus (HA-head trimer interface), as reported in WO2020/232426 and Bangaru et al. (2019), Cell, 177, 1136-1152.

In the present invention, the phrase "specifically binds to FluA-20 antibody" includes the case where the antibody of the present invention or a fragment thereof binds to "FluA-20 antibody" but hardly binds or does not bind at all to other proteins (antibodies, etc.).
Whether an antibody or a fragment thereof specifically binds to the "FluA-20 antibody" can be determined by a method such as enzyme-linked immunosorbent assay (ELISA).

In one embodiment of the present invention, the amino acid sequences of CDR1 to CDR3 of the heavy and light chains of the "FluA-20 antibody-anti-Id antibody" satisfy all of the following [Requirement E]:
[Requirement E]
The heavy chain CDR1 has the amino acid sequence shown in SEQ ID NO:10.
The heavy chain CDR2 has the amino acid sequence shown in SEQ ID NO:11.
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:12.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:14.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:15.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:16.

From the viewpoint of more easily achieving the effects of the present invention, the "FluA-20 antibody-anti-Id antibody" preferably satisfies all of the following [Requirement F].
[Requirement F]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:9.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:13.

The "FluA-20 antibody-anti-Id antibody" according to a preferred embodiment of the present invention includes the antibody referred to in the examples as "R2-150 (R2-150 antibody)." The "FluA-20 antibody-anti-Id antibody" satisfies at least requirement E. Furthermore, the "FluA-20 antibody-anti-Id antibody" preferably satisfies requirement F.

<Method of producing the antibody or fragment thereof of the present invention>
The method for producing the antibody of the present invention or a fragment thereof is not particularly limited, and conventional methods for producing various antibodies (monoclonal antibodies, polyclonal antibodies, etc.) and various polypeptides can be employed.

The antibodies of the present invention are preferably produced as monoclonal antibodies or fusion proteins with other proteins (ferritin protein, hemocyanin, serum albumin, linker proteins, etc.).

The antibody or fragment thereof of the present invention may be purified, etc., as necessary.

Whether an antibody or fragment thereof is an antibody or fragment thereof of the present invention can be determined by any method for identifying an amino acid sequence (such as N-terminal amino acid sequence analysis).

The present invention includes genes encoding the antibodies of the present invention or fragments thereof, expression vectors containing the genes, and the like.
These genes and expression vectors can be prepared by conventionally known methods.

<Uses of the antibody or fragment thereof of the present invention>
The antibody or fragment thereof of the present invention functions as an anti-idiotype antibody against the "MEDI8852 antibody" and "FluA-20 antibody", which are broadly neutralizing anti-influenza HA antibodies.
Therefore, the antibody of the present invention or a fragment thereof can be used for any purpose that utilizes this function.

Anti-influenza vaccines are one example of uses for the antibodies or fragments thereof of the present invention. Therefore, the present invention encompasses anti-influenza vaccines comprising the antibodies or fragments thereof of the present invention.

The present invention also encompasses methods for immunizing against influenza viruses and methods for preventing influenza virus infections, which comprise the step of administering the antibody or fragment thereof of the present invention to any animal (human or non-human animal).

<Anti-influenza vaccine>
The present invention includes vaccines (anti-influenza vaccines) comprising the antibodies of the present invention or fragments thereof, or nucleic acids encoding them.

The antibody or fragment thereof of the present invention functions as an antigen and can produce antibodies or serum that are broadly cross-reactive with known influenza virus HAs of type A or B that infect any organism.
Therefore, the antibody or fragment thereof of the present invention, or a nucleic acid encoding the antibody or fragment thereof can be preferably used as a universal vaccine against influenza viruses.
In the present invention, the term "nucleic acid" includes DNA, mRNA, and the like.

The vaccine of the present invention may be effective in preventing, for example, human seasonal influenza virus infections, novel infectious diseases caused by influenza viruses that have newly acquired infectivity in humans, and influenza virus infections that infect non-humans.

The form of the vaccine of the present invention is not particularly limited, but it can be prepared as a protein vaccine (such as a vaccine containing an antibody of the present invention or a fragment thereof), a nucleic acid vaccine (such as a vaccine containing a nucleic acid encoding an antibody of the present invention or a fragment thereof), etc.

The subjects to which the vaccine of the present invention can be administered are not particularly limited, and examples include mammals (humans and non-human animals) and birds. Non-human animals include rats, mice, rabbits, monkeys, pigs, etc.

The dosage and frequency of administration of the vaccine of the present invention can be determined appropriately depending on the condition of the recipient (age, weight, symptoms, etc.).

The method of administration of the vaccine of the present invention can be selected appropriately depending on the form of the vaccine, etc., and includes, for example, subcutaneous administration, intranasal administration, transdermal administration, etc.

The vaccine of the present invention may contain the antibody or fragment thereof of the present invention, or a nucleic acid encoding the antibody or fragment thereof, as well as any pharmaceutically acceptable carrier or adjuvant, as needed.

The present invention will be explained in more detail below using examples, but the present invention is not limited to these examples.

In this example, the following abbreviations are used where appropriate:
Anti-Id antibody: anti-idiotype antibody CDR: complementarity determining region VH: heavy chain variable region HCDR: heavy chain CDR
VL: light chain variable region LCDR: light chain CDR
HA: hemagglutinin

The above CDRs are defined according to the method of Kabat et al. (Sequences of Proteins of Immunological Interest, Fifth Edition, NIH Publication No. 91-3242, U.S. Department of Health and Human Services, 1991).

<Test 1: Establishment of anti-Id antibody-producing cells against "MEDI8852 antibody">
Based on the following method, a hybridoma clone producing an anti-Id antibody against the "MEDI8852 antibody," a known broadly neutralizing anti-influenza HA antibody, was obtained.

(1) Preparation of "MEDI8852 antibody" Based on the amino acid sequence of "MEDI8852" (Kallewaard et al., Cell, 166, 596-608, 2016, and WO2017/123685), DNA was synthesized by adding a Kozak translation initiation sequence to the 5' end of each of the heavy chain variable region and light chain variable region.
The obtained heavy chain variable region was cloned into the "pFUSE-CHIg-hG1" vector (InvivoGen) to obtain the "MEDI8852 heavy chain expression plasmid."
The obtained light chain variable region was cloned into the "pFUSE2-CLIg-hk" vector (InVivogen) to obtain the "MEDI8852 light chain expression plasmid."
For cloning, the services of Azenta were used.

Next, transient expression was carried out using the "MEDI8852 heavy chain expression plasmid" and the "MEDI8852 light chain expression plasmid" as the introduced plasmids using the "Expi293 Expression System" (Thermo Fisher Scientific).
After transient expression, the culture supernatant was collected and affinity purified using a "HiTrap Protein A HP column" (Cytiva). The resulting eluate was concentrated and purified by gel filtration using a "HiLoad 26/600 Superdex 200 pg column" (Cytiva).
The eluted fractions from the gel filtration purification were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and subjected to the following tests as "MEDI8852 antibody."

