Anti-platelet membrane glycoprotein vi monoclonal antibody   

The present invention relates to an antibody to human platelet membrane glycoprotein VI (hereinafter, sometimes abbreviated as GPVI) and a recognition region of the antibody.

The platelet plays a very important role in the clot formation and the biophylaxis, and its concerning in various clinical condition is being elucidated from the physiological role. In particular, it is remarkable about the function that the platelet forms a hemostatic plug. For example, when the vascular endothelial cell suffers damage, the collagen that is the major matrix protein of the subcutaneous vascular endothelium is exposed and the platelet adheres thereto. Next, the platelet is activated by the signal from the collagen and when finally, the platelet agglutinates through the fibrinogen. Then, since this fact causes morbidity such as thromboembolic disease depending on the situation, it is remarkable as a target for therapy.

In the past, for the purpose of the treatment and the prevention of the thrombosis based on the platelet aggregation, an anti-platelet agent such as aspirin, ticlopidine, GPIIb/IIIa antagonist and the like has been used. However, a lot of problems are pointed out from the aspect of the effectiveness and the side effects such as bleeding. Therefore, the excellent platelet inhibitor having the enough safety and the sure and appropriate function without the above-mentioned problems is desired to arrive.

GPVI that is present on the platelet membrane is the collagen receptor of the platelet, and it has been elucidated that the GPVI plays a central role for an activation of the platelet by collagen stimulation (see, Hiroshi Takayama, The Japanese Journal of Thrombosis and Hemostasis, 2003, volume 14, No. 2, pp. 75-81). That is, Sugiyama et al. has been reported that the membrane protein of 62 kDa is specifically depleted in the platelet of a patient with autoimmune thrombocytopenia and the platelet aggregation by collagen cannot be detected (see, Tateo Sugiyama and five other members, Blood (USA), 1987, volume 69, No. 6, pp. 1712-1720), and further that the protein that has been deleted in the platelet of the patient is GPVI and the Fab fragment of the antibody purified from the serum of the patient suppresses a collagen-induced platelet aggregation (see, Tateo Sugiyama and five other members, Blood (USA), 1987, volume 69, No. 6, pp. 1712-1720, and Masaaki Moroi and three other members, The Journal of Clinical Investigation (USA), 1989, volume 84, No. 5, pp. 1440-1445).

So far, Sugiyama et al. (see, Tateo Sugiyama and five other members, Blood (USA), 1987, volume 69, No. 6, pp. 1712-1720) and Takahashi et al. (see, Hoyu Takahashi and one other member, American Journal of Hematology (USA), 2001, volume 67, No. 4, pp. 262-267) have been reported about the anti-human GPVI autoantibody derived from the patient with autoimmune disease. However, since, according to the report by Sugiyama et al., the anti-human GPVI autoantibody purified from the patient's plasma has a function to induce platelet aggregation, it cannot be applied for medicaments immediately. Takahashi et al. (Hoyu Takahashi and one other member, American Journal of Hematology (USA), 2001, volume 67, No. 4, pp. 262-267) describes that an autoantibody to the protein of ca. 62 kDa that is speculated to be GPVI is present, and that this antibody induces a platelet aggregation. In addition, to apply these anti-GPVI antibodies derived from the patient clinically as a medicament, the antibody with high safety must be produced in quantities in stable quality. However, the method of producing industrially is not yet established.

The anti-GPVI antibody, which has been prepared by the present, includes a monoclonal rat antibody to mouse GPVI (see, Publication number 1228768 of the European Patent Application) and a monoclonal mouse antibody to human GPVI (see, Publication number 01/00810 of International Patent Application and Publication number 02/080968 of International Patent Application, Thromb Haemost. 2003 June; 89(6): 996-1003).

In addition, human single-stranded antibody (scFv: single chain Fv) that recognizes human GPVI has been prepared using the phage display method (see, Publication number 01/00810 of International Patent Application, Publication number 02/080968 of International Patent Application, and Peter A Smethurst and 15 other members, Blood (USA), 2002, volume 100, number 11, p. 474a). These single-stranded antibodies are the antibody that combines heavy chain variable region (VH) and light chain variable region (VL) of the human antibody by the peptide linker and has a variable region derived from the human. However, compared with normal immunoglobulin that the cell produces, it generally possesses a low affinity to an antigen thereof and is short in the half-life in the vivo, too. Further, Smethurst analyzed an epitope on GPVI in the term of clone 10B12 that suppresses platelet aggregation among the single-stranded antibodies, and suggested that lysine at the 59th position (Lys59) has a possibility to involve (Hoyu Takahashi and one other member, American Journal of Hematology (USA), 2001, volume 67, No. 4, pp. 262-267; and Peter A Smethurst and 15 other members, Blood (USA), 2002, volume 100, number 11, p. 474a).

As described above, most of the antibodies to human GPVI that have been reported until now, including the aforementioned human autoantibody, possess an activity of activating the platelet only by the antibody in vitro, and/or an activity of inducing or enhancing platelet aggregation. Thus when administering the same in vivo, a probability to cause thrombocytopenia is considered. In fact, Nieswandt et al. has reported several monoclonal antibodies (JAQ1, JAQ2 and JAQ3) that deplete GPVI on the platelet in vivo. All antibodies caused thrombocytopenia after administration.

Recently, there have been reported that collagen, the agonist for GPVI, convulxin and CRP, and an antibody (9012.2) that inhibits the binding of collagen to GPVI activate the platelet, followed by a shedding of GPVI from the platelet occurs by metalloprotease-mediated cleavage (Stephens G and four other members, Blood, 2005, 105(1): 186-191; Gardiner EE and four other members, Blood, 2004, 104: 3611-3617; Bergmeier W and six other members, Thromb Haemost., 2004, 91: 951-958). Further, the prediction of the amino acid residues on GPVI associated with the interaction with collagen (Val34, Leu36) was performed using the 9012.2 antibody, etc. (Lecut C and seven other members, J Biol Chem, 2004, 279: 52293-52299).

In addition, Takayama et al. cloned an anti-human GPVI antibodies using lymphocytes from the patient with autoimmune disease, and investigated about the feature of the antibodies in vitro (Publication number 05/007800 of International Patent Application).

However, in all reports that have been published until now, no antibody having the activity to deplete GPVI on the platelet membrane without activating the platelet and/or without inducing thrombocytopenia in vivo has been disclosed.

In the context of requirement for a medicament that has high safety and excellent efficacy, and is easy-to-use as an anti-platelet agent as described above, an anti-GPVI antibody applicable in vivois desired.

The present invention intends to provide a novel antibody that specifically binds to GPVI, which is the glycoprotein that exists on the human platelet membrane, preferably a monoclonal antibody. Particularly, there is provided the anti-GPVI antibody that is applicable in vivo, has efficacy and no problem in the aspect of side effects such as thrombocytopenia. Also, the invention provides the antibody that specifically binds to human GPVI and comprises a novel CDR sequence. Moreover, the cell that produces these antibodies is provided.

