Modified HSV gB protein and HSV vaccine including same

Abstract

A modified protein of a herpes simplex virus (HSV) envelope glycoprotein B (gB), in which at least one non-neutralizing antibody-inducing epitope (non-neutralizing epitope) present in domain IV and domain I of wild-type HSV gB is inactivated (de-epitoped).

Claims

1. A modified protein of a herpes simplex virus 2 (HSV-2) envelope glycoprotein B (gB) (modified HSV-2 gB protein), wherein the modified HSV-2 gB protein is derived from a wild-type HSV gB set forth in SEQ ID NO:1, wherein the wild-type HSV-2 gB protein is modified to introduce a N-type glycochain at one or more sites of interest by a substitution of an amino acid residue, a deletion of an amino acid residue, and/or a substitution or deletion of an amino acid residue, such that the three continuous amino acid sequence of the site of interest at which the N-type glycochain is introduced become NXS or NXT, wherein the “X” in the consensus sequences NXS or NXT is any amino acid except a proline, and wherein the one or more sites of interest are selected from the group consisting of amino acids residues D199, R567, R602 and S631 in the amino acid sequence set forth in SEQ ID NO:1.

2. The modified HSV-2 gB protein according to claim 1, wherein the modification includes a modification performed by a substitution of an amino acid residue and a deletion of an amino acid residue.

3. The modified HSV-2 gB protein according to claim 1, wherein the modification includes a modification performed by introducing a glycochain by the substitution or deletion of an amino acid residue.

4. The modified HSV-2 gB protein according to claim 1, wherein a N-type glycochain is introduced at one site of interest selected from the group consisting of amino acid residues D199, R567, R602 and S631 in the amino acid sequence set forth in SEQ ID NO: 1.

5. The modified HSV-2 gB protein according to claim 1, wherein a N-type glycochain is introduced at two or more sites of interest selected from the group consisting of amino acid residues D199, R567, R602 and S631 in the amino acid sequence set forth in SEQ ID NO: 1.

6. The modified HSV-2 gB protein according to claim 5, wherein the sites of interest are amino acid residues R567 and S631 in the amino acid sequence set forth in SEQ ID NO: 1.

7. The modified HSV-2 gB protein according to claim 6, comprising amino acid residue substitutions R567N, P568S, G569S, S631N, H632T and Q633T in the amino acid sequence set forth in SEQ ID NO: 1.

8. The modified HSV-2 gB protein according to claim 5, wherein the sites of interest are amino acid residues D199, R567 and S631 in the amino acid sequence set forth in SEQ ID NO: 1.

9. The modified HSV-2 gB protein according to claim 4, wherein the site of interest is amino acid residue R602 in the amino acid sequence set forth in SEQ ID NO: 1.

10. The modified HSV-2 gB protein according to claim 9, comprising amino acid residue substitutions R602N, D603A, A604T in the amino acid sequence set forth in SEQ ID NO: 1.

11. The modified HSV-2 gB protein according to claim 3, wherein the site of interest is amino acid residue D199 in the amino acid sequence set forth in SEQ ID NO: 1.

12. The modified HSV-2 gB protein according to claim 11, comprising amino acid residue substitutions D199N, D200A, and H201T in the amino acid sequence set forth in SEQ ID NO: 1.

13. The modified HSV-2 gB protein according to claim 2, comprising a substitution of an amino acid residue corresponding to an arginine at position 613 (R613) in the amino acid sequence set forth in SEQ ID NO: 1 with an alanine residue.

14. An HSV vaccine comprising the modified HSV-2 gB protein according to claim 1.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 is a diagram showing the results of SDS-PAGE and Western Blotting using gB of Example 2 and protease cleavage fragments thereof.

(2) FIG. 2 is a correlation diagram obtained by classifying anti-gB antibodies obtained by competitive ELISA of Example 2.

(3) FIG. 3 is a MOE diagram of the results of identifying epitopes of gB antibodies by alanine scanning of Example 3.

(4) FIG. 4 is a diagram illustrating a schematic for a design strategy of a modified gB protein of Example 5.

(5) FIG. 5 is a diagram illustrating a simplified schematic of a crystal structure for a design strategy of a modified gB protein of Example 5.

(6) FIG. 6 is a diagram showing the result of the mouse immunogenicity test of bcev19 in Example 5.

(7) FIG. 7 is a diagram showing the result of the mouse immunogenicity test of bcev50 in Example 5.

(8) FIG. 8 is a diagram showing the result of survival rate of the mouse infection-prevention test of bcev19 in Example 5.

(9) FIG. 9 is a diagram showing the result of symptom score of the mouse infection-prevention test of bcev19 in Example 5.

(10) FIG. 10 is a diagram showing the result of survival rate of the mouse infection-prevention test of bcev50 in Example 5.

(11) FIG. 11 is a diagram showing the result of symptom score of the mouse infection-prevention test of bcev50 in Example 5.

(12) FIG. 12 is a diagram showing the analysis result of immune refocusing of bcev19 in Example 5.

(13) FIG. 13 is a diagram showing the analysis result of immune refocusing of bcev50 in Example 5.

(14) FIG. 14 is a diagram showing the result of property analysis of bcev19, bcev19′, bcev50 and bcev50′ by gel filtration chromatography in Example 5.

(15) FIG. 15 is a diagram showing the comparison result of bcev19 and bcev19′ in the mouse immunogenicity test of Example 5.

(16) FIG. 16 is a diagram showing the comparison result of bcev50 and bcev50′ in the mouse immunogenicity test of Example 5.

(17) FIG. 17 is a diagram showing the comparison result of multiple alignments of the amino acid sequence of HSV-1 derived gB (SEQ ID NO: 2) and the amino acid sequence of HSV-2 derived gB (SEQ ID NO: 3). The italics indicate the leader sequence, and the underlines indicate the amino acid residues at positions 383 to 388 (I383-R388) of HSV-1 derived gB domain II and the amino acid residues at positions 386 to 391 (I386-R391) of HSV-2 derived gB domain II.

DESCRIPTION OF EMBODIMENTS

(18) Hereinafter, the embodiments for carrying out the present invention are described in detail below. However, the present invention is not limited to the following embodiments.

(19) The modified HSV gB protein of the present invention is a modified protein of an envelope glycoprotein B (gB) of herpes simplex virus (HSV), which is modified so that at least one non-neutralizing antibody-inducing epitope (non-neutralizing epitope) present in domains IV and I of wild-type gB, does not function as an epitope.

(20) The present invention is based on the hypothesis proposed by the present inventors that there is a “decoy region” in the HSV gB antigen. The “decoy region” is derived from an English word “Decoy” and is considered one of the immunoediting systems by which the pathogen escapes the host's immune response. The “decoy region” is an antigen region that induces antibodies having no or low neutralizing antibody activity, and is believed to be a mechanism by which pathogens escape the host's immune response such that the neutralizing antibodies are not produced or are produced in small amounts by the deceptive inprinting (also referred to as “immune deviation”).

(21) Until now, the presence of the decoy region has not been confirmed in HSV, not even the concept of the decoy region. The present inventors performed detailed epitope mapping analyses on anti-gB monoclonal antibodies obtained by comprehensive exploration of anti-HSV gB antibodies performed using a human antibody library. As the result, domains IV and I of HSV gB were first revealed to be decoy regions in which unbeneficial or deleterious epitopes are concentrated. Furthermore, in these decoy regions, by de-epitoping the non-neutralizing epitopes and immunologically emphasizing the beneficial epitope, modified HSV gB proteins that are capable of inducing antibodies with high neutralizing activity have been finally obtained.

(22) The “wild-type HSV gB” refers to the full length of HSV-1 derived envelope glycoprotein B (gB) having the amino acid sequence set forth in SEQ ID NO: 2, or HSV-2 derived gB having the amino acid sequence set forth in SEQ ID NO: 3. As the result of comparing the two sequences in multiple alignments, the sequence identity is about 87% (FIG. 17). The conformation of gB has also been analyzed and is known to consist of an intracellular domain, a transmembrane domain, and an ectodomain. The crystal structure of gB derived from HSV-1 has been reported, for example, by Science 313: 217-220 (2006) and J. Virol. 84: 12924-12933 (2010). Although the crystal structure of gB derived from HSV-2 has not been reported, it can be analyzed similarly according to the crystallization method of gB derived from HSV-1 described above. The “ectodomain of wild-type HSV gB” means a soluble, antigenic extracellular region of the wild-type HSV gB. An example of the ectodomain of the wild-type HSV gB is a wild-type gB ectodomain 1-705 derived from a 333 strain of HSV-2 consisting of the amino acid sequence set forth in SEQ ID NO: 1.