(2) Preparation of MEDI8852 Mouse Fc Chimeric Antibody Based on the amino acid sequence of the heavy chain Fab constituent region of the "MEDI8852 antibody," DNA was synthesized by adding a Kozak translation initiation sequence to the 5' end.
The resulting DNA was cloned into the "pFUSE-CHIg-mG2a" vector (InvivoGen) to obtain the "MEDI8852 Fab-mouse Fc chimeric heavy chain expression plasmid."
For cloning, the services of Azenta were used.

Next, using the "MEDI8852 Fab-mouse Fc chimeric heavy chain expression plasmid" and the "MEDI8852 light chain expression plasmid" as the introduction plasmids, a MEDI8852 Fab-mouse Fc chimeric antibody (hereinafter also referred to as "MEDI8852 mouse Fc chimeric antibody") was prepared in the same manner as described above in "(1) Preparation of 'MEDI8852 antibody'".

(3) Preparation of Hybridoma Clones Producing Anti-Id Antibodies Against "MEDI8852 Antibody" Hybridoma clones producing anti-Id antibodies specific to "MEDI8852 antibody" were obtained according to the following method.

(3-1) Immunization of Mice The antigen used was "MEDI8852 mouse Fc chimeric antibody," which was mixed with the adjuvant "TiterMax Gold, cat. G-1" (TiterMax) and administered to the soles of BALB/c mice. The antigen dose was set at 50 μg per mouse.
Booster immunizations were performed 3, 10, and 12 days after administration, with the antigen dose for booster immunization set at 5 μg per mouse, and PBS buffer used instead of an adjuvant.

After the immunization (2 days after the final administration), popliteal lymph nodes were collected from the mice and a cell suspension was prepared.
The resulting cell suspension was mixed with SP2/0-Ag14 myeloma cells, and electrofusion was carried out using an electrofusion device "ECFG21" (Neppa Gene).
After fusion, the resulting cells were suspended in "ClonaCell-HY Medium D, cat. ST-03804" (STEMCELL Technologies) and seeded onto plastic dishes.
Colonies formed 10 days after seeding were isolated into 96-well plastic plates containing hybridoma medium, and the culture supernatant was used for the following evaluation. The hybridoma medium used was "RPMI1640, cat. A1049101" (Thermo Fisher Scientific) supplemented with 1/10 of the amount of "Doma-Drive, cat. T31-1003SF" (Immune Systems) and 1/50 of the amount of "HAT Supplement, cat. 21-60017" (Thermo Fisher Scientific).

(3-2) Selection of clones producing anti-Id antibodies specific to the "MEDI8852 antibody" Using each culture supernatant obtained in the above "(3-1) Immunization of mice," hybridoma clones producing anti-Id antibodies specific to the "MEDI8852 antibody" were screened.
Specifically, clones producing antibodies that meet all of the following criteria were selected by ELISA.
(Criterion 1) Reacts with (binds to) "MEDI8852 antibody."
(Criterion 2) The antibody does not react with the control antibody "human IgG1 kappa-UNLB, cat. 0151K-01" (Southern Biotech) which has the same isotype as the "MEDI8852 antibody."

The ELISA was carried out in an antigen solid-phase format in which "MEDI8852 antibody" or "human IgG1-kappa-UNLB" was immobilized.
The secondary antibody used for detection was "anti-mouse IgG-Fc fragment, HRP conjugated, cat. A90-131" (Bethyl Laboratories).
For color development, "ELISA POD Substrate TMB Kit (popular), cat. 05298-80" (Nacalai Tesque) was used, and absorbance at a wavelength of 450 nm was measured using "Infinite M1000 pro microplate reader" (Tecan).

<Test 2: Selection of anti-Id antibodies that induce antibodies reactive to HA>
An anti-Id antibody against the "MEDI8852 antibody" was obtained based on the following method. This antibody can induce antibodies that react with HA when inoculated as an antigen.

(1) Sequence Analysis of Anti-Id Antibody Hybridoma clones were expanded and cultured, and RNA was extracted from the resulting culture using "RNeasy mini Kit, cat. 74104" (QIAGEN).
Next, cDNA synthesis was carried out using "SuperScript III First-Strand Synthesis Supermix, cat. 18080400" (Thermo Fisher Scientific).

The synthesized cDNA was used as a template to amplify antibody variable region sequences (heavy chain variable region and light chain variable region) by PCR. Both the heavy chain variable region and the light chain variable region were amplified using primers that recognize the upstream of the variable region and the downstream of the constant region.
The obtained DNA fragment was cloned using "Zero Blunt TOPO PCR Cloning Kit, cat. 450159" (Thermo Fisher Scientific), and the DNA sequence of the antibody sequence variable region was analyzed.
For the analysis, the services of Azenta were used.

The antibody sequence identified as a result of DNA sequence analysis of the anti-Id antibody and its CDR region sequence are shown in Table 6. The anti-Id antibody having these sequences is also referred to as "#2-911" hereinafter.
In addition, in the VH sequences shown in Table 6, the 27th amino acid of the heavy chain framework is Tyr, the 29th amino acid is Phe, the 73rd amino acid is Lys, and the 76th amino acid is Ser, based on the Kabat definition.

(2) Preparation of Recombinant Anti-Id Antibody Based on the sequence of the variable region of the anti-Id antibody (#2-911), DNA was synthesized by adding a Kozak translation initiation sequence to the 5' end of each of the heavy chain variable region and light chain variable region.
The obtained heavy chain variable region was cloned into the "pFUSE-CHIg-mG2a" vector (InvivoGen) to obtain an "anti-Id antibody heavy chain expression plasmid."
The obtained light chain variable region was cloned into the "pFUSE2-CLIg-ml1" vector (InvivoGen) to obtain an "anti-Id antibody light chain expression plasmid."
For cloning, the services of Azenta were used.

Next, transient expression was carried out using the "anti-Id antibody heavy chain expression plasmid" and the "anti-Id antibody light chain expression plasmid" as the introduced plasmids using the "Expi293 Expression System" (Thermo Fisher Scientific).
After transient expression, the culture supernatant was collected and affinity purified using a "HiTrap Protein G HP column" (Cytiva). The resulting eluate was concentrated and purified by gel filtration using a "HiLoad 26/600 Superdex 200 pg column" (Cytiva).
The eluted fractions from the gel filtration purification were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and subjected to the following tests as recombinant antibodies (hereinafter also referred to as "recombinant anti-Id antibodies").

(3) Verification of reactivity to HA Based on the following method, the "recombinant anti-Id antibody"(#2-911) was verified from the viewpoint of whether it functions as an antigen capable of inducing antibodies that react with HA in mouse serum.