To solve the above-mentioned problems, the inventors have conceived of establishing a lot of mouse hybridomas that produce an antibody to GPVI and analyzing properties of the antibodies produced by the hybridomas. Based on the conception, the present inventors have made extensive investigations and have succeeded to obtain the hybridomas that produce the antibody having a binding ability to GPVI and an activity for decreasing the collagen-induced platelet aggregation. Then the inventors further analyzed recognition region on GPVI of each antibody to obtain useful information about epitope of GPVI. As a result of further investigation after isolating the clones, the inventors succeeded in obtaining a gene encoding the antibody. In addition, it was found that the amino acid sequence of CDR of the antibody is a novel one. Moreover, the inventors have completed the present invention by preparing recombinant antibodies using gene recombination technology.

In addition, throughout the specification, an antibody that is produced by a hybridoma, e.g. clone F1232-18 is sometimes referred to as F1232-18 antibody.

The first embodiment of the present invention is an antibody which exhibits a specific function or feature, and specifically binds to human GPVI, preferably a monoclonal antibody (hereinafter, sometimes referred to as anti-human GPVI antibody and human GPVI monoclonal antibody, respectively), its active fragment, or derivatives thereof. Specifically, it includes as follows.

(1) An antibody having the following features, its active fragment, or derivatives thereof: a) It specifically binds to human platelet membrane glycoprotein VI (GPVI); b) The action to activate a platelet and/or the action to induce a thrombocytopenia in vivo is weak; and c) It at least partially depletes GPVI on the platelet membrane by contacting with a platelet. (2) An antibody, its active fragment, or derivatives thereof, which at least partially depletes GPVI on the platelet membrane by contacting with a platelet without shedding of platelet GPVI, especially, shedding of platelet GPVI by metalloprotease-mediated cleavage accompanying platelet activation. (3) The antibody of (1) or (2), its active fragment or derivatives thereof, which at least partially depletes GPVI on the platelet membrane by contacting with a platelet in vivo. (4) The antibody of (1) or (3), its active fragment or derivatives thereof, which decreases or deletes the ability of platelet to aggregate responsive to collagen by being administered in vivo and contacting with a platelet. (5) The antibody of (1) to (4), its active fragment, or derivatives thereof, wherein its action to prolong bleeding time is weak. (6) The antibody of (1) to (5), its active fragment or derivatives thereof, wherein the dissociation constant with GPVI is equal to or less than 4×10−8 M.

The antibody of the aforementioned (1) to (6) is preferably an antibody which does not induce human platelet aggregation solely. Adequate examples of the antibody include antibody clones listed in Tables 6 and 11, preferably an antibody that recognizes loop 9 of GPVI, or a chimeric antibody or a humanized antibody, wherein the above antibody is recombined with human IgG, more preferably human IgG4. Further, the antibody of the present invention is an antibody, wherein the dissociation constant (Kd value) between human GPVI and the antibody is preferably equal to or less than 10−8 M, more preferably equal to or less than 4×10−9 M. The antibody, its active fragment or derivatives thereof of the present invention encompass, as long as they have a binding ability to GPVI, for example, a chimeric antibody and a humanized antibody, Fab (Fragment of antigen binding), Fab′, F(ab′)2, a single-chain antibody (scFv), a disulfide stabilized antibody (dsFv), diabody, nanobody and a peptide comprising CDR, and a labeled antibody, a conjugated antibody and a antibody-fused protein, etc.

Moreover, the antibody of the first embodiment of the present invention, or the like, is preferably an antibody which specifically binds to human GPVI and specifically decreases platelet aggregability against collagen, but has no effect on its aggregability against other agonists, e.g. ADP or thrombin. Preferably, the antibody cannot induce human platelet aggregation solely. The antibody cannot significantly induce human platelet aggregation in the concentration or the dosage equivalent to those, preferably 10-fold, more preferably 100-fold, further preferably 1000-fold of those, in which it suppresses collagen-induced human platelet aggregation.

Herein, the antibody of the aforementioned (1) to (6) may be an antibody which inhibits the binding of human GPVI to collagen as long as it has its properties, preferably an antibody which inhibits the binding of human GPVI to collagen with the dissociation constant (Kd) equal to or less than 10−8 M, more preferably equal to or less than 10−9 M, further preferably equal to or less than 10−10 M.

The antibody of the present invention is not necessarily limited to the specific clone, and the antibody having a similar function to that of the preferred examples of the present invention (or antibody, etc.) is encompassed within the scope of the present invention. The existence or non-existence of the function of the antibody of the present invention can be confirmed by the method shown in Examples or the publicly known method.

In addition, an antibody, wherein its recognition region, binding site or epitope on GPVI is the same or at least partially common to those of the preferred antibody of the present invention, for example, an antibody which competes with each other for binding with GPVI, is included within the scope of the present invention. Whether an antibody has a commonality in recognition region or binding site with the antibody of the present invention or not can be confirmed according to the method described in EXAMPLES or by publicly known methods. That is, the present invention provides an antibody which competes with a specified antibody of the present invention for GPVI binding. In the classification based on the competition experiment in EXAMPLES of the present invention, antibodies classified into eight groups in Table 1, preferably groups d, e or h, more preferably group d or e are exemplified as the antibody of the present invention.

The second embodiment of the present invention is an anti-GPVI antibody which is defined by novel recognition region, binding site or epitope on GPVI, preferably a monoclonal antibody. Specifically, it includes:

(7) An antibody, its active fragment or derivatives thereof, which specifically recognizes an amino acid sequence or a structure on the GPVI comprising at least a part of loop 2, loop 3 and loop 5, or loop 4 and loop 5 in human GPVI domain 1, or loop 9, or loop 9 and loop 11 in domain 2, preferably loop 9 in domain 2 or loop 2 in domain 1, more preferably loop 9 in domain 2; (8) The antibody of (7), its active fragment or derivatives thereof, wherein at least a part of loop 2, loop 3 and loop 5, or loop 4 and loop 5 in human GPVI domain 1, or loop 9, or loop 9 and loop 11 in domain 2 are E21, K22 and P23 of loop 2, G33 of loop 3 and A57, K59 and L62 of loop 5, or S43, S44, S45, R46, and E48 of loop 4 and A57, K59 and L62 of loop 5, or T116, R117, G119 and Q122 of loop 9 or T116, R117, G119, and Q122 of loop 9 and R139 of loop 11; (9 The antibody of (7) or (8), its active fragment or derivatives thereof, which specifically binds to loop 2, loop 3 and loop 5, or loop 4 and loop 5 in human GPVI domain 1, or loop 9, or loop 9 and loop 11 in domain 2, preferably loop 9 in domain 2 or loop 2 in domain 1, more preferably loop 9 in domain 2; (10) The antibody of (1) to (6), its active fragment or derivatives thereof, which recognizes an amino acid sequence or a structure on the GPVI comprising at least a part of loop 2, loop 3 and loop 5, or loop 4 and loop 5 in human GPVI domain 1, or loop 9, or loop 9 and loop 11 in domain 2, preferably loop 9 in domain 2 or loop 2 in domain 1, more preferably loop 9 in domain 2; (11) The antibody of (1) to (6), its active fragment or derivatives thereof, which specifically recognizes an amino acid sequence or a structure on the GPVI comprising E21, K22 and P23 of loop 2, G33 of loop 3 and A57, K59 and L62 of loop 5, or S43, S44, S45, R46, and E48 of loop 4 and A57, K59 and L62 of loop 5, or T116, R117, G119 and Q122 of loop 9 or T116, R117, G119, and Q122 of loop 9 and R139 of loop 11; (12) The antibody of (1) to (6), its active fragment or derivatives thereof, which specifically binds to loop 2, loop 3 and loop 5, or loop 4 and loop 5 in human GPVI domain 1, or loop 9, or loop 9 and loop 11 in domain 2, preferably loop 9 in domain 2 or loop 2 in domain 1, more preferably loop 9 in domain 2.