(23) In the crystal structure of gB, domains IV and I, located at the top (Crown) and bottom (Bottom) respectively, are “more conspicuous” than domain II located at the middle (Middle), resulting in higher antigen presentation. Indeed in antibodies in blood, as investigated by the present inventors, the percentage of antibodies to domains IV and I is higher, and the percentage of antibodies to domain II is lower. However, according to the investigation by the present inventors, whereas domain II of wild-type HSV gB only contains an epitope that induce a neutralizing antibody (herein referred to as a “neutralizing epitope”), domains IV and I of wild-type HSV gB contain not only neutralizing epitopes but also an epitope that induces a non-neutralizing antibody (herein referred to as a “non-neutralizing epitope”). Since non-neutralizing antibodies bind antigens (i.e., viruses) but are unable to suppress the viral activity, it is believed that ability to induce production of neutralizing antibodies rather than non-neutralizing antibodies is important, particularly in the manufacture of vaccines.

(24) The modified HSV gB proteins of the present invention can derive antibodies having high neutralizing activity by “de-epitoping” the “conspicuous” domains IV and I, which are less beneficial for the production of neutralizing antibodies, and making them “inconspicuous”, thereby making the domain II containing neutralizing epitopes “conspicuous”.

(25) The “modified HSV gB protein” (“modified protein of HSV gB” or “variant”) is a protein in which at least one amino acid residue or a region of contiguous amino acid residues is substituted, deleted or added to a wild-type HSV gB, and includes a protein in which a protein modification that is not present in the wild-type is performed, such as a protein in which a glycochain is introduced by substitution or deficiency of the amino acid residue. The modified HSV gB protein of the invention has higher neutralizing antibody-inducing activity than the wild-type protein.

(26) The “neutralizing antibody-inducing activity” refers to the ability to induce neutralizing antibodies of an antigen protein, and can be evaluated by the neutralizing antibody titer in immune serum obtained by inoculating the antigen protein in a subject animal. The “neutralizing antibody” refers to an antibody that is capable of losing the infectivity of a viral particle. The neutralizing antibody is, for example, evaluated by the intensity of neutralizing activity of the antibody at a concentration (NT50) necessary to reduce the plaque number of the subject virus by 50%.

(27) The “de-epitoping” refers to a modification of a site contributed to antibody production as an epitope in the wild-type HSV gB so as not to function as an epitope. Examples of the method of de-epitoping include a method of substituting an amino acid residue at the site of an epitope with another amino acid residue; a method of defecting (deleting) an amino acid residue at the site of an epitope; and a method of introducing a glycochain by substitution or deficiency of an amino acid residue at the site of an epitope. As the method of de-epitoping, the method of introducing a glycochain, in particular an N-type glycochain (N-glycoside-linked glycochain) is preferable. The method has advantages in that not only the portion where the glycochain is introduced, but also the epitopes in the periphery can be masked at the same time due to its bulkiness. Considering the size ratio to proteins such as antibodies or receptors that interact with gB, it is expected that the dot-to-dot interaction such that binding is formed in a very narrow range of about a few amino acids is less likely to occur. In the binding between gB and receptors, it is believed that a network of facet-to-facet interaction is formed such that a wide range of amino acids cooperate to form a binding. Introduction of a glycochain is believed to be an effective way of de-epitoping to hide peripheral residues extensively by its own bulkiness, and simultaneously to inhibit access of the binding partner. It is also reported that the glycochain is difficult to induce anti-glycochain antibodies, thus it is believed that the likelihood of new immunogenicity due to modification can be reduced.

(28) An example of the non-neutralizing antibody-inducing epitope (non-neutralizing epitope) present in domains IV and I of wild-type HSV gB includes an epitope containing at least one amino acid residue present in a region at a distance of 1.5 nm or less from an amino acid residue corresponding to an arginine residue at position 567 (R567), an arginine residue at position 602 (R602), a serine residue at position 631 (S631) or an aspartic acid residue at position 199 (D199), more preferably a region at a distance of 1 nm or less from the amino acid residue, in the amino acid sequence set forth in SEQ ID NO: 1 in a surface of a crystal structure of an ectodomain of the wild-type HSV gB. Here, the “distance from an amino acid residue” refers to the linear distance from the amino acid residue described above, regardless of the shape of the surface of the crystal structure of the wild-type HSV gB ectodomain. Since non-neutralizing epitopes induce the production of unbeneficial or deleterious antibodies and are not beneficial for the production of neutralizing antigens, de-epitoping of these non-neutralizing epitopes can reduce the production of unbeneficial or deleterious antibodies and increase the production of neutralizing antibodies by emphasizing beneficial epitopes. The crystallization methods are not particularly limited, but examples thereof include a crystallization method described in J. Virol. 84: 12924-12933 (2010). For example, crystals of gB can be grown with 15% PEG 4000-0.3 M NaCl-0.1 M sodium citrate, pH 5.5.

(29) Examples of the non-neutralizing epitope include epitopes containing an amino acid residue corresponding to an arginine residue at position 567 (R567), an arginine residue at position 602 (R602), a serine residue at position 631 (S631) or an aspartic acid residue at position 199 (D199) in the amino acid sequence set forth in SEQ ID NO: 1. Since these epitopes have been proven to be non-neutralizing epitopes by the present inventors, de-epitoping of these epitopes can reduce the production of unbeneficial or deleterious antibodies in preventive activity expression. Furthermore, de-epitoping of these epitopes can also increase the percentage of production of neutralizing antibodies by emphasizing beneficial epitopes.

(30) The modification for de-epitoping include a modification performed by introducing a glycochain by a substitution of an amino acid residue, a deficiency of an amino acid residue, and/or a substitution or deficiency of an amino acid residue.

(31) It is preferred that the modification includes a modification for introducing the glycochain to a position of at least one amino acid residue, preferably at least two amino acid residues, selected from the group consisting of amino acid residues corresponding to an aspartic acid residue at position 199 (D199), an arginine residue at position 567 (R567), an arginine residue at position 602 (R602), and a serine residue at position 631 (S631) in the amino acid sequence set forth in SEQ ID NO: 1. It is preferred that the modification includes a modification for introducing a glycochain to each position of amino acid residues corresponding to R567 and 5631 in the amino acid sequence set forth in SEQ ID NO: 1, and it is preferred that the introducing a glycochain is performed by amino acid residue substitutions of R567N, P568S, G569S, S631N, H632T and Q633T in the amino acid sequence set forth in SEQ ID NO: 1. It is further preferred that the modification includes a modification for introducing a glycochain to each position of amino acid residues corresponding to D199, R567 and 5631 in the amino acid sequence set forth in SEQ ID NO: 1. It is preferred that the modification further includes a modification for introducing a glycochain to a position of an amino acid residue corresponding to R602 in the amino acid sequence set forth in SEQ ID NO: 1, and it is preferred that the introducing a glycochain is performed by amino acid residue substitutions of R602N, D603A, A604T in the amino acid sequence set forth in SEQ ID NO: 1. It is preferred that the modification further includes a modification for introducing a glycochain to a position of an amino acid residue corresponding to D199 in the amino acid sequence set forth in SEQ ID NO: 1, and it is preferred that the introducing a glycochain is performed by amino acid residue substitutions of D199N, D200A and H201T in the amino acid sequence set forth in SEQ ID NO: 1.

(32) The method of introducing a glycochain is not particularly limited as long as it is a conventional method. For example, when an N-type glycochain is introduced, a cDNA of wild-type gB protein (GenBank: M15118.1, SEQ ID NO: 4) is used as a template, and the primer is designed such that the three continuous amino acid sequences of the site of interest at which the N-type glycochain is introduced become N-X-S/T (X is any amino acid other than proline), then a mutation is introduced by PCR. Examples of the mutations for introducing a glycochain include the following mutations in the amino acid sequence set forth in SEQ ID NO: 1: (D199N, D200A, H201T), (R567N, P568S, G569S), (S631N, H632A, Q633T), and (S631N, H632T, Q633T). The nucleic acid sequence of the modified gB protein of interest, or the nucleic acid sequence further linked to a tag such as 6×His as required, can be cloned into an appropriate vector, then expressed to acquire a gB variant. Then, an N-type glycochain is added to asparagine at the site of interest of the gB variant by a conventional method.

(33) The modification of the non-neutralizing epitope may further include a substitution of a charged amino acid residue that is an epitope, with a non-characteristic amino acid residue, e.g., an alanine residue. This method, unlike the introduction of an N-type glycochain, has an advantage that it can achieve de-epitoping by pinpoint. For example, it is preferred that the alanine substitution includes a substitution of an amino acid residue corresponding to an arginine at position 613 (R613) in the amino acid sequence set forth in SEQ ID NO: 1 with alanine.