(3-1) Immunization of Mice The antigen used was a recombinant anti-Id antibody, which was mixed with the adjuvant TiterMax Gold (TiterMax) and intraperitoneally administered to BALB/c mice (3 mice per group). The antigen dose was set at 200 μg per mouse.
After administration, booster immunizations were performed twice at two-week intervals from the start of administration. The antigen dose for booster immunization was set at 20 μg per mouse, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant.

(3-2) Verification of reactivity with HA One week after the first booster immunization, partial blood samples were taken from the mice immunized with the "recombinant anti-Id antibody," and the production of antibodies against the anti-Id antibody was confirmed by ELISA.
Therefore, this anti-Id antibody was found to function as an antigen.
Therefore, one week after the second booster immunization, whole blood was collected from the mice, and serum was prepared (immune serum), which was then used to verify reactivity with HA.
As a control, non-immunized mouse serum (naive serum) was also prepared.

The reactivity with HA was verified by ELISA using an antigen solid-phase format in which an antigen panel consisting of various HAs was immobilized. These HAs correspond to H1 to H18, which represent the antigenicity of circulating influenza A viruses.
As a "positive control," to compare the immune induction ability of anti-Id antibodies, serum from a mouse immunized with Ab2 antibody, an anti-Id antibody against an antibody specific to A/BK79 (H3N2) HA, as described in a previous paper (Betakova et al. J. Gen. Virol. 1998), was reacted with immobilized A/BK79 HA.
Each serum was diluted 300-fold with PBS buffer containing 1% BSA (hereinafter also referred to as "1% BSA/PBS") and used as a sample.
As the secondary antibody for detection, "anti-mouse IgG-Fc fragment, HRP conjugated" (Bethyl Laboratories) was used.
As a color-developing reagent, "ELISA POD Substrate TMB Kit" (popular) (Nacalai Tesque) was used.
After the ELISA reaction, the absorbance at a wavelength of 450 nm was measured using an "Infinite M1000 pro microplate reader" (Tecan). ELISA was performed in duplicate for each point, and the average value was calculated.

The results based on the ELISA are shown in Figure 1. This figure shows the reactivity of naive or immune sera with a panel of HA antigens.
These results demonstrate that the anti-Id antibody (#2-911) functions as an antigen capable of inducing antibodies against all known influenza A virus HA subtypes (except H9).

(4) Verification of binding competition between "#2-911" and "MEDI8852 antibody" From the results of (3) above, it was found that "#2-911", obtained using "MEDI8852 antibody" as a template, is an anti-Id antibody of "MEDI8852 antibody" and further functions as an antigen that induces antibodies that react with HA.
When mice are immunized with "#2-911" as an antigen, the resulting serum may contain antibodies similar to the "MEDI8852 antibody" (Figure 2). In such cases, such antibodies similar to the "MEDI8852 antibody" should compete with the "MEDI8852 antibody" for binding to HA.
Therefore, the competition between the serum of mice inoculated with "#2-911" and "MEDI8852 antibody" was examined by competitive ELISA.

As the HA antigen in the competitive ELISA, one of the following was used:
NC99/H1N1: Recombinant ectodomain protein of human influenza A/New Caledonia/20/99 (H1N1) Vic361/H3N2: Recombinant ectodomain protein of A/Victoria/361/2011 (H3N2)

Each serum (diluted 100-fold, 300-fold, or 900-fold) was reacted with the antigen-immobilized wells at room temperature for 0.5 hours, and then washed with PBS buffer containing 0.05% Tween 20 (hereinafter also referred to as "PBST").
Next, the wells containing the immobilized NC99/H1N1 antibody were reacted with MEDI8852 antibody (more than three times the EC50 of 13.5 ng/ml for NC99/H1N1) prepared at 45 ng/ml using 1% BSA/PBS for 1 hour at room temperature, and then washed with PBST.
On the other hand, the wells containing Vic361/H3N2 were reacted with MEDI8852 antibody (90 ng/ml) prepared using 1% BSA/PBS (more than three times the EC50 of 26.7 ng/ml for Vic361/H3N2) at room temperature for 1 hour, followed by washing with PBST.
As a control, 20 μg/ml of mouse isotype control antibody "mouse IgG2a isotype control (clone C1.18.4), cat. I-118" (Leinco Technologies) was used instead of serum.

After the above reaction, the MEDI8852 antibody bound to HA was detected using anti-human IgG-Fc fragment, HRP-conjugated, cat. A80-104P (Bethyl Laboratories) as the secondary antibody. ELISA was performed in duplicate for each point, and the average value was calculated.

FIG. 3 shows the results based on the above ELISA.
As shown in FIG. 3, the binding of "MEDI8852 antibody" to HA decreased in a concentration-dependent manner with the serum from mice inoculated with "#2-911."
This indicates that the mouse serum obtained after inoculation with "#2-911" contains antibodies that compete with the "MEDI8852 antibody" in binding to HA, i.e., it contains antibodies similar to the "MEDI8852 antibody."

Furthermore, analysis of the complex formed by binding of the above-mentioned "#2-911" and MEDI8852 suggested that the heavy chain CDR1 (amino acid sequence represented by SEQ ID NO: 2) and heavy chain CDR2 (amino acid sequence represented by SEQ ID NO: 3) of "#2-911" were directly involved in the binding.

<Test 3: Evaluation of anti-Id antibody (#2-911) against "MEDI8852 antibody"-1>
As shown in Test 2, it was found that the mouse serum obtained by inoculation with "#2-911" competes with "MEDI8852 antibody" in binding to HA, i.e., antibodies similar to "MEDI8852 antibody" were induced.
Therefore, we prepared nanoparticles (hereinafter also referred to as "#2-911Fab nanoparticles") using a self-associating protein designed to enable multivalent antigen presentation of the Fab region of "#2-911." Nanoparticles using self-associating proteins are known as one form of vaccine antigen.
Furthermore, serum from animals immunized with the nanoparticles as an antigen was obtained and its binding to HA molecules was evaluated.

(1) Preparation of #2-911 Fab Nanoparticles A DNA encoding a fusion protein sequence in which a ferritin protein sequence was bound was ligated downstream of the sequence of the heavy chain Fab structural region of the "#2-911" expression plasmid via a linker sequence containing repeats (3, 5, or 7 times) of a sequence consisting of the four amino acids serine-glycine-glycine-glycine.
This fusion protein was designed according to Cell, 162, 1090-1100, 2015, and has a Kozak translation initiation sequence added to the 5' side and a stop codon added to the 3' side.
The obtained gene was cloned into the "pcDNA3.4" vector (Thermo Fisher Scientific) to obtain an expression plasmid (heavy chain) for "#2-911Fab-nanoparticles."

In addition, the light chain Fab component region of the "#2-911" expression plasmid was also cloned into the "pcDNA3.4" vector (Thermo Fisher Scientific) to obtain the "#2-911Fab-nanoparticle" expression plasmid (light chain).

Azenta's services were used for the cloning.