Herein, at least a part of each loop described above is, for example, the residues, which are different from the corresponding amino acid residues of heterogenous GPVI such as mouse or rat GPVI. A modeling structure of GPVI such as human GPVI is presumable by the method described in EXAMPLES. A position of each loop structure is shown in FIGS. 1 and 3. Among the abovementioned loops, loop 9 in domain 2 and loop 2 in domain 1, preferably loop 9 in domain 2 are important as a recognition region of the antibody of the present invention, and the antibody which can recognize the region is preferred. Preferred examples are an antibody listed in Tables 6 and 11, or a chimeric or humanized antibody which is recombined with human IgG, more preferably human IgG4.

The antibody of the second embodiment of the present invention can be classified or can be confirmed as for its binding region by the binding property with the peptide of the eighth embodiment of the present invention and/or the polypeptide of the ninth embodiment of the present invention. That is, the present invention provides an anti-GPVI antibody which has a distinct binding property, preferably a decreased binding property with the specified substance among the peptides of the eighth embodiment of the present invention and/or the polypeptides of the ninth embodiment of the present invention. Specifically, it is an anti-GPVI antibody which has a binding property with the specified GPVI mutant significantly different from that with human GPVI and/or other GPVI mutants, preferably a decreased binding property. Specific examples of the method of confirmation, polypeptides to be used, preferred classification and preferred antibodies are illustrated in EXAMPLES.

An antibody of the present invention is not necessarily limited regarding its antigen-binding valency and may be a monovalent antibody such as Fab or scFv. From the aspects of stability in vivo, especially in the blood, or binding property to GPVI or strength of action, preferred is a multivalent antibody having two or more valencies, e.g., a divalent, trivalent, tetravalent or decavalent antibody; a divalent antibody is more preferable. Accordingly, in the second embodiment of the present invention, a monovalent antibody and a multivalent antibody having two or more valency which recognizes a specific region, especially loop 9 on GPVI, such as a divalent, trivalent, tetravalent or decavalent antibody; preferably a divalent antibody can be provided. Herein, examples of tetravalent antibody include IgA, and examples of decavalent antibody include IgM. However, they are not limited to these examples. Further, a trivalent antibody does not physiologically exist. However, by chemically or through genetic engineering binding a natural or synthetic peptide having an intrinsic trimerization property such as the domain of tenascin molecule (AA 110-139, Swissprot #P10039 (chicken) or Swissprot #P24821 (human)) to a monovalent antibody (scFv or Fab and so on), a trivalent antibody can be prepared (see, JP 2004-508828 publication). In addition, the antibody of the second embodiment of the invention is preferably an antibody which exhibits a specific function or property of the antibody of the first embodiment.

The third embodiment of the present invention is an anti-GPVI antibody which comprises a novel amino acid sequence of CDR or variable region, preferably a chimeric antibody which is recombined with human IgG, especially human IgG4, more preferably a CDR grafted antibody, especially a humanized antibody. Specifically, it includes:

(13) An anti-GPVI antibody, its active fragment or derivatives thereof, wherein at least three CDR of either H-chain or L-chain in the antibody, preferably six CDR of both H-chain and L-chain in the antibody comprises an amino acid sequence of CDR in the clones listed in Tables 8, 9, 12 and 13, preferably an antibody which recognizes loop 9 of GPVI as an amino acid sequence of the corresponding CDR; (14) A heavy chain of anti-GPVI antibody, its active fragment, or derivatives thereof, wherein the amino acid sequences of SEQ ID NOs: 15, 16, and 17, the amino acid sequences of SEQ ID NOs: 18, 19, and 20, the amino acid sequences of SEQ ID NOs: 21, 22, and 23, the amino acid sequences of SEQ ID NOs: 24, 25, and 26, the amino acid sequences of SEQ ID NOs: 27, 28, and 29, the amino acid sequences of SEQ ID NOs: 30, 31, and 32, the amino acid sequences of SEQ ID NOs: 33, 34, and 35, the amino acid sequences of SEQ ID NOs: 36, 37, and 38, the amino acid sequences of SEQ ID NOs: 39, 40, and 41, the amino acid sequences of SEQ ID NOs: 42, 43, and 44, the amino acid sequences of SEQ ID NOs: 45, 46, and 47, or the amino acid sequences of SEQ ID NOs: 48, 49, and 50, or VH CDR1, VH CDR2, and VH CDR3 of any clone listed in Table 12 are comprised in VH CDR1, VH CDR2, and VH CDR3, respectively; (15) A light chain of anti-GPVI antibody, its active fragment, or derivatives thereof, wherein the amino acid sequences of SEQ ID NOs: 51, 52 and 53, the amino acid sequences of SEQ ID NOs: 54, 55 and 56, the amino acid sequences of SEQ ID NOs: 57, 58 and 59, the amino acid sequences of SEQ ID NOs: 60, 61 and 62, the amino acid sequences of SEQ ID NOs: 63, 64 and 65, the amino acid sequences of SEQ ID NOs: 66, 67 and 68, the amino acid sequences of SEQ ID NOs: 69, 70 and 71, the amino acid sequences of SEQ ID NOs: 72, 73 and 74, the amino acid sequences of SEQ ID NOs: 75, 76 and 77, the amino acid sequences of SEQ ID NOs: 78, 79 and 80, the amino acid sequences of SEQ ID NOs: 81, 82 and 83, or the amino acid sequences of SEQ ID NOs: 84, 85 and 86, or VL CDR1, VL CDR2, and VL CDR3 of any clone listed in Table 13 are comprised in VL CDR1, VL CDR2 and VL CDR3, respectively; (16) An anti-GPVI antibody, its active fragment, or derivatives thereof, wherein the amino acid sequences of SEQ ID NOs: 15, 16, 17, 51, 52 and 53, the amino acid sequences of SEQ ID NOs: 18, 19, 20, 54, 55 and 56, the amino acid sequences of SEQ ID NOs: 21, 22, 23, 57, 58 and 59, the amino acid sequences of SEQ ID NOs: 24, 25, 26, 60, 61 and 62, the amino acid sequences of SEQ ID NOs: 27, 28, 29, 63, 64 and 65, the amino acid sequences of SEQ ID NOs: 30, 31, 32, 66, 67 and 68, the amino acid sequences of SEQ ID NOs: 33, 34, 35, 69, 70 and 71, the amino acid sequences of SEQ ID NOs: 36, 37, 38, 72, 73 and 74, the amino acid sequences of SEQ ID NOs: 39, 40, 41, 75, 76 and 77, the amino acid sequences of SEQ ID NOs: 42, 43, 44, 78, 79 and 80, the amino acid sequences of SEQ ID NOs: 45, 46, 47, 81, 82 and 83, or the amino acid sequences of SEQ ID NOs: 48, 49, 50, 84, 85 and 86, or VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2, and VL CDR3 of any clone listed in Tables 12 and 13 are comprised in VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3, respectively; (17) An anti-GPVI antibody, its active fragments or derivatives thereof, wherein at least variable region of either H-chain or L-chain in the antibody, preferably variable regions of both H-chain and L-chain in the antibody comprises an amino acid sequence of variable region in the clones listed in Table 7 or Table 14, preferably an antibody which recognizes loop 9 of GPVI as an amino acid sequence of the corresponding variable regions, especially a chimeric antibody which is recombined with human IgG, preferably human IgG4.