(34) It is preferred that the modified HSV gB protein is a gB protein in which a glycochain is introduced into each of the amino acid residues corresponding to D199, R567 and 5631 in the amino acid sequence set forth in SEQ ID NO: 1, and the amino acid residue corresponding to an arginine at position 613 (R613) in the amino acid sequence set forth in SEQ ID NO: 1 is substituted with alanine. As another example, it is further preferred that the modified HSV gB protein is a gB protein in which a glycochain is introduced into each of the amino acid residues corresponding to D199, R567, R602 and 5631 in the amino acid sequence set forth in SEQ ID NO: 1, and the amino acid residue corresponding to an arginine at position 613 (R613) in the amino acid sequence set forth in SEQ ID NO: 1 is substituted with alanine.

(35) The modified HSV gB protein of the present invention can be produced by genetic engineering methods. Production methods thereof are not particularly limited, but examples thereof include a production method including: using the cDNA of the wild-type gB protein as a template; designing a primer to introduce the mutation of interest; obtaining a nucleic acid into which the mutation has been introduced by PCR; functionally linking the nucleic acid to an expression promoter, and optionally a tag; and introducing the resultant into an appropriate expression vector to express a modified HSV gB protein. Alternatively, the modified HSV gB protein by introducing a glycochain can be obtained as described above.

(36) The produced modified HSV gB protein may be purified as needed. Purification methods are not particularly limited, but examples thereof include purifications with affinity chromatography column.

(37) HSV infections include infections by HSV-1 and HSV-2, and examples thereof include lip herpes, corneal herpes, genital herpes, systemic neonatal herpes, and stomatitis, skin diseases, encephalitis, meningitis and myelitis due to HSV.

(38) The HSV vaccine of the present invention contains the modified HSV gB protein of the present invention.

(39) Examples of the dosage form of the HSV vaccine of the present embodiment include, liquid, powder (lyophilized powder, dried powder), capsule, tablet, and frozen form.

(40) The HSV vaccine of the present embodiment may contain a pharmaceutically acceptable carrier. As the carrier, carriers commonly used in vaccine production can be used without limitation, and specific examples thereof include saline, buffered saline, dextrose, water, glycerol, isotonic aqueous buffers, and combinations of these. The vaccine may further contain emulsifiers, preservatives (e.g., thimerosal), isotonic agents, pH adjusting agents, and the like.

(41) It is also possible that the HSV vaccine of the present embodiment further contains an adjuvant to further enhance the immunogenicity. Examples of the adjuvant include aluminum adjuvants; oil-in-water emulsion adjuvants containing squalene (AS03, MF59, or the like), ligands of Toll-like receptors such as CpG and 3-O-deacylated-4′-monophosphoryl lipid A (MPL); saponin-based adjuvants; polymer-based adjuvants such as polyγ-glutamic acid; and polysaccharides such as chitosan and inulin.

(42) The HSV vaccine of the present embodiment can be obtained by mixing a modified HSV gB protein of the present invention with a carrier, adjuvant, or the like, as needed. The adjuvant may be mixed when used.

(43) Examples of the administration route of the HSV vaccine of the present embodiment include transdermal administration, sublingual administration, ophthalmic administration, intradermal administration, intramuscular administration, oral administration, enteral administration, nasal administration, intravenous administration, subcutaneous administration, intraperitoneal administration, and oral to pulmonary inhalation administration.

(44) Examples of the administration method of the HSV vaccine of the present embodiment include a syringe, a transdermal patch, a microneedle, an implantable sustained-release device, a syringe with microneedle, a needleless device, and a spray.

EXAMPLES

(45) Hereinafter, the present invention is described in more detail based on examples. However, the present invention is not limited to the following examples.

Example 1 Isolation of Anti-gB Antibodies

(46) The cDNA (SEQ ID NO: 4) of gB ectodomain 1-705aa (gB1-705) derived from a wild-type HSV-2 333 strain was cloned into pCAGGS1-dhfr-neo. The gB was designed so that StrepTag II was added to the C-terminal. For expression, a FreeStyle 293 or Expi293 expression system (Life Technologies) was used. Expression plasmids were transfected into cells and culture supernatants were collected at 4-6 days. The culture supernatants containing gB were concentrated with UF membranes to remove biotin contained in the medium. The concentrated culture supernatant was purified with a StrepTactin column to acquire purified gB2. The gene encoding amino acids from 370 to 457aa of gB were amplified by PCR and cloned into pET43.1b(+) to construct a gB370-457 expression plasmid. To facilitate purification, a Nus-Tag gene and a His-Tag gene were added to this plasmid. The plasmid was transformed into Rosetta2 (Novagen). After culturing the transformant in LB medium at 37° C. until the mid-log growth period, expression was induced with 1 mM IPTG and culturing was performed at 25° C. overnight. For the extraction of soluble proteins, BugBuster mix was used. The obtained protein was purified with Ni NTA Agarose (QIAGEN).

(47) The scFv-phage display library prepared using human VH and VL cDNAs from human B cell-derived mRNA from tonsils and spleens was screened to isolate 44 scFv clones having reactivity with HSV-2 gB. The DNA nucleotide sequences of the VH and VL chain genes of the isolated scFv gene were determined with Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Each clone had an independent sequence and was expected to have variations in its epitopes. Multiple scFv clones, except for 44 clones, were obtained, but most of which had sequences similar to D1 or D2 (data not shown). D1 and D2 are the VH5 family and their antibody populations in organisms are small.

Example 2 Classification of Anti-gB Antibodies

(48) <Preparation of Antibody Fragment scFv-hFc>

(49) The variable region of the isolated scFv gene was linked to human Fc gene, and cloned into a pCAG vector to construct a scFv-hFc expression plasmid. For expression, a FreeStyle 293 or Expi 293 expression system was used. Expression plasmids were transfected into cells and culture supernatant was collected at 4-6 days. From the culture supernatant, scFv-hFc was purified using Ab Rapid Pure 10 (ProteNova Co., Ltd). Furthermore, E. coli TG1 strains with phagemid vectors in which each scFv gene was cloned were cultured in 2×YTCG medium (37° C.). After M13K07 helper phage was infected with moi=20, phage expression was performed in 2×YTCK medium (25° C.) overnight. The obtained scFv-phages were concentrated with 20%-PEG-2.5M NaCl.

(50) The binding activities of the acquired scFv-phage and scFv-hFc were evaluated by ELISA. The gB1-705 was diluted with PBS to 1 μg/mL, then 50 μL of the dilution was placed in Maxisorp plate (Nunc) and incubated at 4° C. overnight or at room temperature for 1-2 hours to make the recombinant gB immobilized. After the immobilization, the plate was washed with PBS, and 100 μL of the acquired scFv-phage, IgG was added to wells of the plate and incubated at 37° C. After 1 hour, the plate was washed with PBST, and 100 μL of the detection antibody anti-M13/HRP (GE Healthcare) or anti-hFc/HRP (Cosmo Bio Co., Ltd.) was added to the wells of the plate and incubated at 37° C. After 1 hour, the plate was washed with PBST and colored by adding 100 μL of TMB to the wells of the plate. After 30 minutes, the reaction was stopped with 1N sulfuric acid, and absorbance (O.D. 450 nm/650 nm) was measured with a microplate reader (Molecular Devices, LLC.) (data not shown).

(51) <Classification of Anti-gB Antibodies by Western Blotting Using gB and its Protease Cleavage Fragments>

(52) To perform grouping of the obtained 44 anti-gB antibody clones, western blotting was performed using gB fragments obtained by digesting gB1-705 with trypsin or chymotrypsin, gB1-705 and cleavage variants thereof as materials.

(53) 500 ng of denatured or non-denatured gB1-705 was loaded to 8-16% SDS-PAGE and electrophoresed. The denatured gB1-705 was obtained by adding 2% 2-mercaptoethanol to gB1-705 and boiling the mixture at 96° C. for 5 minutes. The non-denatured gB1-705 did not undergo these procedures and was directly loaded. The gB fragments digested with trypsin or chymotrypsin were obtained as follows. To 0.5 M Tris-HCl (pH 8.0), 500 ng of gB1-705 and 1 ng, 5 ng or 1000 ng of trypsin, or 1 ng of chymotrypsin were added, and the mixture was stood still at 37° C. for 1 hour or 3 hours to be treated enzymatically, and SDS-PAGE loading buffer with 2% 2-mercaptoethanol (FUJIFILM Wako Pure Chemical Corporation) was added and the mixture was boiled at 96° C. for 5 minutes to stop the reaction. Each sample was loaded to 8-16% SDS-PAGE and electrophoresed, the gel was transferred to a nitrocellulose membrane (Millipore) and blocked with 2% fat-free milk/PBS-T. After washed with PBS-T, the membrane was reacted with each scFv-hFc at a concentration of 1 μg/mL 2% fat-free milk-PBS-T at room temperature for 30 minutes. After washed again, the membrane was reacted with anti-hFc/HRP in 2% fat-free milk-PBS-T, and colored with Immobilon Western Detection Regent (Millipore). Silver staining was performed to detect SDS-PAGE bands.