(2) Preparation of "#2-911" Immune Mouse Serum The resulting "#2-911Fab-nanoparticle" expression plasmids (heavy chain and light chain) were introduced into the "Expi293 Expression System" (Thermo Fisher Scientific) to perform transient expression.
After transient expression, the culture supernatant was collected and subjected to affinity purification using "CaptureSelect LC-lambda (mouse) affinity matrix, cat. 194323010" (Thermo Fischer Scientific).
The resulting eluate was concentrated and purified by gel filtration using a "HiPrep 26/60 Sephacryl S-500HR column" (Cytiva).
The eluted fractions from the gel filtration purification were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and used hereinafter as "#2-911 Fab nanoparticles."

(3) Immunization of Mice The antigen used was "#2-911 Fab nanoparticles," which were mixed with the adjuvant "TiterMax Gold" (TiterMax), and intraperitoneally administered to BALB/c mice (6 mice per group). The antigen dose was set at 200 μg per mouse.
After administration, booster immunizations were performed twice at two-week intervals from the start of administration. The antigen dose for booster immunization was set at 20 μg per mouse, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant.
One week after the final immunization, whole blood was collected and serum was prepared and used for the following evaluation.

(4) Verification of reactivity with HA Based on the same method as in "(3-2) Verification of reactivity with HA" in Test 2, the reactivity of antibodies contained in serum obtained using "#2-911 Fab nanoparticles" as an antigen with a panel of HA antigens was evaluated by ELISA using a solid-phase antigen format.

The results based on the ELISA are shown in Figure 4. This figure shows the reactivity of naive or immune sera with a panel of HA antigens.
These results confirmed that mouse serum obtained using "#2-911 Fab nanoparticles" as an antigen was capable of inducing antibodies that exhibited broad cross-reactivity to various HAs, similar to the anti-Id antibody (#2-911).

Furthermore, it was confirmed that the #2-911 Fab nanoparticles induced a more stable immune response than immunization using the #2-911 antibody as the antigen, and that the serum of mice immunized with this antigen showed stronger reactivity to HAs from H4 to H15.

Furthermore, it was confirmed that the antibodies contained in the immune serum reacted with both the HA of influenza B Victoria and Yamagata strains, which the template antibody for #2-911, MEDI8852 antibody, could not react with (Figure 5).

(5) Evaluation of binding to HA molecules expressed on the cell surface Using serum obtained using "#2-911 Fab nanoparticles" as an antigen, binding to HA molecules expressed on cells was evaluated according to the following method.

(5-1) Preparation of HA-Expressing Cells The full-length HA gene of "A/Indonesia/5/2005 (H5N1)" (hereinafter also referred to as "Indo05/H5N1") containing a signal sequence was cloned into the "pCMVbeta" vector (Takara Bio Inc.).
The obtained full-length HA expression plasmid was used as a transfection plasmid and transiently expressed in Expi293 cells using the "Expi293 Expression System" (Thermo Fisher Scientific).
Hereinafter, the obtained cells are also referred to as "HA-expressing cells."

(5-2) Verification by Flow Cytometry 100,000 HA-expressing cells and 100,000 Expi 293 cells transfected with a plasmid not containing the HA gene (hereinafter also referred to as "mock cells") were prepared.
These cells were reacted at 4°C in the presence of any of the following, washed with PBS, and then stained with the detection antibody "BD Pharmingen PE goat anti-mouse Ig (multiple adsorption), cat. 550589" (BD Biosciences).
- Immune serum for "#2-911 Fab nanoparticles" - Negative control: mouse IgG2a (mIg2a) isotype control (clone C1.18.4) (Leinco Technologies)
Positive control: MEDI8852 mouse Fc chimera

After staining, flow cytometry analysis was performed using "FACS Canto II" (BD Biosciences).
FIG. 6 shows the histogram obtained by flow cytometry.
In the negative control group, no change was observed between the HA-expressing cells and the mock cells.
In the positive control, a peak shift was observed in HA-expressing cells compared to mock cells, confirming that HA molecules that react with MEDI8852 mouse Fc chimera are expressed on the cell surface.
On the other hand, a reaction to HA-expressing cells was also observed in the immune serum group of "#2-911 Fab nanoparticles." This confirmed that the immune serum contained antibodies that reacted with HA expressed on the cell surface.

<Test 4: Evaluation of anti-Id antibody (#2-911) against "MEDI8852 antibody"-2>
In the above-mentioned Test 3, it was confirmed that when mice were immunized with nanoparticles containing the Fab region of "#2-911" in a multivalent form, antibodies reactive to HA were contained.
Therefore, we examined whether antibodies that react to HA can be similarly induced in animals other than mice.

(1) Immunization of Animals Rats and rabbits were used instead of mice and were immunized as follows.

(1-1) Immunization of rats: "#2-911 Fab nanoparticles" were used as an antigen, mixed with the adjuvant "TiterMax Gold" (TiterMax), and intraperitoneally administered to Jcl:Wistar rats (3 rats per group). The antigen dose was set at 400 μg per rat.
After administration, booster immunizations were performed twice at two-week intervals from the start of administration. The antigen dose for booster immunization was set at 40 μg per rat, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant.
One week after the final immunization, whole blood was collected and serum was prepared, which was used for the following verification.

(1-2) Immunization of rabbits: "#2-911Fab nanoparticles" were used as an antigen, mixed with the adjuvant "TiterMax Gold" (TiterMax), and intradermally inoculated into Scl:NZW rabbits (2 rabbits per group). The antigen dose was set at 300 μg per rabbit.
After administration, booster immunizations were administered twice at two-week intervals. The antigen dose for booster immunizations was set at 300 μg per rabbit, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant. Japan Bioserum Corporation provided services for immunization and blood collection.
One week after the final immunization, whole blood was collected and serum was prepared, which was used for the following verification.

(2) Verification of reactivity with HA Verification of reactivity with HA was performed by ELISA using an antigen solid-phase format in which an antigen panel consisting of various HAs was immobilized, as in Test 2, "(3-2) Verification of reactivity with HA."

The results based on the above ELISA are shown in Figures 7 and 8. Figure 7 shows the results using rats, and Figure 8 shows the results using rabbits.
The results below confirm that "#2-911" is capable of inducing antibodies in immune serum that react broadly against various subtypes of influenza A not only in mice but also in rats and rabbits.

Furthermore, it was confirmed that the antibodies contained in each immune serum reacted with both the HA of influenza B Victoria strain and the HA of Yamagata strain, which cannot be reacted with the template antibody for "#2-911,""MEDI8852antibody" (Figures 9 and 10).
These results indicate that by using "#2-911" as an antigen, for example, by immunizing an animal to obtain an antibody, it is possible to obtain a new antibody that is similar in function and morphology to the template antibody (MEDI8852 antibody), or that has functions that exceed those of the template antibody.

<Test 5: Preparation of humanized antibodies>
Based on the following method, an antibody (hereinafter referred to as a humanized antibody) was obtained that has the CDR sequences of "#2-911" and has other sequences converted to human antibody sequences. Humanized antibodies can be a form of more effective vaccine antigen. Therefore, the humanized antibody was tested to determine whether it exhibited reactivity equivalent to that of the original antibody "#2-911."