The antibody of the third embodiment of the present invention is preferably an antibody which possesses characteristics and/or specificity for recognition region of the antibody of the first and/or the second embodiment.

The fourth embodiment of the present invention is a polynucleotide or nucleic acid which comprises the base sequence encoding at least three CDR of either H-chain or L-chain, preferably six CDR of both H-chain and L-chain, or variable region in the antibody of the first to the third embodiments, its active fragment or derivatives thereof. Specifically, it includes:

(18) A polynucleotide which comprises the base sequence encoding H-chain and/or L-chain of the antibody of the first to the third embodiments, its active fragment or derivatives thereof, (19) The polynucleotide of (18) which comprises the base sequence encoding the corresponding CDR in the clones listed in Tables 8 and 9, or Tables 12 and 13, preferably an antibody which recognizes loop 9 of GPVI, as the base sequence encoding at least three CDR of either H-chain or L-chain in the antibody, preferably six CDR of both H-chain and L-chain in the antibody; (20) A polynucleotide which comprises the base sequence encoding the corresponding variable region in the clones listed in Tables 7 and 14, an antibody which recognizes loop 9 of GPVI, as the base sequence encoding at least variable region of either H-chain or L-chain in the antibody, preferably variable regions of both H-chain and L-chain in the antibody; (21) A polynucleotide comprising the base sequence having SEQ ID NO: 280 and encoding variable region of the H-chain and the base sequence having SEQ ID NO: 284 and encoding variable region of the L-chain, and a polynucleotide comprising the base sequence having SEQ ID NO: 282 and encoding variable region of the H-chain and the base sequence having SEQ ID NO: 284 and encoding variable region of the L-chain.

Further, the present invention provides an anti-human GPVI antibody gene, or its heavy chain or light chain variable region gene thereof, which is derived from an antibody gene comprising a combination of specific mouse germ-line antibody gene segments. That is,

(22) An anti-human GPVI antibody gene, or a heavy chain variable region gene thereof, which is derived from an antibody heavy chain gene comprising any combination of mouse germ-line antibody gene segments VH, DH and JH listed in Table 16. (23) An anti-human GPVI antibody gene, or a heavy chain variable region gene thereof, comprising a nucleotide sequence encoding CDR amino acid sequence in the aforementioned antibody heavy chain variable gene. (24) An anti-human GPVI antibody gene, or a light chain variable region gene thereof, which is derived from an antibody light chain gene comprising any combination of mouse germ-line antibody gene segments VL and JL listed in Table 16. (25) An anti-human GPVI antibody gene, or a light chain variable region gene thereof comprising a nucleotide sequence encoding CDR amino acid sequence in the aforementioned antibody light chain variable gene. Herein, among the mouse germ-line antibody gene segments listed in Table 16, a combination of segments having a high score indicated on the first line of each antibody clones, for example, a combination of VH (3:3.9), DH (DSP2.7 or DSP2.5) and JH (JH4) in the heavy chain gene of clone F1246-1-1, is preferred. Further, a gene derived from the above-mentioned antibody gene includes the antibody gene itself or a gene with mutation in one base or more as long as an antibody encoded by the gene exhibits a similar antigen-binding specificity. In addition, both of naturally occurring or artificially introduced mutations may be acceptable. At the same time as the above, the present invention provides an antibody, its active fragment or derivatives thereof, encoded by the anti-human GPVI antibody gene or its heavy chain or light chain variable region gene, which is derived from an antibody gene comprising a combination of specific mouse germ-line antibody gene segments. That is, (26) An anti-human GPVI antibody or its heavy chain variable region polypeptide encoded by the antibody gene or its heavy chain variable region gene of (22) to (23). (27) An anti-human GPVI antibody or its light chain variable region polypeptide encoded by the antibody gene or its light chain variable region gene of (24) to (25).

In addition, the present invention provides an anti-human GPVI antibody, specifically, the above-mentioned antibody of the present invention, preferably the antibody recognizing loop 9 of GPVI, or its active fragment or derivatives thereof, which is polyethyleneglycolated (PEGylated). The method for PEGylating the antibody etc. may be in accordance with publicly known method (for example, Roberts M. J. et al., Advanced Drug Delivery Reviews 54 (2002) 459-476), and specifically is described in EXAMPLE 31.

The fifth embodiment of the present invention is a cell which produces the antibody of the first through the third embodiment, its active fragment or derivatives thereof, or a cell which comprises the polynucleotide of the fourth embodiment. Specifically, it includes:

(28) A cell which produces any antibody described in the aforementioned (1) through (17), its active fragment, or derivatives thereof, especially a transformant or a hybridoma; (29) A cell which comprises any polynucleotide described in the aforementioned (18) through (21), especially a transformant or a hybridoma.

The sixth embodiment of the present invention is a method of manufacturing the antibody of the first through the third embodiment, which is characterized by using the polynucleotide of the fourth embodiment or an expression vector comprising the same, or the cell of the fifth embodiment. Specifically, it includes:

(30) A method of manufacturing the antibody of the first through the third embodiment, which comprises a process, in which the cell of the aforementioned (28) or (29) is cultured, and a process, in which a monoclonal antibody produced by the cell is collected; (31) A method of manufacturing the antibody of the first through the third embodiment, which comprises a process, in which any among the polynucleotide of the aforementioned (17) through (21), an expression vector comprising the same, and the cell of the above (28) or (29) is used.