(54) The result is shown in FIG. 1. Samples for each lane of SDS-PAGE and Western Blotting in FIG. 1 are as follows. In SDS-PAGE, lane 1: non-reduced, non-denatured gB1-705; lane 2: boiled only gB1-705; lane 3: gB1-705 boiled and denatured with a denaturing agent; lane 4: gB370-457 boiled and denatured with a denaturing agent; lane 5: gB1-705 treated with 1 ng (1:500) of trypsin at 37° C. for 1 hour, boiled and denatured with a denaturing agent; lane 6: gB1-705 treated with 5 ng (1:100) of trypsin at 37° C. for 3 hours; lane 7: gB1-705 treated with 100 ng (1:5) of trypsin at 37° C. for 3 hours, boiled and denatured with a denaturing agent; lane 8: gB1-705 treated with 1 ng (1:500) of chymotrypsin at 37° C. for 1 hour, boiled and denatured with denaturing agent; lane M: BenchMark prestained Ladder; and lane M′: Magic Western standard.

(55) In analysis of non-reduced, non-boiled gB1-705 with SDS-PAGE, a band near about 300 kDa was detected (lane 1). Meanwhile, a band of boiled gB1-705 was detected near about 100 kDa (lane 2). This refers to trimers and monomers of gB, respectively, suggesting that the trimer is not formed via S—S bond because boiling changed trimers into monomers.

(56) Treatments with trypsin or chymotrypsin respectively gave fragmented gBs. The reactivity of these gBs with antibody E7 scFv-hFc was detected by Western blotting. E7 scFv-hFc showed reactivity with non-reduced, non-boiled gB1-705 (lane 1), boiled gB1-705 (lane 2), reduced, boiled gB1-705 (lane 3), and protease-treated gB fragments (lanes 5, 6, 7) (FIG. 1). Similar reaction patterns were observed in antibodies E17 and E31 (Table 1). These suggest that antibodies E7, E17 and E31 scFv-hFc are continuous epitopes. Antibodies A17, D1, D2, D3, D37, D48, and E15 scFv-hFc showed reactivity with non-reduced, non-boiled gB1-705, boiled gB1-705, reduced, boiled gB1-705, and protease-treated gB fragments, although their reactivity with gB fragments showed a different pattern from those of E7, E17, E31 (Table 1). The reaction pattern between D1 and D2 was the same, and the reaction pattern between D3 and D37 was the same, suggesting that they recognize the same or proximal epitope, respectively.

(57) The reactivity pattern of gB fragments were able to be classified into six groups. E41, F13, F18, F19, F22, F30 and F78 scFv-hFc showed reactivity only with non-reduced, non-boiled gB1-705 (Table 1). These antibody clones are trimeric-specific, suggesting that they are discontinuous epitopes. F7, F11, F12, F33, F52, F65, F67, F68, F69, F76, F80, F87, G39, G64, G76, H15, H34, H57, H61, H65, G10, G25 and G65 showed reactivity with boiled gB1-705 as well as non-reduced, non-boiled gB1-705 (Table 1). It is suggested that these antibody clones are not trimeric-specific, but discontinuous epitopes recognizing confirmation to some extent. E8, E35, E82, and E88 showed reactivity with non-reduced, non-boiled gB1-705, boiled gB1-705, and reduced boiled gB1-705, but their reactivity with cleavage fragments by proteases could not be confirmed (Table 1). It is suggested that these antibody clones are continuous epitopes because they were reactive with completely denatured gB, but the region may be susceptible to proteases. With analysis by Western blotting using the above gB and gB fragments, it was revealed that the acquired 44 anti-gB antibody clones are classified into nine groups. Of these, it was suggested that 30 clones are discontinuous epitopes and 14 clones are continuous epitopes.

(58) TABLE-US-00001 TABLE 1 Reaction patterns of gB1-705 and its fragments with anti-gB antibodies Protease-treatment ※2 2ME− 2ME− 2ME+ Fragment Fragment Fragment Fragment Fragment Fragment boil− boil+ boil+ pattern A pattern B pattern C pattern D pattern E pattern F Clone ※1 ※1 ※1 ※1 ※1 ※1 ※1 ※1 ※1 A17 + + + + − − − − − D1 + + + − + − − − − D2 + + + − + − − − − D3 + + + − − + − − − D37 + + + − − + − − − D48 + + + − − − + − − E7 + + + − − − − + − E8 + + + − − − − − − E15 + + + − − − − − + E17 + + + − − − − + − E31 + + + − − − − + − E35 + + + − − − − − − E41 + − − − − − − − − E82 + + + − − − − − − E88 + + + − − − − − − F7 + + − − − − − − − F11 + + − − − − − − − F12 + + − − − − − − − F13 + − − − − − − − − F18 + − − − − − − − − F19 + − − − − − − − − F22 + − − − − − − − − F30 + − − − − − − − − F33 + + − − − − − − − F52 + + − − − − − − − F65 + + − − − − − − − F67 + + − − − − − − − F68 + + − − − − − − − F69 + + − − − − − − − F76 + + − − − − − − − F78 + − − − − − − − − F80 + + − − − − − − − F87 + + − − − − − − − G10 + + − − − − − − − G25 + + − − − − − − − G39 + + − − − − − − − G64 + + − − − − − − − G65 + + − − − − − − − G76 + + − − − − − − − H15 + + − − − − − − − H34 + + − − − − − − − H57 + + − − − − − − − H61 + + − − − − − − − H65 + + − − − − − − − ※1 +: with reactivity, −: without reactivity, 2ME+: 2 mercaptcethanol-treated, boil+: boiled ※2 Protease-treatment was performed with trypsin and chymotrypsin. Reactivity patterns of gB fragments with antibodies were classified into six patterns A, B, C, D, E and F.

(59) <Classification of Acquired Antibodies by Competitive ELISA>

(60) To further perform grouping of the acquired 44 antibody clones, competitive ELISA using scFv-phage and scFv-hFc of each anti-gB clone was performed.

(61) Competition between the acquired scFv-phage and scFv-hFc was evaluated by ELISA. The gB1-705 was diluted with PBS to 1 μg/mL, then 50 μL of the dilution was placed in Maxisorp Plate, and incubated at 4° C. overnight to make gB1-705 immobilized. After the immobilization, the plate was washed with PBS, and 50 μL of the acquired scFv-hFc was added to the wells of the plate and incubated at 25° C. After 1 hour, 50 μL of scFv-phage was added to the wells of the plate and incubated at 25° C. After 1 hour, the plate was washed with PBST, and 100 μL of the detection antibody anti-M13/HRP was added to the wells of the plate and incubated at 25° C. After 1 hour, the plate was washed with PBST and colored by adding 100 μL of TMB to the wells of the plate. After 30 minutes, the reaction was stopped with 1N sulfuric acid, and absorbance (O.D. 450 nm/650 nm) was measured with a microplate reader.

(62) Based on the results, a correlation diagram of the antibodies was made (FIG. 2). The group including antibodies E41, F18, F19, F13, F78, F22 and F30 was a group showing reactivity only with non-reduced, non-boiled gB1-705, but unlike other groups, it was confirmed that this group was classified into a group of antibodies E41, F18, F19, F13 and F78 and a group of F22 and F30, based on the results of competitive ELISA. Meanwhile, the other groups were not further subdivided from the groups described above. Thus, anti-gB clones were classified into a total of 10 groups. It was also suggested that an epitope is present within 370aa to 457aa of gB because D48 was reactive with gB370-457.

Example 3 Identification of Epitopes of gB Antibodies by Alanine Scanning

(63) The genes in which each charged amino acid residue (187 locations) in the ectodomain of gB, 1-705aa, was modified with alanine were constructed by PCR and cloned into pCAGGS1-dhfr-neo. For expression, a FreeStyle 293 or Expi 293 expression system was used. The expression level of the acquired gB alanine substitute and the binding activity of the alanine substitute to the antibody fragment were evaluated by ELISA.

(64) The culture supernatant containing the gB alanine substitute was placed in a Maxisorp plate and incubated at room temperature for 1 hour to make the gB alanine substitute immobilized. After the immobilization, the plate was washed with PBST, and 100 μL of the detection antibody StrepTactin/HRP (IBA) was added to wells of the plate and incubated at room temperature. After 1 hour, the plate was washed with PBST and colored by adding 100 μL of TMB to the wells of the plate. After 30 minutes, the reaction was stopped with 1N sulfuric acid, and absorbance (O.D. 450 nm/650 nm) was measured with a microplate reader to determine expression level.