(1) Design of humanized sequence of "#2-911" The humanized sequence was designed using the CDR grafting method with reference to the method of Kuramochi et al. (Human monoclonal antibodies, Methods and protocols, Methods in Molecular Biology 1904, ISBN 978:1-4939-8957-7, Chapter 9).

First, known structures of mouse antibodies with high amino acid sequence homology to "#2-911" were searched for, and the positions of amino acids thought to be important for CDR structure formation and reaction with antigen were predicted.
In parallel with this estimation work, cDNA databases for human antibody heavy chain variable regions and human antibody light chain variable regions were searched for sequences highly homologous to the framework regions of the heavy chain variable region and light chain variable region of "#2-911."

A sequence was designed by linking the framework sequence of the human antibody sequence identified as a result of the search with the CDR sequence of "#2-911".
Amino acid sequences at positions thought to be important for CDR structure formation and antigen reactivity were grafted into this sequence to design the humanized antibody sequence "Hu #2-911" shown in Table 7. The CDR sequences in this humanized antibody sequence are identical to those in the original mouse antibody. Furthermore, in this humanized antibody sequence, as in the original mouse antibody sequence, the 27th amino acid in the heavy chain framework is Tyr, the 29th amino acid is Phe, the 73rd amino acid is Lys, and the 76th amino acid is Ser, based on the Kabat definition.

(2) Evaluation of the antigenic reactivity of "Hu #2-911" Whether "Hu #2-911" has the same antigenic reactivity as "#2-911" was evaluated based on the following method.

(2-1) Preparation of "Hu #2-911" Mouse Chimera DNAs encoding the heavy and light chain variable regions of "Hu #2-911" were synthesized by adding a Kozak translation initiation sequence to the 5' end.
The obtained heavy chain variable region was cloned into the "pFUSE-CHIg-mG2a" vector (InvivoGen).
The obtained light chain variable region was cloned into the "pFUSE2-CLIg-ml1" vector (InvivoGen). For the cloning, the services of Genscript were used.
By these manipulations, a chimeric antibody expression plasmid containing the variable region of the humanized anti-Id antibody and the constant region of the mouse antibody was obtained.

Next, the obtained chimeric antibody expression plasmid was used as a transfection plasmid and transient expression was carried out using the "Expi293 Expression System" (Thermo Fisher Scientific).
After transient expression, the culture supernatant was collected and affinity purified using "rProtein A Sepharose fast flow, cat. 17127903" (Cytiva). The resulting eluate was concentrated and buffer exchanged into PBS using "PD-10 column, cat. 17085101" (Cytiva).
The eluted fractions of the buffer-exchanged eluate were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and subjected to the following tests as "Hu #2-911" mouse chimeric antibody (hereinafter also referred to as "Hu #2-911 mouse chimera").

(2-2) Verification of reactivity with "MEDI8852" The "MEDI8852" antibody corresponds to the antigen of "Hu #2-911."
Therefore, the reactivity of "Hu #2-911 mouse chimera" with "MEDI8852" was examined based on the following method.

The verification was carried out using ELISA in an antigen solid-phase format in which "MEDI8852" was immobilized.
Note that "MEDI8852" is a human antibody, and "#2-911" is a mouse antibody. In this example, "Hu #2-911" was prepared as a "Hu #2-911 mouse chimera" so that the reactivity of "Hu #2-911" to "MEDI8852" and the reactivity of "#2-911" to "MEDI8852" could be compared using the same secondary antibody.

For each of "#2-911" and "Hu #2-911 Mouse Chimera", a seven-step dilution series was prepared with 1% BSA/PBS at a common ratio of 3, starting from 10 μg/ml.
The resulting dilution series was reacted with the antigen and detected using a secondary antibody, "anti-mouse IgG-Fc fragment, HRP conjugated" (Bethyl Laboratories).
For color development, an "ELISA POD Substrate TMB Kit" (popular) (Nacalai Tesque) was used, and absorbance at a wavelength of 450 nm was measured using an "Infinite M1000 pro microplate reader" (Tecan). ELISA was performed in duplicate for each point, and the average value was calculated.

The results based on the above ELISA are shown in Figure 11. This figure shows the reactivity of "Hu #2-911 mouse chimera" with "MEDI8852."
The EC50 of each antibody calculated from these results is as follows:
・"#2-911": 1.67μg/mL
"Hu #2-911 VH1/VL1" (combination of SEQ ID NOs: 17 and 19): 1.71 μg/mL
"Hu #2-911 VH1/VL2" (combination of SEQ ID NOs: 17 and 20): 2.34 μg/mL
"Hu #2-911 VH2/VL1" (combination of SEQ ID NOs: 18 and 19): 1.43 μg/mL
"Hu #2-911 VH2/VL2" (combination of SEQ ID NOs: 18 and 20): 0.9 μg/mL
Therefore, it was found that the "Hu #2-911 mouse chimera" exhibited reactivity to the antigen "MEDI8852" equivalent to that of the original antibody "#2-911."

Even though the framework sequence was replaced with a different sequence due to humanization of the antibody sequence, it still exhibited the same reactivity as "#2-911," suggesting that the antigen reactivity of "#2-911" is determined by the CDR sequence.

<Test 6: Establishment of anti-Id antibody-producing cells against "Flu A-20 antibody">
Based on the following method, a hybridoma clone producing an anti-Id antibody against "FluA-20 antibody," a known broadly neutralizing anti-influenza HA antibody, was obtained.

(1) Preparation of "FluA-20 Antibody" Based on the amino acid sequence of "FluA-20" (Bangaru, et al., Cell, 177, 1136-1152, 2019, WO2020/232426), DNA was synthesized by adding a Kozak translation initiation sequence to the 5' end of each of the heavy chain variable region and light chain variable region.
The obtained heavy chain variable region was cloned into the "pFUSE-CHIg-hG1" vector (InvivoGen) to obtain a "FluA-20 heavy chain expression plasmid."
The obtained light chain variable region was cloned into the "pFUSE2-CLIg-hk" vector (InvivoGen) to obtain the "FluA-20 light chain expression plasmid."
For cloning, the services of Azenta were used.

Next, transient expression was carried out using the "FluA-20 heavy chain expression plasmid" and the "FluA-20 light chain expression plasmid" as the introduced plasmids using the "Expi293 Expression System" (Thermo Fisher Scientific).
After transient expression, the culture supernatant was collected and affinity purified using a "HiTrap Protein A HP column" (Cytiva). The resulting eluate was concentrated and purified by gel filtration using a "HiLoad 26/600 Superdex 200 pg column" (Cytiva).
The eluted fractions from the gel filtration purification were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and subjected to the following tests as "FluA-20 antibody."