The seventh embodiment of the present invention relates to a medical composition comprising the antibody of the first to the third embodiment of the present invention, its active fragment or derivatives thereof as an effective ingredient, and preferably is a medical composition for prophylaxis and/or therapy of thrombotic, embolic or arteriosclerotic disease. The antibody of the present invention hardly has side effects such as activation of platelet, platelet aggregation, thrombocytopenia and prolongation of bleeding time and so on, and is useful for prevention and/or therapy of the above-mentioned diseases.

The eighth embodiment of the present invention is a peptide which constitutes the specific structure, especially a loop structure on GPVI, specifically,

(32) a peptide which comprises loop 2, loop 3 and loop 5, or loop 4 and loop 5 in domain 1, or loop 9, or loop 9 and loop 11 in domain 2 of human GPVI, especially, a peptide which consists of any one of amino acid sequence among them. Herein, the peptide may comprise an amino acid sequence derived from heterogeneous GPVI or an amino acid sequence of a polypeptide other than GPVI such as Fc.

The ninth embodiment of the present invention is a specified GPVI mutant, for example, a mutant with an amino acid substitution, a domain substitution between species or a partial sequence substitution between species such as a loop substitution, and the like. Preferred is a mutant wherein amino acids constituting one or more loop structures of GPVI shown in FIGS. 1 and 3 may be substituted with other amino acids or amino acids of the corresponding loop of other species such as human, mouse and rat. Specific examples are described in Table 4 or EXAMPLES. Specifically,

(33) a polypeptide which comprises the amino acid sequence of SEQ ID NOs: 137 through 151.

The tenth embodiment of the present invention is a method of screening an antibody, its active fragment or derivatives thereof, which comprises the following process:

a) A process for measuring the binding property with human platelet membrane glycoprotein VI (GPVI); b) A process for measuring the action to activate a platelet and/or the function to induce thrombocytopenia in vivo; and c) A process for measuring the activity to at least partially deplete GPVI on the platelet membrane disappear by contacting with a platelet.

The eleventh embodiment of the present invention is a method of estimating an epitope of an antibody or a method of identifying a recognition region of an antibody, which comprises a process for measuring the reactivity, for example, the binding property with the antibody of the peptide of the eighth embodiment or the polypeptide of the ninth embodiment.

The twelfth embodiment is a method of manufacturing an antibody specific for GPVI, which is characterized by using the peptide of the eighth embodiment or the polypeptide of the ninth embodiment, specifically,

(34) a method of manufacturing an antibody specific for GPVI, preferably the antibody of the first to the third embodiments of the present invention, which is characterized by using the peptide of the eighth embodiment or the polypeptide of the ninth embodiment as an antigen for immunization or as an antigen for in vitro immunization; (35) a method of manufacturing an antibody specific for GPVI, preferably the antibody of the first to the third embodiments of the present invention, which is characterized by using the peptide of the eighth embodiment or the polypeptide of the ninth embodiment as an antigen for detection or identification of the antibody. That is, using a recombinant GPVI as an immunogen and/or an antigen for detection, wherein the amino acid sequence on human GPVI, such as the amino acid sequence corresponding to the loop structure, which can be recognized by the antibody of the present invention, is integrated into mouse GPVI, a novel antibody which can recognize the same recognition region can be obtained. As a more preferred method of preparing an antibody for human therapy, the method using human antibody gene-transgenic non-human animal has been disclosed (WO 2002/070648 (Tokuhyou 2005-504507) and WO 2002/043478 (Tokuhyou 2004-515230)). When using a protein, wherein a partial amino acid sequence of human GPVI is integrated into a heterogeneous GPVI, for example, mouse GPVI, the above-mentioned transgenic animal such as a mouse is immunized, it will be considered that a human antibody which reacts with the integrated amino acid sequence from human, preferably epitope, but not the amino acid sequence from mouse on GPVI can be obtained efficiently. Therefore, the human antibody obtained by such method is useful as a human antibody having the features of the antibody of the first or the second embodiment, and said method is particularly useful.

The thirteenth embodiment of the present invention is a method of detecting or quantifying GPVI in the test sample using the antibody of the first to the third embodiments. By the method, GPVI on the platelet or in the body fluid, especially in the blood can be measured. Further, it can be applied to a method of diagnosing diseases, preferably a method of diagnosing diseases associated with clot formation. Furthermore, the method can be applied to monitoring for treatment relating to GPVI, particularly prediction or determination of efficacy of anti-GPVI antibody, or prognostic determination.

FIG. 1 is the alignment of the amino acid sequence of human soluble GPVI and mouse soluble GPVI. The squares show the positions of each domain region of GPVI and the loop region deduced by the modeling (L1-L14).

FIG. 2 presents the results of the competitive test of an antibody from a patient with GPVI deficiency with mouse anti-human GPVI monoclonal antibody. YA-Abs-88 and YA-Abs-03 mean anti-GPVI antibodies derived from a patient with GPVI deficiency.

FIG. 3 is the alignment of the amino acid sequence of human soluble GPVI and rat soluble GPVI. The squares show the positions of each domain region of GPVI and the loop region deduced by the modeling (L1-L14).

FIG. 4 shows the results of reactivity for mouse hybridoma antibody and chimeric antibody.

FIG. 5 shows the inhibition of binding between GPVI and collagen by the chimeric antibody and the mouse hybridoma antibody.

FIG. 6 presents the results of a binding property of anti-human GPVI antibody to GPVI mutants.

FIG. 7 shows the results of a reactivity of cF1232-37-2 with various hGPVI mouse loop substitution mutants.

FIG. 8 shows the activating action on human and cynololgus platelets. Expression level of CD62P (P-selectin) on human platelet (A) and cynomolgus monkey platelet (B) was assayed by FACS and the results were shown by mean fluorescent intensity (MFI).

FIG. 9 shows the aggregation-inducing action on human platelet.

FIG. 10 shows the action of F1232-37-2 on collagen-induced human platelet aggregation.

FIG. 11 shows the action of F1232-37-2 on ADP-induced human platelet aggregation.

FIG. 12 shows the results of the test of intravenous administration of mouse anti-human GPVI monoclonal antibodies F1232-37-2 and F1199-6 into cynomolgus.

FIG. 13 shows the results of ex vivo test of single intravenous administration of mouse/human chimeric anti-human GPVI antibody (cF1232-37-2) in cynomolgus.

FIG. 14 shows the results of ex vivo test of repetitive intravenous administrations of mouse/human chimeric anti-human GPVI antibody in cynomolgus. When 0.3 mg/kg of cF1232-37-2 was administered four times to cynomolgus monkey every other day, collagen-induced platelet aggregation (A) and platelet GPVI level (B) were measured.