(65) Meanwhile, the culture supernatant containing the gB alanine substitute was placed in a Streptactin-immobilized Maxisorp plate and incubated at room temperature for 1 hour to make the gB alanine substitute immobilized. After the immobilization, the plate was washed with PBST and 100 μL of antibody fragments were added to wells of the plate and incubated at room temperature. After 1 hour, the plate was washed with PBST, and 100 μL of the detection antibody anti-human Fc/HRP (Cosmo Bio Co., Ltd.) was added to the wells of the plate and incubated at room temperature. After 1 hour, the plate was washed with PBST and colored by adding 100 μL of TMB to the wells of the plate. After 30 minutes, the reaction was stopped with 1N sulfuric acid, color value (O.D. 450 nm/650 nm) was measured with a microplate reader to determine binding activity. Epitope candidates were selected based on whether the reactivity per expression level was changed compared to that of wild-type gB1-705, non-alanine substitute. In addition, in the case of antibodies requiring scrutiny, epitope candidates were narrowed down by ELISA using purified gB alanine substitutes.

(66) Based on the binding activity of alanine substitutes and antibody fragments, the epitopes of each antibody were estimated. The results are shown in Table 2. There were 28 clones of antibodies to domain IV forming Crown portion of the gB trimer, which were most abundant. Subsequently, there were 8 clones of antibodies to domain I forming Bottom portion, and there was only one clone of antibody to domain II forming Middle portion. In the antibodies for which epitopes were able to be identified, many were at distant positions in the amino acid sequence. Although gB of HSV2 was modeled in MOE (FIG. 3) using gB of HSV1 as a template and their positions were confirmed, they were structurally proximate amino acids to each other (data not shown). For only D48 that recognizes domain II, R391 and D362 were identified, but since D48 reacts with gB37-457, R391 is likely to be the key of the epitope. The epitopes of 7 clones (E82, E88, F7, F65, F80, G10, G76) in 44 clones were not achieved to be identified. However, since there are clones, whose epitopes were able to be identified, in the same group by grouping of previous competitive ELISA or the like, it is likely that amino acid residues in the vicinity are their epitopes. For example, epitopes of E82 and E88 could not be identified, but it is known that the epitopes of E8 and E35 are R532, R613, H632. Since E82 and E88 compete with E8 and E35 in competitive ELISA, the epitopes of E82 and E88 are expected to be in the vicinity of R532, R613, H632. The antibodies whose epitopes were unable to be identified are considered that they likely have epitopes of amino acid residues other than charged amino acids from a viewpoint that this analysis is limited to charged amino acid residues.

(67) TABLE-US-00002 TABLE 2 Identification of epitopes of anti-gB antibodies by alanine scanning Clone Epitope Domain A17 D172 K177 R308 I D1 R308 I D2 R308 I D3 K45 K46 K49 R50 E52 VI D37 K45 K46 K49 R50 E52 VI D48 R391 D362 II E7 D583 IV E8 R613 R537 H632 IV E15 R277 E278 H281 I E17 D583 D554 R563 R580 IV E31 D583 R563 D554 R277 R379 IV E35 R532 H632 R379 IV E41 R458 R277 E278 H281 I E82 ND IV E88 ND IV F7 ND IV F11 D583 D554 R563 D258 IV F12 D583 R602 IV F13 D200 H201 E259 I F18 D200 D224 K306 I F19 D199 D200 H201 I F22 H632 IV F30 D647 H632 IV F33 D583 D554 R563 IV F52 D583 D554 R563 IV F65 ND IV F67 E627 E626 R613 H632 IV F68 D583 IV F69 R580 D583 D554 IV F75 R567 IV F78 D199 D200 H201 K401 I F80 ND IV F87 H632 IV G10 ND IV G25 R567 IV G39 H632 IV G64 D554 R563 D583 IV G65 H632 IV G76 ND IV H15 D583 D554 R563 IV H34 D583 D554 R563 IV H57 D583 R498 IV H61 D583 D554 R563 IV H65 R580 IV ND: Not detected

Example 4 Analysis of the Presence of Neutralizing and Non-Neutralizing Epitopes in Crown and Bottom Portions of gB by Neutralization Test

(68) <Cells and Viruses>

(69) Vero cells purchased from ATCC (CCL.81) were used for viral culture, measurement of infectivity titer, and measurement of neutralizing antibody titer. Human herpes virus 2 (HSV-2) MS strains [VR-540] were used for neutralization test and infection prevention ability analysis. For HSV-1, KOS strain (VR-1493) were purchased from ATCC and used. Vero cells were cultured in 10% FBS-containing MEM medium until full-sheeted, and the HSV-2 MS strain and HSV-1 KOS strain were inoculated at m.o.i=0.01-1. After culturing the cells in 2% FBS-containing MEM medium for 2-3 days, viruses in the cells were released into the medium by three freeze-thaws. After centrifugation, supernatants were collected and used as HSV-2 virus bank and HSV-1 virus bank.

(70) <Viral Neutralization Test>

(71) The neutralization activity of the acquired anti-gB clones was evaluated by viral neutralization test. The neutralization test was performed using two methods, a measurement of plaque number-reducing activity (plaque reduction activity) and a measurement of cell-to-cell infection spread-suppression activity. Two types of virus of interest, HSV-2 MS strain and HSV-1 KOS strain, were used. For the measurement of plaque reduction activity, a predetermined concentration of the subject antibody was prepared and mixed with about 100 PFU of HSV-2 MS strains or HSV-1 KOS strains, then reacted at 37° C. for 1 hour. The reaction solution was plated to Vero cells full-sheeted in a 48-well plate, and after adsorption at 30° C. for 1 hour, cultured in 1% methylcellulose-containing medium for 24 hours, then inactivation and immobilization were performed at −20° C. for 30 minutes with 50% methanol/50% ethanol (−20° C.) obtained by mixing methanol and ethanol at 1:1. The anti-HSV gB monoclonal antibody was then reacted at 37° C. for 1 hour, the resultant was immunostained with anti-mouse IgG-HRP (Dako P0447) and TMBH (MOS TMBH-1000), and images of each well were captured with an ELISpot analyzer to count plaque numbers with analytical software.

(72) For the measurement of cell-to-cell infection spread-suppression activity, about 100 PFU of HSV-2 MS strain or HSV-1 KOS strain were inoculated into Vero cells full-sheeted in a 48-well plate, and adsorption was performed at 30° C. for 1 hour. Thereafter, a predetermined concentration of the subject antibody in 1% methylcellulose medium (antibody concentrations was 5 μg/mL, 25 μg/mL, and 125 μg/mL) was added. The HSV-2 MS strain was cultured for about 40 hours and the HSV-1 KOS strain was cultured for about 48 hours, then inactivation and immobilization were performed at −20° C. for 30 minutes with 50% methanol/50% ethanol (−20° C.). Anti-HSV gB monoclonal antibodies were then reacted at 37° C. for 1 hour, then the resulting cells were immunostained with anti-mouse IgG-HRP (Dako P0447) and TMBH, and images of each well were captured with an ELISpot analyzer to analyze average plaque size values with analytical software.

(73) The results are shown in Table 3. The acquired 44 clones were categorized into two of neutralizing and non-neutralizing antibodies. Among the neutralizing antibodies, there were clones that neutralize both HSV-1 and HSV-2, type-specific clones that neutralize only one of them, clones with strong plaque number-reducing activity but no (weak) cell-to-cell infection spread-suppression activity, clones with strong both plaque number-reducing activity and cell-to-cell infection spread-suppression activity, and the like. It was also confirmed that this neutralization pattern is linked to the previous grouping results (Table 3, FIG. 2).

(74) TABLE-US-00003 TABLE 3 Neutralization activities of anti-gB antibodies HSV-1 KOS strain HSV-2 MS strain Plaque Cell-to-cell Plaque Cell-to-cell number spread number spread Clone reduction suppression reduction suppression A17 ++ ++ ++ N.T. D1 ++ ++ − − D2 ++ ++ − − D3 − − ± ± D37 − − ± ± D48 ++ ++ ++ ++ E7 ++ ++ ++ ++ E8 − − − − E15 ++ ++ ++ ++ E17 ++ − ++ ++ E31 ++ − ++ ++ E35 ± − ± − E41 ++ ++ ++ ++ E82 − − − − E88 ++ N.T. ++ ++ F7 − − − − F11 + + + + F12 + + + + F13 − − − − F18 − − − − F19 − − − − F22 − − − − F30 − − − − F33 + + + + F52 + − + + F65 − − − − F67 − − − − F68 − − − − F69 + + + + F76 − − − − F78 − − − − F80 − − − − F87 − − − − G10.3 ± − ± N.T. G25 − − − N.T. G39 − − − N.T. G64 ++ ++ ++ N.T. G65 + + + N.T. G76 + + + N.T. H15 ++ + ++ N.T. H34 ++ + ++ N.T. H57 ++ + ++ N.T. H61-1 ++ + ++ N.T. H65 ++ + ++ N.T. N.T.: Not tested

(75) The plaque number-reducing activity was analyzed at 5 μg/mL, 25 μg/mL, and 125 μg/mL, and those with neutralizing ability at up to 5 μg/mL were determined to be “++”, those with neutralizing ability at up to 25 μg/mL were determined to be “+”, and those with neutralizing ability at up to 125 μg/mL were determined to be “±”. The cell-to-cell infection spread-suppression was analyzed at 5 μg/mL, 10 μg/mL, and 20 μg/mL, and those with neutralizing ability at up to 5 μg/mL were determined to be “++”, those with neutralizing ability at up to 10 μg/mL were determined to be “+”, and those with neutralizing ability at up to 20 μg/mL were determined to be “±”. All measurements were performed in duplicate.