(2) Preparation of Hybridoma Clones Producing Anti-Id Antibodies Against "FluA-20 Antibody" Hybridoma clones producing anti-Id antibodies specific to "FluA-20 antibody" were obtained according to the following method.

(2-1) Immunization of Mice The antigen used was "FluA-20 antibody," which was mixed with the adjuvant "TiterMax Gold" (TiterMax) and administered to the soles of BALB/c mice. The antigen dose was set at 50 μg per mouse.
Booster immunizations were performed 3, 10, and 12 days after administration, with the antigen dose for booster immunization set at 5 μg per mouse, and PBS buffer used instead of an adjuvant.

After the immunization (2 days after the final administration), popliteal lymph nodes were collected from the mice and a cell suspension was prepared.
The resulting cell suspension was mixed with SP2/0-Ag14 myeloma cells, and electrofusion was carried out using an electrofusion device "ECFG21" (Neppa Gene).
After fusion, the resulting cells were suspended in "ClonaCell-HY Medium D" (STEMCELL Technologies) and seeded onto plastic dishes.
Colonies formed 10 days after seeding were isolated in 96-well plastic plates containing hybridoma medium, and the culture supernatant was used for the following evaluation. The hybridoma medium used was "RPMI1640" (Thermo Fisher Scientific) supplemented with 1/50 of the amount of "Nutridoma-CS, cat. 1136374001" (Merck Millipore) and "HAT Supplement" (Thermo Fisher Scientific).

(2-2) Selection of clones producing anti-Id antibodies specific to "Flu A-20 antibody" Using each culture supernatant obtained in "(2-1) Immunization of mice" above, hybridoma clones producing anti-Id antibodies specific to "Flu A-20 antibody" were screened.
Specifically, clones producing antibodies that meet all of the following criteria were selected by ELISA.
(Criterion 1) Reacts with (binds to) "FluA-20 antibody."
(Criterion 2) The antibody does not react with the control antibody "human IgG1 kappa-UNLB" (Southern Biotech) which has the same isotype as the "FluA-20 antibody."

The ELISA was carried out in an antigen solid-phase format in which "FluA-20 antibody" or "human IgG1-kappa-UNLB (Southern Biotech)" was immobilized.
The secondary antibody used for detection was "anti-mouse IgG-Fc fragment, HRP conjugated" (Bethyl Laboratories).
For color development, an "ELISA POD Substrate TMB Kit" (popular) (Nacalai Tesque) was used, and absorbance at a wavelength of 450 nm was measured using an "Infinite M1000 pro microplate reader" (Tecan).

<Test 7: Selection of anti-Id antibodies that induce antibodies reactive to HA>
An anti-Id antibody against the "FluA-20 antibody" was obtained according to the following method. This antibody can induce antibodies that react with HA when inoculated as an antigen.

(1) Sequence Analysis of Anti-Id Antibody Hybridoma cells were expanded and cultured, and RNA was extracted from the resulting culture using "RNeasy mini Kit, cat. 74104" (QIAGEN).
Next, the DNA sequences of the antibody variable regions were analyzed by an external contractor (Azenta).

The antibody sequence and its CDR region sequence identified as a result of DNA sequence analysis of the anti-Id antibody are shown in Table 8. The anti-Id antibody having these sequences will also be referred to as "R2-150" hereinafter.

(2) Preparation of anti-Id antibody Fab nanoparticles "R2-150 Fab nanoparticles" DNA encoding the sequence of a fusion protein to which ferritin protein was bound was linked downstream of the sequence of the heavy chain Fab constituent region of "R2-150" via a linker sequence containing five repeats of a sequence consisting of the four amino acids serine-glycine-glycine-glycine.
This fusion protein was designed according to Cell, 162, 1090-1100, 2015, and has a Kozak translation initiation sequence added to the 5' side and a stop codon added to the 3' side.
The obtained gene was cloned into the "pcDNA3.4" (Thermo Fisher Scientific) vector to obtain an anti-Id antibody Fab-nanoparticle expression plasmid (heavy chain).

Furthermore, the light chain Fab component region of "R2-150" was also cloned into the "pcDNA3.4" (Thermo Fisher Scientific) vector using the same procedure as above to obtain an anti-Id antibody Fab-antibody Fab-nanoparticle expression plasmid (light chain).

Next, the obtained expression plasmids (heavy chain and light chain) were used as introduction plasmids to carry out transient expression using the "Expi293 Expression System" (Thermo Fisher Scientific).
After transient expression, the culture supernatant was collected and subjected to affinity purification using "CaptureSelect LC-kappa (mur) affinity matrix, cat. 191315005" (Thermo Fischer Scientific).
The resulting eluate was concentrated and purified by gel filtration using a "HiPrep 26/60 Sephacryl S-500HR column" (Cytiva).
The eluted fractions from the gel filtration purification were confirmed by SDS-PAGE, and the fractions containing the target protein were collected and used hereinafter as "R2-150Fab nanoparticles."

(3) Immunization of Mice The antigen used was "R2-150Fab nanoparticles," which were mixed with the adjuvant "TiterMax Gold" (TiterMax) and intraperitoneally administered to BALB/c mice. The antigen dose was set at 200 μg per mouse.
After administration, booster immunizations were performed twice at two-week intervals from the start of administration. The antigen dose for booster immunization was set at 20 μg per mouse, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant.
One week after the first booster immunization, partial blood samples were taken from the mice, and the production of antibodies against the anti-Id antibody used as the antigen was confirmed by ELISA. One week after the final immunization, whole blood was collected, and serum was prepared and used for the following verification.

(4) Verification of reactivity with HA Verification of reactivity with HA was performed by ELISA using an antigen solid-phase format in which an antigen panel consisting of various HAs was immobilized, as in Test 2, "(3-2) Verification of reactivity with HA."

The results based on the ELISA are shown in Figure 12. This figure shows the reactivity of serum from mice immunized with "R2-150 Fab nanoparticles" with a panel of recombinant influenza A HA antigens.
These results demonstrate that the anti-Id antibody (R2-150) functions as an antigen capable of inducing antibodies against all known influenza A virus HA subtypes (except H9).

Furthermore, it was confirmed that the antibodies contained in the immune serum reacted with both HA of influenza B Victoria strain and Yamagata strain, which cannot be reacted with "FluA-20 antibody," the template antibody of "R2-150" (Figure 13).
These results indicate that new antibodies similar in function and morphology to, or even superior in function to, the template antibody (FluA-20 antibody) can be obtained by using "R2-150" as an antigen, for example, by immunizing an animal to obtain an antibody.

<Test 8: Analysis of the interaction between "#2-911" and template antibody>
In this example, a complex of Fab of anti-Id antibody "#2-911" and its template Fab of "MEDI8852 antibody" was crystallized, and its three-dimensional structure was analyzed.
This analysis identified detailed structural coordinates, such as the positions of the amino acid side chains that make up the Fab and the hydration water located on the surface of the protein molecule.