FIG. 15 shows the results of ex vivo test of subcutaneous administration of mouse/human chimeric anti-human GPVI antibody in cynomolgus. After cF1232-37-2 was subcutaneously administered to cynomolgus monkey, the blood was collected with time, and platelet aggregation (A) induced by 2 μg/mL of collagen and platelet GPVI level (B) were measured.

FIG. 16 shows an inhibiting action of F1232-37-2 Fab on collagen-induced platelet aggregation.

FIG. 17 shows the results of ex vivo test of F1232-37-2 F(ab′)2 in cynomolgus.

FIG. 18 shows the construction of a stable co-expression plasmid for both chains of cF1232-37-2.

FIG. 19 shows a reactivity for antigen binding of cF1232-37-2 expressed in COS cells and CHO cells.

FIG. 20 shows an amino acid sequence of heavy chain variable region and humanization thereof.

FIG. 21 shows an amino acid sequence of light chain variable region and humanization thereof.

FIG. 22 shows a binding specificity of humanized antibody to GPVI.

FIG. 23 presents a result of collagen-induced aggregation test of platelet from cynomolgus monkey administered with chimeric anti-GPVI antibody.

FIG. 24 presents a result of the bleeding time test in cynomolgus monkey. A: shows the collagen-induced platelet aggregation at five minutes after intravenous administration of eptifibatide and at 48 hours after administration of anti-GPVI antibody cF1232-37-2. B: shows a result of the bleeding time at five minutes after intravenous administration of eptifibatide and at 48 hours after administration of anti-GPVI antibody cF1232-37-2 compared with the bleeding time prior to administration.

FIG. 25 shows a result of confirmation test for platelet GPVI antigen shedding by anti-GPVI antibodies.

FIG. 26 presents a result of PEGylation of anti-GPVI whole antibody and Fab antibody. Lanes shows 1: F1232-37-2 whole antibody; 2: PEGylated product of F1232-37-2; 3: PEGylated purified product of F1232-37-2; 4: F1232-37-2 Fab antibody; 5: PEGylated product of F1232-37-2 Fab; and 6: PEGylated purified product of F1232-37-2 Fab, respectively.

FIG. 27 shows the result of GPVI antigen binding assay for PEGylated anti-GPVI antibody.

The antibody of the present invention is the one that specifically recognizes GPVI, the membrane glycoprotein existing on human platelet. Further, the GPVI that is recognized by the antibody of the present invention is not always limited to the GPVI on the platelet, and for example, GPVI on megakaryocyte can also be recognized. The present invention will be described in detail as follows. In addition, in the specification, sometimes, an amino acid residue is denoted as one-letter code or three-letter code.

The antibody of the present invention may be a polyclonal antibody, or preferably is a monoclonal antibody. A method of preparing the monoclonal antibody is not limited to the specified method. The monoclonal antibody may be any of a monoclonal antibody produced by hybridoma, a monoclonal antibody produced by a recombinant cell, in which a gene encoding the antibody is incorporated, or a monoclonal antibody produced by the transformed cell with Epstein-Barr virus (EBV). In addition, it may be a mixture of antibodies or a polyclonal antibodies comprising at least one of the monoclonal antibodies of the present invention, or a mixture of plural monoclonal antibodies of the present invention. Further, the antibody of the present invention encompasses a bispecific antibody or a polyspecific antibody.

The antibody of the present invention is an antibody that specifically binds to human GPVI. The binding of the antibody of the present invention with human GPVI can be measured by publicly known methods, specifically the method described in EXAMPLES. Dissociation constant (Kd value) between human GPVI and the antibody of the present invention is 4×10−8 M, preferably equal to or less than 10−8 M, more preferably equal to or less than 4×10−9 M, further preferably equal to or less than 10−9 M. A method of determining dissociation constant between human GPVI and the antibody is not limited to the specified method, and the conventional method can be used. For example, it can be measured with the protein interaction analyzer such as BIACORE3000 using GPVI-Fc immobilized on chip. Specifically, it can be illustrated in EXAMPLE 5.

The antibody of the first embodiment of the present invention has an activity for disappearing GPVI on the platelet membrane at least partially by bringing in contact with the platelet, in particular, by bringing in contact with the platelet in vivo. The activity can be confirmed by bringing the antibody of the present invention in contact with the platelet for a given time followed by isolating the platelet, and assaying the expression level of GPVI on the surface. The expression level of GPVI can be measured by the conventional method using FACS, etc., and the specific method is illustrated in EXAMPLES. The antibody of the present invention has an activity for depleting GPVI on the platelet preferably equal to or more than 40%, more preferably equal to or more than 80% in comparison with the value prior to administration or the control value at the dosage of preferably 1 mg/kg, more preferably 0.3 mg/kg.

The antibody of the present invention is an antibody that has GPVI-depleting ability without mediating shedding of platelet GPVI, in particular, shedding of GPVI from platelets by metalloprotease-mediated cleavage accompanying with platelet activation or an antibody that at least partially depletes GPVI on the platelet membrane by contacting with platelets with little or virtually no shedding. Herein, shedding can be detected by the publicly known method (Stephens G and other four members, Blood, 2005 Jan. 1; 105 (1): 186-191; Gardiner E E et al., Blood 2004, 104: 3611-3617; Bergmeier W et al., Thromb Haemost. 2004; 91: 951-958), and specifically the method described in EXAMPLE 30 may be applicable.

The antibody of the present invention per se has a low activity for activating a platelet and/or inducing thrombocytopenia, preferably no activity. The activation of platelet can be measured by known methods, and an expression level of a platelet surface antigen, preferably CD62P can be used as an index. For example, a method of isolating a platelet from a living body that the antibody has been administered after a given period and measuring the expression level of CD62P by a conventional method, a method of bringing the antibody of the present invention in contact with a platelet isolated from a living body, and assaying the expression level of CD62P after a given period by a conventional method, and the like are included. The specific method is shown in EXAMPLES. The activation of platelet by the antibody of the present invention is, when the expression level of CD62P is used as an index, at the dose or the concentration that at least partially depletes GPVI on the platelet, equal to or less than five-fold, preferably equal to or less than two-fold, more preferably equal to or less than 1.5-fold, further preferably almost the same as the platelet for control.

Thrombocytopenia in vivo can be confirmed by collecting blood with time after in vivo administration of the antibody of the present invention, calculating the number of platelets with a conventional method and comparing the number with a value prior to administration or the number of platelets of an individual as a control. The specific method is shown in EXAMPLES. The platelet number by the antibody of the present invention is, at the dose or the concentration that at least partially disappears GPVI on the platelet, when the value prior to administration or the control value is set as 100%, equal to or more than 50%, preferably equal to or more than 70%, more preferably equal to or more than 90%, further preferably almost the same.