(76) From Table 2, antibodies to domains IV and I on the gB2 antigen accounted for a large percentage to the total antibody. The antibodies to domains IV and I were partially with neutralizing activity, but mostly non-neutralizing antibodies that did not exhibit neutralizing activity (Table 3). In contrast, antibodies to domain II accounted for a small percentage to the total antibody, but they were neutralizing antibodies. In the MOE diagram (FIG. 3) which is the identification result of epitopes of each gB antibody, positions of these epitopes are illustrated. From these results, it was revealed that neutralizing epitopes and non-neutralizing epitopes are present in domains IV and I, and neutralizing epitopes are present in domain II, and that many “conspicuous” antigen epitopes are present in domains IV and I.

Example 5 Design and Evaluation of Vaccine Antigens

(77) <Design Strategy of Modified gB Antigen>

(78) Comprehensive acquisition of anti-HSV2-gB antibodies and their epitope mapping and functional classification were performed. As the result, it was found that neutralizing epitopes and non-neutralizing epitopes are present in domain IV and domain I, and neutralizing epitopes are present in domain II in the gB2 antigen. From the viewpoint of preventive activity expression, modified gB antigens were designed by setting the neutralizing epitopes as beneficial epitopes, and the non-neutralizing epitopes as unbeneficial or deleterious epitopes, and de-epitoping the non-neutralizing epitopes present in gB domains I and IV. The base wild-type gB was the ectodomain gB1-705, and Streptag II was added to the C-terminal side for StrepTactin purification (FIG. 4(A)).

(79) First, the modification of gB by N-type glycosylation was considered. The N-type glycochain, unlike 0-type glycochain, is added to a consensus sequence, NXT or NXS (X is any amino acid other than proline). Antibodies are generally difficult to be generated against glycopeptides, and due to the bulkiness of the glycochain, antibodies are also difficult to be generated against peripheries of the glycopeptides (Non Patent Literature 16). Domain IV of HSV2-gB is thought to be an important region for binding to the receptor, but is furthest away from the surface of the viral membrane and is exposed to the surface so that it is susceptible to antibody binding. Indeed, as shown in the result described above, it has been suggested that many antibodies that recognize domain IV are contained in human serum. Thus, it was decided to introduce an N-type glycochain at three locations of the non-neutralizing epitopes in domain IV (FIGS. 4(B) and 5). Meanwhile, domain I is located at the root of gB and is an important region for fusion with the host cell, but when only the ectodomain is expressed, the region that has been originally in contact with the viral membrane surface is exposed to the surface. In addition, the vaccine antigen design using only ectodomain of gB is assumed in the manufacture of vaccines. There is a possibility that a region in contact with the viral membrane may be newly exposed, thus it was decided to introduce an N-type glycochain at one location of the non-neutralizing epitope region (FIGS. 4(B) and 5). The glycochains originally attached to the wild-type gB appear to be positioned at N115, N371, N649 with reference to HSV-1 gB (PDB No. 3 NWF) (FIG. 4(B)). Thus, D199, R567, R602, S631 were newly selected as positions for introducing a glycochain.

(80) Next, the modification of gB by alanine substitution was considered. Charged amino acid residues are often included as epitopes of antibodies. Thus, a method for de-epitoping by substituting a charged amino acid residue with a non-characteristic amino acid residue is sometimes taken. This method, unlike the introduction of an N-type glycochain, has an advantage that it can achieve de-epitoping by pinpoint. Not only non-neutralizing epitopes but also neutralizing epitopes are present in domain IV, thus it was decided to perform alanine substitution at one location (R613A) to be performed (FIGS. 4(B) and 5).

(81) <Preparation of Modified gB1-705>

(82) The cDNA (SEQ ID NO: 4) of gB ectodomain (1-705aa) derived from wild-type HSV-2 333 strain was cloned into pCAGGS1-dhfr-neo. The gB was designed so that Streptag II is added to the C-terminal. Using this sequence as a template, variants in which the following mutations were introduced were designed:

(83) Modified product bcev1-3: D199N, D200A, H201T

(84) Modified product bceg13: R567N, P568S, G569S

(85) Modified product bcev11: D199N, D200A, H201T, R613A

(86) Modified product bcev12: D199N, D200A, H201T, R567N, P568S, G569S, R613A

(87) Modified product bcev13: D199N, D200A, H201T, R567N, P568S, G569S, R613A, S631N, H632A, Q633T

(88) Modified product bcev19: D199N, D200A, H201T, R567N, P568S, G569S, R613A, S631N, H632T, Q633T

(89) Modified product bcev19′: R567N, P568S, G569S, R613A, S631N, H632T, Q633T

(90) Modified product bcev50: D199N, D200A, H201T, R567N, P568S, G569S, R613A, R602N, D603A, A604T, S631N, H632T, Q633T

(91) Modified product bcev50′: R567N, P568S, G569S, R613A, R602N, D603A, A604T, S631N, H632T, Q633T

(92) For expression, a FreeStyle 293 or Expi 293 expression system was used. Expression plasmids were transfected into cells and culture supernatants were collected at 4-6 days. The culture supernatant containing gB was concentrated with a UF membrane to suppress the effects of biotin contained in the medium. The concentrated culture supernatant was purified with a StrepTactin column to acquire purified gB.

(93) For properties of the acquired purified gB1-705, the multimeric state thereof was determined by SDS-PAGE and gel filtration chromatography. In the gel filtration chromatography, Superdex 200 Increase 5/150 GL (GE Healthcare) was used as the column, and each modified gB purified product was applied at a concentration of 100 μg/mL. D-PBS was used as the electrophoresis buffer at a flow rate of 0.4 mL/min to detect A280.

(94) Table 4 shows the modification sites and characteristics of the six modified gB s.

(95) TABLE-US-00004 TABLE 4 Expression Artificial level Multimeric glycochain Clone Mutation site (μg/mL) state introduction gB1-705 — 2.25 Trimer 0 (Wild- type) bcev1-3 D199N, D200A, H201T 4.18 Trimer 1 bceg13 R567N, P568S, G569S 0.64 Trimer 1 bcev11 D199N, D200A, H201T, 2.04 Trimer 1 R613A bcev12 D199N, D200A, H201T, 4.01 Trimer 2 R567N, P568S, G569S, R613A bcev19 D199N, D200A, H201T, 1.38 Multimer 3 R567N, P568S, G569S, Trimer R613A, S631N, H632T, Q633T bcev50 D199N, D200A, H201T, 0.35 Multimer 4 R567N, P568S, G569S, Trimer R613A, R602N, D603A, A604T, S631N, H632T, Q633T N.T.: Not tested

(96) The bcev1-3 to which D199N, D200A, H201T modifications were added is a gB modified to which a mutation of introducing a glycochain into domain I was added, and the expression level thereof is enhanced approximately 2-fold compared to that of wild-type gB1-705. It is believed that this region is originally stabilized in a state contact with the surface of the viral membrane. It is presumed that the wild-type gB1-705 is destabilized due to secretion expression which leads to the exposure of this region that is not otherwise solvent contacted, then by the addition of a glycochain to the region, bcev1-3 is stabilized to increase the expression level.

(97) The bceg13 is a gB to which modifications of R567N, P568S, G569S were added in domain IV. In contrast to bcev1-3, the expression level thereof is 0.3-fold compared to that of wild-type gB1-705. Since the trimeric structure similar to that of the wild-type is maintained, it is suggested that mutation is not a critical mutation to maintain the structure, but the single-chain structural changes and folding thereof is slower than those of the wild-type.

(98) The bcev11 is a variant in which, in addition to the D199N, D200A, H201T mutations in bcev1-3, R613A positioned in domain IV were introduced. The expression level of bcev11 was equivalent to that of wild-type gB1-705.

(99) Given that the bcev1-3 mutation enhanced the expression level, it is suggested that the R613A mutation is a contrary mutation that makes expression level reduced or equal to that of the wild-type.

(100) The properties of bcev11 were also equivalent to those of the wild-type and a trimer thereof was formed.

(101) The bcev12 is a gB in which further modifications of R567N, P568S, G569S were added to bcev11. Based on the result of bceg13, the addition of R567N, P568S, G569S mutations should have reduced the expression level, but expression level thereof unexpectedly doubled compared to that of the wild-type. This expression level is equivalent to that of bcev1-3 in which only modifications of D199N, D200A, H201T were added. It is suggested that R613A mutation contributes to reduction of expression level, but the addition of R567N, P568S, G569S makes the structure and folding rate of domain IV closer to bcev1-3.