The analysis results are shown in Figures 14, 15 and 16. These figures show the relationship of the interaction (binding) between "#2-911" and "MEDI8852 antibody" from three perspectives.
The numbers assigned to each amino acid in Figures 14, 15, and 16 are numbers according to the definition by Kabat (Sequences of Proteins of Immunological Interest, Fifth Edition, NIH Publication No. 91-3242, U.S. Department of Health and Human Services, 1991).

FIG. 14 shows the crystal structure of the complex between Fab #2-911 and Fab #MEDI8852, with the interaction region between the two displayed and analyzed using the software "Pymol" (Schrodinger). The presence or absence of contact between amino acids between Fab #2-911 and Fab #MEDI8852 was objectively determined by calculation from the structural coordinates using the "InterfaceResidues" script (https://pymolwiki.org/index.php/InterfaceResidues) on the software. More specifically, amino acids with a contact area of 1.0 Å^2 or greater between Fab #2-911 and Fab #MEDI8852 were determined to have a direct interaction.
FIG. 14 illustrates the main chain structures of the "#2-911" Fab and the "MEDI8852 antibody" Fab, which showed direct binding to it, in a "ribbon model." For each Fab, the CDR regions are shown in dark color. To facilitate understanding of the interaction, a "space-filling model" including the side chains is shown for HCDR1 and HCDR3 of the "MEDI8852 antibody." In addition, the side chains of the amino acids constituting the "#2-911" HCDR1 and HCDR2 are shown in a "stick model." In the crystal structure, direct binding to the "MEDI8852 antibody" was observed other than those in the "#2-911" HCDR1 and HCDR2. For amino acids on the framework that were considered to be particularly important, their side chains are shown in a "stick model," and the observed hydrogen bonds are illustrated. In the figure, the CDR numbers and residue numbers of particularly noteworthy amino acids are shown (those of the "MEDI8852 antibody" are shown in italics).

15 and 16, like FIG. 14, show the interaction between Fab of "#2-911" and Fab of "MEDI8852 antibody" observed from a different direction than that of FIG.
Of these, Figure 16 more clearly illustrates the three-dimensional positional relationship between the "#2-911" Fab (shown as a "cartoon model") and the "MEDI8852 antibody" Fab (shown as a "space-filling model") that make up the complex. The three CDRs of the "#2-911" heavy chain and the "#2-911" light chain are each shown in dark colors in the figure.

As can be seen from these figures, the crystal structure of the complex shows that the "MEDI8852 antibody" directly interacts with a region composed of HCDR1 and HCDR2 of "#2-911" and the surrounding framework regions. Specifically, the crystal structure of the complex shows that HCDR1 and HCDR2 of "#2-911" have a large binding area (area of direct interaction), particularly with HCDR3, which is composed of highly hydrophobic amino acids of the "MEDI8852 antibody," and are responsible for the primary interaction with the antigen (FIGS. 14 and 16). It was revealed that not only HCDR1 and 2, but also specific amino acids described below are particularly involved in binding.
On the other hand, no amino acids that directly interact with the "MEDI8852 antibody" were detected in the HCDR3 and light chain of "#2-911," and the HCDR3 and light chain did not bind to the "MEDI8852 antibody." This was also visually apparent, particularly in Figure 16, where a clear distance was observed between the HCDR3 and light chain and the molecules that make up the "MEDI8852 antibody."

The side chains of amino acids Tyr27 and Phe29 on the N-terminal framework (also referred to as FR1) of HCDR1 of "#2-911" showed direct interactions with the heavy chain amino acids of "MEDI8852 antibody."
Depending on the antibody, for example, Phe, Leu, Ile, or Glu can be used instead of Tyr as the 27th amino acid, and for example, Leu, Ile, Val, or Ala can be used instead of Phe as the 29th amino acid.
Using the structural coordinates identified in this example, Tyr27 and Phe29 were substituted with the above-mentioned amino acids on the "Pymol" program, and it was confirmed that these amino acids, like Tyr27 and Phe29, can show direct interactions with the heavy chain amino acids of the "MEDI8852 antibody."

Furthermore, the amino acid Lys73 in the framework (also referred to as FR3) between HCDR2 and HCDR3 of "#2-911" showed direct interaction with the side chain of Ser53 on the light chain of "MEDI8852 antibody." Lys73 formed hydrogen bonds with the side chains of amino acids Ser30 and Ser31 on the light chain of "MEDI8852 antibody" (hydrogen bonds are indicated by dotted lines extending from Lys73 in Figures 14 and 15).
Furthermore, the side chain of amino acid Ser76 on the heavy chain FR3 of "#2-911" formed a hydrogen bond with the main chain of the heavy chain Phe100a of "MEDI8852 antibody" (in Figure 15, the hydrogen bond is indicated by the dotted line extending from Ser76).

These results indicate that Tyr27, Phe29, Lys73, and Ser76 present on the heavy chain framework of #2-911, along with HCDR1 and HCDR2 of #2-911, are particularly involved in the binding between #2-911 and the MEDI8852 antibody.

The importance of HCDR1 and HCDR2 of "#2-911" deduced from the above-mentioned three-dimensional structure for the binding to the "MEDI8852 antibody" and the importance of specific amino acids in the FR region for the binding to the "MEDI8852 antibody" were demonstrated by the changes in binding affinity of the alanine-substituted mutants to the "MEDI8852 antibody" in Figure 17.
That is, when His35 contained in the HCDR1 sequence, Phe53 contained in the HCDR2 sequence, Tyr27 in FR1, and Lys73 in FR3 of "#2-911" are substituted with alanine, the binding strength with the "MEDI8852 antibody" is significantly reduced or lost. Furthermore, it was found that substituting Phe29 in FR1 with alanine strengthens the binding strength between "#2-911" and the "MEDI8852 antibody."

<Test 9: Obtaining a new antibody that exceeds the function of the template antibody using an anti-Id antibody>
By using an anti-Id antibody as an antigen, it is expected that antibodies similar in function and morphology to the template antibody, or new antibodies with functions exceeding those of the template antibody, can be obtained.
To verify this point, five new monoclonal antibodies (clones A to E) were obtained from a hybridoma expressing an antibody that reacts with the anti-Id antibody "#2-911" as an antigen.
Specifically, the "MEDI8852 antibody," which is the template antibody for "#2-911," broadly cross-reacts with influenza A HA, but does not bind to influenza B HA. Therefore, an attempt was made to obtain a clone expressing a monoclonal antibody that cross-reacts with both influenza A HA and influenza B HA from a hybridoma expressing an antibody against "#2-911" according to the following procedure.

(1) Immunization of Mice The #2-911 Fab nanoparticles used in Experiments 3 and 4 were used as an antigen. They were mixed with the adjuvant TiterMax Gold (TiterMax) and intraperitoneally administered to 13 BALB/c mice. The antigen dose was set at 200 μg per mouse.
After administration, booster immunizations were performed twice at two-week intervals from the start of administration. The antigen dose for booster immunization was set at 20 μg per mouse, and "Imject Alum Adjuvant, cat. 77161" (Thermo Fischer Scientific) was used as the adjuvant.
One week after the final immunization, popliteal lymph nodes and spleens were collected from the mice, and a cell suspension was prepared from a portion of the spleens.