The antibody of the present invention has the activity that suppresses the human platelet aggregability against the collagen, that is, the activity that attenuates or deletes aggregability of the platelet responsive to collagen by contacting with the platelet in vivo. The activity can be confirmed by administering the antibody of the present invention in vivo to make contact with the platelet, followed by isolating the platelet with time, and measuring a collagen-induced platelet aggregation. Herein, the platelet aggregation can be measured by the publicly known method. For example, it can be measured by calculating an aggregation rate using a light transmission with the platelet aggregometer as an index, and in general, may be represented by the aggregation rate at the point that exhibits a maximum light transmission (hereinafter, sometimes referred to as the maximum aggregation rate). In the method described in EXAMPLE 8 later, the antibody of the present invention has an activity that at the dosage of preferably 1 mg/kg, more preferably 0.3 mg/kg, makes the collagen-induced aggregability of the platelet decrease by preferably equal to or more than 40%, more preferably equal to or more than 80% in comparison to the value prior to the administration or the control value.

Preferably, the antibody of the present invention hardly has an effect on aggregation induced by platelet aggregation-inducing substances except for collagen, for example, ADP or thrombin. At the dosage or the concentration that affects on collagen-induced aggregability, the maximum aggregation rate is preferably equal to or more than 80% of the control, more preferably equal to or more than 90% of the control, further preferably equal to or more than 95% of the control. A method of assaying suppression of a human platelet aggregation by platelet aggregation-inducing substances except for collagen can be performed by the conventional methods.

In addition, the antibody of the present invention is an antibody that has a weak prolonging activity on bleeding time, preferably no significant prolonging activity, more preferably no substantial prolonging activity. The bleeding time can be assayed by the publicly known method, and specifically the method described in EXAMPLE 28 may be applicable.

Since most of the antibodies to the human GPVI, including the above-mentioned human autoantibody, that have been reported until now, possess, in vitro, an activity that activates a platelet by the antibody itself, and/or an activity that induces or enhance a platelet aggregation, when they are administered in vivo, the possibility to cause a thrombocytopenia may be considered. In the form of the Fab fragment and so on, the one that does not induce a platelet aggregation is also reported, meanwhile in vivo, a possibility that Fab behaves in a similar manner to IgG by cross-linking or aggregating from any cause cannot fully be denied. Therefore, in an intact antibody molecule, but not an active fragment of the antibody, for example, IgG form, an anti-GPVI antibody that does not exhibit the above-mentioned activity, or has a low activity is preferred.

Also, for behavior and stability in vivo, the antibody molecule that is the natural form, e.g. IgG, is superior. Generally, a half-life of the IgG in the blood is much longer than that of the fragment such as Fab. Thus especially, for chronic diseases such as thrombosis, especially thrombosis associated with atrial fibrillation, or clinical conditions that necessitate an antibody administration over long period, a molecular form having a long half-life in the blood, particularly IgG is desirable.

The antibody of the present invention may specifically inhibit the binding of GPVI on the platelet to collagen. For example, in the method described in EXAMPLES later, the antibody of the present invention is an antibody that inhibits the binding of GPVI and collagen by 50% at the concentration preferably equal to or less than 100 μg/mL, more preferably equal to or less than 10 μg/mL, further preferably equal to or less than 1 μg/mL, especially preferably equal to or less than 0.1 μg/mL. A method of measuring the binding of collagen and GPVI is not limited to a specific method and can also be done by the other conventional method.

The second embodiment of the present invention is an anti-GPVI antibody that is defined by novel recognition region, binding site or epitope on GPVI, preferably a monoclonal antibody. The recognition region on GPVI by the antibody of the present invention and the like can be confirmed or deduced by publicly known methods. For example, by applying the method of the eleventh embodiment of the present invention and measuring a reactivity with the peptide of the eighth embodiment or the polypeptide of the ninth embodiment, it can be performed. The specific method is illustrated in EXAMPLES. For example, in the method described in EXAMPLE 7 or EXAMPLE 18, on the basis where the reactivity or inhibition rate is significantly varied in comparison to the control (e.g., hGPVI-Fc), for example, it is reduced to 50%, preferably 30%, more preferably less or equal to 10%; or where a value for IC50 and the like are significantly changed, for example, it is increased three-fold, ten-fold, more preferably 30-fold, further preferably 100-fold, the antibody can be confirmed or deduced. The antibody, wherein recognition region, binding site or epitope on GPVI is confirmed, is useful for detection of specific GPVI molecular species, or for analysis of relationship between the structure and the function of GPVI, solely or by combination with other antibodies.

As the third embodiment of the present invention, there is an anti-GPVI antibody comprising a novel amino acid sequence of CDR or variable region.

On the N-terminal end of heavy chain and light chain of antibody, variable region exists, and is designated heavy chain variable region (VH) and light chain variable region (VL), respectively. Within the variable region, complementarity determining region (CDR) is present, and assumes specificity for recognition of antigen. A part except CDR in variable region has the role that maintains the structure of CDR and is called framework region (FR). On the C-terminal end of heavy chain and light chain of antibody, constant region exists, and is designated heavy chain constant region (CH) and light chain constant region (CL), respectively.

In the heavy chain variable region, three complementarity determining regions exist: the first complementarity determining region (CDR1), the second complementarity determining region (CDR2) and the third complementarity determining region (CDR3). These three complementarity determining regions in the heavy chain variable region collectively mean a heavy chain complementarity determining region. As is the case with the heavy chain, in the light chain variable region, three complementarity determining regions exist: the first complementarity determining region (CDR1), the second complementarity determining region (CDR2) and the third complementarity determining region (CDR3). These three complementarity determining regions in the light chain variable region collectively mean a light chain complementarity determining region.

CDR sequence in the antibody of the present invention is not always limited. Preferred combination of the amino acid sequences as VH CDR1, VH CDR2 and VH CDR3, preferred combination of the amino acid sequences as VL CDR1, VL CDR2 and VL CDR3, and further, preferred combination of the amino acid sequences as VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3, are listed in Tables 8 and 9, and Tables 12 and 13. Preferably, it is an antibody that comprises any one or more among the amino acid sequences of CDR of an antibody that recognizes loop 9 of GPVI. The amino acid sequences except for CDR are not particularly limited. The antibody of the present invention includes so-called CDR-grafted antibody, wherein the amino acid sequences except for CDR are derived from another antibody, especially the antibody from other species. Among them, preferred is a humanized antibody, wherein the amino acid sequences except for CDR are derived from human. The humanized antibody may have addition, deletion, substitution and/or insertion of one to several amino acid residues in the framework (FR) region, if desired. As a method of preparing the humanized antibody, publicly known methods can be used, and the specific method is shown in EXAMPLES.

The amino acid sequence of VH and VL of the antibody of the present invention is not always limited, but preferred antibody is an antibody that comprises any one or more among the amino acid sequence of SEQ ID NO: 281 as VH or the amino acid sequence of SEQ ID NO: 285 as VL, or an antibody that comprises any one or more among the amino acid sequence of SEQ ID NO: 283 as VH or the amino acid sequence of SEQ ID NO: 285 as VL.