(102) The bcev19 is a gB variant in which mutations of S631N, H632T, Q633T were introduced into bcev12. The expression level of bcev19 was 0.6-fold compared to that of wild-type, and not only trimer but also trace amounts of multimer were contained therein. Other mutations to 5631, H632, Q633 or glycosylation to a vicinity location other than 631-633aa were also examined, but no significant improvement effect on the properties was observed (data not shown).

(103) The bcev50 is a gB variant in which mutations of R602N, D603A, A604T were introduced into domain IV of bcev19. The expression level thereof was further reduced compared to that of bcev19 and 0.3-fold compared to that of wild-type. For properties, trimers were contained as in bcev19, but multimers were also contained.

(104) <Binding Activity Test of Modified gB1-705>

(105) To confirm the aim of low or no reactivity with non-neutralizing antibodies, the reactivity of the produced variants bcev1-3, bceg13, bcev12, bcev19 and bcev50 with 44 clones of the anti-gB2 monoclonal antibody was evaluated by ELISA. As a control, wild-type gB1-705 was used. The gB1-705 was diluted with PBS to 1 μg/mL, then 50 μL of the dilution was placed in Maxisorp Plate, and incubated at 4° C. overnight to make gB1-705 immobilized. After the immobilization, the plate was washed with PBS, and 100 μL of the acquired antibody was added to wells of the plate and incubated at 37° C. After 1 hour, the plate was washed with PBST, and 100 μL of the detection antibody anti-human IgG Fc/HRP (Rockland Immunochemicals, Inc.) was added to the wells of the plate and incubated at 37° C. After 1 hour, the plate was washed with PBST and colored by adding 100 μL of TMB to the wells of the plate. After 30 minutes, the reaction was stopped with 1N sulfuric acid, and absorbance (O.D. 450 nm/650 nm) was measured with a microplate reader.

(106) The results are shown in Table 5.

(107) TABLE-US-00005 TABLE 5 Reactivities of the modified gB and acquired antibodies Ab Clone gB1-705 bcev1-3 bceg13 bcev11 bcev12 bcev19 bcev50 Non- F22 + + + + + − − neutralizing F30 + + + N.T. ± − − antibodies F7 + + + N.T. + + − F65 + + + N.T. + ± − F67 + + + ± − − − F68 + + + N.T. + + − F76 + + − N.T. − − − F80 + + + N.T. ± ± − F87 + + + N.T. − − − G39 + + + N.T. ± − − G76 + + + N.T. − ± − G65 + + + N.T. − − − E8 + + + − − − − E35 + + + N.T. + − − E82 + + + N.T. − − − E88 + + + N.T. − − − G25 + + − + − − − F13 + − + − − − − F18 + + + N.T. ± + − F19 + − + N.T. − ± − F78 + − ± N.T. − ± − Neutralizing E7 + + + N.T. + + ± antibodies E17 + + + N.T. + + + E31 + + + + + + − F11 + + + N.T. + + − F12 + + + + + + − F33 + + + N.T. + + ± F52 + + + N.T. + + ± F69 + + + + + + − G64 + + + N.T. + + + H15 + + + N.T. + + − H34 + + + N.T. + + ± H57 + + + N.T. + + ± H61 + + + N.T. + + − H65 + + + N.T. + + − G10 + + + N.T. − − − D3 + + + ± + + + D37 + + + N.T. + + + D48 + + + + + + + E15 + + + + + + + D1 + + + + + + ± D2 + ± + N.T. ± + ± A17 + + + + + + + E41 + + + + + + − N.T.: Not tested

(108) The bcev1-3 was no longer reactive with F13, F19, F78 whose epitope is D200, H201 after mutations of D199N, D200A, H201T were introduced. The bceg13 was no longer reactive with F76, G25 whose epitope is R567 after mutations of R567N, P568S, G569S were introduced.

(109) The bcev12 into which the modifications of bcev1-3 and bceg13 and R613A were introduced was almost no longer reactive with antibodies whose epitopes are each mutation site.

(110) The bcev19 in which mutations of S631N, H632A, Q633T were introduced into bcev12 maintained reactivity with 22 types of neutralizing antibodies, while became less reactive with non-neutralizing antibodies.

(111) Among the 21 types of non-neutralizing antibodies, a total 8 type, F7, F65, F68, F80 and G76 which recognize domain IV and F18, F19 and F78 which recognize domain I, maintained reactivity.

(112) The bcev50 in which mutations of R602N, D603A, A604T were introduced into bcev19 was no longer reactive with all 21 types of non-neutralizing antibodies. Meanwhile, for the 23 types of neutralizing antibodies, it maintained reactivity with 14 types, but became less reactive with 9 types.

(113) These mean a possibility that immunizing the bcev50 makes not only non-neutralizing antibodies difficult to induce, but also neutralizing antibodies slightly difficult to induce. The difference between bcev19 and bcev50 is whether or not there are R602N, D603A, A604T mutations in domain IV. Nevertheless, differences have arisen in reactivity with antibodies to domain I. It is suggested that modification of domain IV keeps the trimer but may change the structure of domain I. In summary, bcev19 and bcev50 are less likely to generate non-neutralizing antibodies compared to wild-type antigen, gB1-705, and can be expected to be novel vaccine antigens that can induce an ideal immune response.

(114) <Mouse Immunogenicity Test of Bcev19 and Bcev50>

(115) Mouse immunogenicity tests of the produced modified gB antigens bcev19 and bcev50 were performed separately. In both experiments, mice were subcutaneously immunized with antigen dose of 0.3 μg/mouse and 1 μg/mouse at 2-week intervals for 3 times. The experiment was performed setting the number of mouse cases in each group as n=4.

(116) An immunogenicity test of the modified gB antigen was performed using the wild-type gD antigen gB1-705 (gB WT) as a positive control and saline as a negative control. The predetermined amount of antigen was dissolved in saline for injection (saline) to immunize BALB/c mouse (5 weeks old, female) subcutaneously in back at a volume of 200 μL/mouse with MPLA (10 μg/mouse) and CpG (1 μg/mouse) at 2-week intervals for a total of 3 times. Two weeks after the final immunization (third dose), blood was collected for each individual and serum was prepared. The prepared serum was serially diluted, and the binding antibody titer to wild-type gB antigen (anti-gB ELISA) and neutralizing antibody titer against HSV-2 (50% plaque number-reducing activity) were evaluated.

(117) The result of bcev19 is shown in FIG. 6 and the result of bcev50 is shown in FIG. 7. In the graph, mean values of n=4 are plotted with ±SE error bars. Using serum collected 2 weeks after the final immunization, the binding antibody-inducing activity against wild-type gB antigen (anti-gB ELISA) and neutralizing antibody-inducing activity against HSV-2 (plaque number-reducing rate) were evaluated. As the result, both bcev19 and bcev50 were confirmed to induce higher neutralizing antibody activity with less binding antibody activity than wild-type gB antigen (gB1-705) at any dose.

(118) The results are believed to be a result of being able to more efficiently and effectively induce an immune response to the remaining neutralizing epitopes (beneficial epitopes) by de-epitoping non-neutralizing epitopes (deleterious and unbeneficial epitopes) in the wild-type gB antigen by introduction of N-type glycosylation and alanine substitution. In other words, a biased immune response (immune deviation) to a wild-type gB antigen can have been ideally corrected (immune correction) by the present inventors' immune refocusing strategies.

(119) <Mouse Infection-Prevention Test of Bcev19 and Bcev50>

(120) An infection prevention ability in the preventive administration of modified gB antigens bcev19 and bcev50 was evaluated separately using a mouse genital herpes infection model. In both experiments, wild-type gB1-705 (gB WT) was used as a positive control. Mice were immunized subcutaneously with 0.03 μg/mouse, 0.1 μg/mouse, 0.3 μg/mouse, and 1 μg/mouse for all antigens at 2-week intervals for 3 times. Depo-Provera was inoculated subcutaneously at 2 mg/mouse 6 days prior to viral inoculation to improve multiplicity of infection upon viral inoculation 2 weeks after the final immunization (third dose). Mouse was inoculated transvaginally with 5×10.sup.5 PFU/20 μL/mouse HSV-2 MS strain under anesthesia and observed for 21 days. The infection prevention ability was evaluated using survival time (survival rate) and symptom score as indexes. Symptom scores were classified by the extent of vaginal lesion symptoms and systemic symptoms, and scores were set for 3 and 2 stages, respectively. The total of the following vaginal lesion and systemic symptom scores were taken as the symptom scores: score for vaginal lesions (0: no change, 1: partial erythema/swelling, 2: extensive swelling/edema, 3: ulceration/bleeding); score for systemic symptoms (0: no change, 1: piloerection, 2: hind limb paralysis). Furthermore, death or sacrifice was scored 6. The experiment was performed setting the number of mouse cases in each group as n=10 and their mean values were plotted in the graph.