The resulting cell suspension was mixed with SP2/0-Ag14 myeloma cells, and electrofusion was carried out using an electrofusion device "ECFG21" (Neppa Gene).
After fusion, the resulting cells were suspended in "ClonaCell-HY Medium D, cat. ST-03804" (STEMCELL Technologies) and seeded onto plastic dishes.

Colonies formed 10 days after seeding were isolated in 96-well plastic plates containing hybridoma medium, and the culture supernatant was used for the following evaluation. The hybridoma medium used was "RPMI1640, cat. A1049101" (Thermo Fisher Scientific) supplemented with 1/10 the amount of "Doma-Drive, cat. T31-1003SF" (Immune Systems) and 1/50 the amount of "HAT Supplement, cat. 21-60017" (Thermo Fisher Scientific).

(2) Screening of monoclonal antibodies cross-reactive with influenza A and B HAs Using each culture supernatant obtained in the above "(1) Immunization of mice," hybridoma clones producing monoclonal antibodies cross-reactive with both influenza A HA and influenza B were screened.
Specifically, clones producing antibodies that meet all of the following criteria were selected by ELISA.
(Criterion 1) Reacts with (binds to) "#2-911 Fab nanoparticles."
(Criterion 2) No reaction with mouse polyclonal Fab "mouse IgG Fab fragment, cat. 010-0105" (Rockland).
(Criterion 3) Reacts to a mixed antigen (mixed influenza A HA antigen) consisting of influenza A HA (8 types shown in Table 9).
(Criterion 4) Reacts to a mixed antigen (mixed influenza B HA antigen) consisting of influenza B HA (8 types shown in Table 5).

The ELISA was carried out in an antigen solid-phase format in which each of the above antigens was immobilized.
The secondary antibody used for detection was "anti-mouse IgG-Fc fragment, HRP conjugated, cat. A90-131" (Bethyl Laboratories).
For color development, "ELISA POD Substrate TMB Kit (popular), cat. 05298-80" (Nacalai Tesque) was used, and absorbance at a wavelength of 450 nm was measured using "Infinite M200 pro microplate reader" (Tecan).

The following three types of controls were set up.
"#2-911Fab-nps" is the antigen used for immunization, "#2-911Fab nanoparticles" (positive antigen).
"mIgG polyclonal Fab" is a mouse polyclonal Fab "mouse IgG Fab fragment, cat. 010-0105" (Rockland) (negative antigen).
"SARS-CoV-2 Spike" is the spike protein of the SARS-CoV-2 virus (Wuhan strain) (negative antigen) prepared by adding the same sequence as the purification peptide tag added to the C-terminus of each HA protein during the preparation of the HA antigen panel.

Five clones, clones A to E, were selected as hybridoma clones that fulfilled all of the above criteria (Criteria 1) to (Criteria 4). The results of confirming their individual reactivity with 14 types of influenza A and two types of influenza B HA are shown in Figure 18. It was confirmed that multiple clones were obtained that produced antibodies that showed cross-reactivity with various subtypes of influenza A HA and also cross-reacted with HA of the Yamagata and Victoria lineages (known as influenza B HA).

The template antigen for "#2-911,""MEDI8852antibody," is an antibody that specifically reacts with influenza A HA and strongly reacts with all known subtypes from H1 to H18 (Kallewaard et al., Cell, 166, 596-608, 2016, and WO2017/123685). These 18 types of influenza A HA are classified into "Group 1" and "Group 2" based on differences in amino acid sequences resulting from differences in their common ancestors (see Table 9), and it is known that there are significant differences in antigenicity between the two groups.
However, all of the antibodies obtained from mice immunized with "#2-911" as an antigen not only broadly cross-react with HAs belonging to both groups, like the "MEDI8852 antibody," but also cross-react with influenza B HA, which the "MEDI8852 antibody" cannot bind to.
In other words, the approach of this example yielded an antibody that is similar to the template antibody but has better functionality.

Claims (8)

An antibody or fragment thereof that specifically binds to the "MEDI8852 antibody," an anti-influenza HA broadly neutralizing antibody, and that satisfies all of the following requirement A:
[Requirement A]
- Heavy chain CDR1 is the amino acid sequence shown in SEQ ID NO:2.
- Heavy chain CDR2 is the amino acid sequence shown in SEQ ID NO:3.
The 27th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Tyr, Phe, Leu, Ile, and Glu.
The 29th amino acid of the heavy chain FR1 based on the Kabat definition is an amino acid selected from the group consisting of Phe, Leu, Ile, Val, and Ala.
The 73rd amino acid in the heavy chain FR3 based on the Kabat definition is Lys.
The 76th amino acid of the heavy chain FR3 based on the Kabat definition is Ser. The antibody or fragment thereof according to claim 1, wherein the 27th amino acid of the heavy chain FR1 is Tyr and the 29th amino acid of the heavy chain FR1 is Phe. The antibody or fragment thereof according to claim 1 or 2, wherein the amino acid sequences of CDR1 to CDR3 of the heavy and light chains further satisfy all of the following [Requirement B]:
[Requirement B]
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:4.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:6.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:7.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:8. The antibody or fragment thereof according to any one of claims 1 to 3, further satisfying all of the following [Requirement C]:
[Requirement C]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:1.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:5. The antibody or fragment thereof according to any one of claims 1 to 3, further satisfying all of the following [Requirement D]:
[Requirement D]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO: 17 or 18.
The light chain variable region has the amino acid sequence shown in SEQ ID NO: 19 or 20. An antibody or fragment thereof that specifically binds to the "FluA-20 antibody," an anti-influenza HA broadly neutralizing antibody, in which the amino acid sequences of CDR1 to CDR3 of the heavy and light chains satisfy all of the following [Requirement E]:
[Requirement E]
The heavy chain CDR1 has the amino acid sequence shown in SEQ ID NO:10.
The heavy chain CDR2 has the amino acid sequence shown in SEQ ID NO:11.
The heavy chain CDR3 has the amino acid sequence shown in SEQ ID NO:12.
The light chain CDR1 has the amino acid sequence shown in SEQ ID NO:14.
The light chain CDR2 has the amino acid sequence shown in SEQ ID NO:15.
The light chain CDR3 has the amino acid sequence shown in SEQ ID NO:16. The antibody or fragment thereof according to claim 6, further satisfying all of the following [Requirement F]:
[Requirement F]
The heavy chain variable region has the amino acid sequence shown in SEQ ID NO:9.
The light chain variable region has the amino acid sequence shown in SEQ ID NO:13. An anti-influenza vaccine comprising the antibody or fragment thereof described in any one of claims 1 to 7, or a nucleic acid encoding the antibody or fragment thereof.