(121) The results of bcev19 are shown in Table 6 (survival time), FIG. 8 (survival rate), and FIG. 9 (symptom score). In both of gB WT and bcev19, a significant survival time prolongation effect relative to the negative control group (saline administration group) was shown at all set dosages (0.03 to 1 μg/mouse). However, in the gB WT administration group, median survival time (MST) showed no clear dose dependency and the MST ratio to the saline administration group remained below 2 at any dosage. In contrast, in the bcev19 administration group, MST showed approximately dose dependency, and MST ratios were >2.8 at three doses of 0.1 μg/mouse or more, indicating clear improvement in survival rate (Table 6, FIG. 8). For the symptom score, in the gB WT administration group, there was no clear dose dependency and severe symptoms were accompanied even in the highest dose of 1 μg/mouse. However, in the bcev19 administration group, dose-dependent and significant improvement effects were observed (FIG. 9).

(122) TABLE-US-00006 TABLE 6 Significant difference Dosage Survival MST MST ratio test Antigen (μg) days (days) (days) (vs saline) (vs saline) gB1-705 1 9.9, 10, 11, 11.5 1.9 *** (WT) 11, 12, 13, 14, >21, >21 0.3 6, 7, 8, 8, 9 1.5 ** 8, 10, 10, 11, >21, >21 0.1 8, 8, 8, 9, 9.5 1.6 *** 9, 10, 10, 12, >21, >21 0.03 6, 8, 8, 8, 8 1.3 ** 8, 11, 17, >21, >21 bcev19 1 10, 11, 14, >21 >3.5 *** >21, >21, >21, >21, >21, >21, >21 0.3 5, 11, 12, 12, >16.5 >2.8 *** 12, >21, >21, >21, >21, >21 0.1 6, 9, 9, 12, >17 >2.8 *** 13, >21, >21, >21, >21, >21 0.03 5, 6, 7, 7, 9.5 1.6 * 9, 10, 14, >21, >21, >21 saline 5, 5, 5, 6, 6 6, 6, 7, 7, 8 MST: Mean survival time ***: p < 0.0001/**: 0.0001 < p < 0.001/*: 0.001 < p < 0.01 (Kaplan-Meier method)

(123) The results of bcev50 are shown in Table 7 (survival time), FIG. 10 (survival rate), and FIG. 11 (symptom score). Similar to bcev19, bcev50 also showed a clear superiority to gB WT in any index of survival time, survival rate, and symptom score.

(124) TABLE-US-00007 TABLE 7 Significant difference Dosage Survival MST MST ratio test Antigen (μg) time (days) (days) (vs saline) (vs saline) gB1-705 1 8, 9, 11, 13, 15 2.3 *** (WT) 15, >23, >23, >23, >23 0.3 7, 8, 8, 12, >23 >3.5 ** >23, >23, >23 >23, >23, >23 0.1 7, 7, 7, 9, 9 1.4 ** 9, 9, 9, 9, >23, >23 0.03 6, 6, 6, 6, 7 1.1 N.S. 7, 7, 8, 11, >23, >23 bcev50 1 12, >23, >23, >23 >3.5 *** >23, >23, >23 >23, >23, >23 0.3 >23, >23, >23, >23 >3.5 *** >23, >23, >23, >23, >23, >23, >23 0.1 7, 9, 12, >23, >23 >3.5 *** >23, >23, >23, >23, >23, >23 0.03 5, 6, 6, 9, >17 >2.6 * 11, >23, >23, >23, >23, >23 saline 6, 6, 6, 6, 6.5 6, 7, 7 ,7, 8, 8 MST: Mean survival time ***: p < 0.0001/**: 0.0001 < p < 0.001/*: 0.001 < p <0.01 (Kaplan-Meier method)

(125) <Analysis of Immune Refocusing>

(126) For immune sera of the modified gB antigens bcev19 and bcev50, which were found to be superior to wild-type gB (gB1-705) in the mouse immunogenicity test and the mouse infection-prevention test, whether immune refocusing was induced or not was analyzed by gB1-457 and gB111-457 immobilized ELISAs, respectively.

(127) The analysis result of bcev19 is shown in FIG. 12 and the analysis result of bcev50 is shown in FIG. 13. Both bcev19 immune serum and bcev50 immune serum were confirmed to have increased binding antibody activity against gB1-457 and gB111-457 compared to wild-type gB immune serum. The above results are believed to be a result of being able to more efficiently and effectively induce an immune response to the neutralizing epitopes (beneficial epitopes) remaining a lot in domains I and II by de-epitoping the non-neutralizing epitopes present mainly in domain IV, which is thought to be a decoy region in wild-type gB antigen, by introduction of N-type glycosylation and alanine substitution. In other words, it is believed that a biased immune response (immune deviation) to the decoy region in the wild-type gB antigen can have been ideally corrected (immune correction) by the present inventors' immune refocusing strategies.

(128) <Effect of N-Type Glycochain Introduced into gB Domain I>

(129) In the gB variant, bcev19, modifications of D199N, D200A, H201T are introduced into domain I, and modifications of R613A, R567N, P568S, G569S, S631N, H632T, Q633T are introduced into domain IV. In bcev50, modifications of D199N, D200A, H201T are introduced into domain I, and modifications of R613A, R567N, P568S, G569S, S631N, H632T, Q633T, R602N, D603A, A604T are introduced into domain IV. To further investigate the effect of the modifications introduced into bcev19 and bcev50, variants bcev19′ and bcev50′ in which only modifications D199N, D200A, H201T of domain I contained in bcev19 and bcev50 returned to the amino acid sequences of the original, respectively, were made.

(130) Transient expression using an Expi293 expression system was performed to compare the expression levels (Table 8). As the result, the expression levels of bcev19, bcev19′, bcev50 and bcev50′ were 10.81 μg/mL, 1.28 μg/mL, 6.29 μg/mL, and 4.36 μg/mL, respectively. When comparing the expression levels of bcev19 and bcev19′, bcev19 was 8.45 times higher than bcev19′, and when comparing the expression levels of bcev50 and bcev50′, bcev50 was 1.44 times higher than bcev50′. This means that D199N, D200A, H201T in bcev19 and bcev50 are mutations that contribute to expression level enhancement. This is also supported by similar results obtained in bceg1-3 in which modification of D199N, D200A, H201T were introduced to gB1-705 (Table 4).

(131) TABLE-US-00008 TABLE 8 Artificial Yield glycochain Clone Mutation site (μg/mL) introduction bcev19 D199N, D200A, H201T, R567N, 10.81 3 P568S, G569S, R613A, S631N, H632T, Q633T bcev19′ R567N, P568S, G569S, R613A, 1.28 2 S631N, H632T, Q633T bcev50 D199N, D200A, H201T, R567N, 6.29 4 P568S, G569S, R613A, R602N, D603A, A604T, S631N, H632T, Q633T bcev50′ R567N, P568S, G569S, R613A, 4.36 3 R602N, Q603A, A604T, S631N, H632T, Q633T

(132) Furthermore, property analysis of bcev19, bcev19′, bcev50 and bcev50′ was performed by gel filtration chromatography. The result is shown in FIG. 14. The bcev19 had a higher native trimer content than bcev19′ (FIG. 14(A)) and bcev50 had a higher trimer content than bcev50′ (FIG. 14(B)). This means that D199N, D200A, H201T in bcev19 and bcev50 are mutations that contribute to property improvement.

(133) Next, bcev19 and bcev19′, and bcev50 and bcev50′ were respectively compared in mouse immunogenicity tests. Mice were subcutaneously immunized with antigen dose of 0.3 μg/mouse and 1 μg/mouse at 2-week intervals for 3 times. The experiment was performed setting the number of mouse cases in each group as n=4. A comparison of bcev19 and bcev19′ is shown in FIG. 15, and a comparison of bcev50 and bcev50′ is shown in FIG. 16, respectively (Mean values of n=4 are plotted with ±SE error bars in the graph). Using serum collected 2 weeks after final immunization, binding antibody-inducing activity against wild-type gB antigen (anti-gB ELISA) and neutralizing antibody-inducing activity against HSV-2 (plaque number-reducing rate) were evaluated. As the result, it was found in both indexes that higher antibody-inducing activity was tend to be exhibited for bcev19 than bcev19′ and for bcev50 than bcev50′.

(134) From the above results, it was found that among the mutations introduced into the variants bcev19 and bcev50, the N-type glycochains (D199N, D200A, H201T) of domain I are mutations that contribute not only to neutralizing antibody-inducing ability enhancement, but also to protein expression level enhancement and property improvement.

INDUSTRIAL APPLICABILITY

(135) Modified HSV gB proteins of the present invention can be used in the production of vaccines effective for prevention and treatment of HSV infections.