In addition, the antibody of the present invention is not always limited to that with the specific amino acid sequence, and within the range where there is virtually no influence on its activity and/or antigenecity, as for the amino acid sequence of the antibody of the present invention, for example, variable region, especially FR, addition, deletion, substitution and/or insertion of one to several amino acid residues are permissible.

The antibody of the present invention is an antibody that the constant region of the antibody consists of an amino acid sequence derived from preferably human antibody, more preferably human IgG, further preferably human IgG4.

The antibody of this invention is not always limited to the specific molecular species. The structure of an antibody, i.e. an immunoglobulin consists of heavy chain (H-chain) and light chain (L-chain) and is divided into five isotypes (IgG, IgA, IgM, IgD, IgE) based on the class of the heavy chain (γ, α, μ, δ, ε). Among them, IgG and IgA are divided, based on difference of the heavy chain (e.g., in the case of the human, γ1, γ2, γ3, γ4, α1, α2), into subclasses (e.g., in the case of the human, IgG1, IgG2, IgG3, IgG4, IgA1, IgA2). The light chain is classified into either κ or λ type. The class, the subtype or the isotype of the antibody of the present invention is not limited, and may be the one that is classified into either. Preferred isotype is IgG, and further preferably, the subclass is IgG4 in the point that there is no complement fixation.

The antibody of the present invention, as long as the antibody has the activity such as a binding ability to GPVI, may be the fragment or the part of the antibody. For example, Fab (fragment of antigen binding), Fab′, (Fab′)2, single-stranded antibody (scFv), disulfide stabilizing antibody (dsFv), diabody, sc(Fv)2 (see, e.g., Orita T, Blood. 2005; 105: 562-566), nanobody (see, e.g., Cortez-Retamozo V., Cancer Research 64, 2853-2857, 2004) and the peptide comprising CDR and so on are included. An antibody of the present invention is not always limited regarding antigen binding valency and may be monovalent antibody such as Fab or scFv. From the perspective of stability in vivo, especially in the blood, and/or binding property to GPVI or strength of action, preferred is a polyvalent antibody having two or more valencies, e.g., a divalent, trivalent, tetravalent or decavalent antibody; a divalent antibody is more preferable.

In the fourth embodiment of the present invention, a polynucleotide or nucleic acid encoding the antibody of the first embodiment to the third embodiment of the present invention is provided. The polynucleotide is not always limited as long as that encodes the amino acid sequence of the antibody of the present invention, and includes DNA and RNA.

A polynucleotide encoding CDR sequence in the antibody of the present invention is not always limited. Preferred combination of the base sequences encoding the amino acid sequence as VH CDR1, VH CDR2 and VH CDR3, preferred combination of the base sequences encoding the amino acid sequence as VL CDR1, VL CDR2 and VL CDR3, and further, preferred combination of the base sequences encoding the amino acid sequence as VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3, are listed in Tables 8 and 9, and Tables 12 and 13.

The polynucleotide encoding the amino acid sequence of VH and VL of the antibody of the present invention is not always limited, and preferably is a polynucleotide that comprises any one of the base sequences of SEQ ID NOs: 87, 89, 91, 93, 95, 97, 99, 101, 103, 105, 107 and 109 encoding the amino acid sequence as VH or the base sequences of SEQ ID NOs: 88, 90, 92, 94, 96, 98, 100, 102, 104, 106, 108 and 110 encoding the amino acid sequence as VL, more preferably both base sequences, or a polynucleotide that comprises either the base sequences having the corresponding sequence identification number and encoding the amino acid sequence as VH or the base sequences having the corresponding sequence identification number and encoding the amino acid sequence as VL, more preferably both base sequences.

The polynucleotide encoding the constant region of the antibody of the present invention comprises a base sequence derived from preferably human antibody, more preferably human IgG, further preferably human IgG4.

By transferring the vector comprising a nucleotide sequence encoding the antibody of the present invention, or the gene into a cell, a cell that produces the antibody of the present invention can be manufactured. A method of transfer may be according to publicly known methods, and the specific methods are illustrated in EXAMPLES.

As the fifth embodiment of the present invention, a cell that produces the antibody of the present invention is provided. Examples of such cell include hybridoma, transformant, or recombinant cell, in which a gene encoding the antibody of the present invention is introduced. Hybridoma that produces the antibody specifically includes clones listed in Tables 6 and 11. In addition, the present invention provides an antibody that is produced by the abovementioned cells of the present invention.

The eighth and the ninth embodiments of the present invention provide a novel peptide or polypeptide relating to GPVI. These peptides can be prepared by publicly known methods, and the specific method is shown in EXAMPLES. The peptide of the eighth embodiment and the polypeptide of the ninth embodiment can be used as an antigen for immunization, or as an antigen for detection of the anti-GPVI antibody.

The sixth embodiment of the present invention provides a method of manufacturing an antibody. The method of preparing the antibody of the present invention is not always limited, and the antibody can be prepared by the method described below. That is, by administering human GPVI, its fragment or derivatives thereof, for example, human GPVI-Fc as an antigen to animal such as mouse, collecting lymphocytes from the peripheral blood, hybridoma with mouse myeloma cell may be prepared. An antibody that is produced by the prepared hybridoma is recovered to select the antibody having a binding ability to GPVI having characteristics of the first to the third antibody. Thereby, a cell, in which the antibody is produced, can be obtained. By culturing the cell, the antibody of the present invention can be obtained.

The antibody of the present invention can be prepared as the recombinant human antibody using publicly known methods (a lot of methods are developed since Nature, 312: 643, 1984, and Nature, 321: 522, 1986 were published, respectively). Firstly, from the cells that produce the antibody of the present invention, e.g. the lymphocytes, preferably, the hybridoma which produces anti-GPVI monoclonal antibody, nucleic acid encoding VH or VL, e.g. cDNA may be obtained, and the base sequence and the amino acid sequence are determined. Then, by inserting the obtained cDNA encoding VH and VL into the expression vector for animal cells comprising a gene encoding human antibody CH and/or human antibody CL that has been prepared from the same or other human cell, respectively, human antibody-expressing vector may be constructed. By introducing the vector into the animal cell and expressing it, the antibody of the present invention can be manufactured. A method of preparing the gene to be introduced to the animal cell is not limited, and may be obtained from genomic DNA or cDNA derived from hybridoma, by PCR from mRNA of hybridoma, or also by the chemical synthesis.

The vector, into which to the nucleic acid encoding VH or VL of the antibody of the present invention is incorporated, is not always limited, but a vector or a vector for high expression that generally is used for expression of gene encoding protein and adapted to expression of the antibody gene is preferred. Preferred example includes a vector containing EF promoter and/or CMV enhancer, specifically pEF-BOS or the vector used in EXAMPLES. In addition, the vector that the nucleic acid encoding VH or VL is incorporated is usually prepared independently, and co-transfected into host cells. However, the nucleic acid may be incorporated into a single expression vector.

The host cell, in which the expression vector is introduced, is not always limited, and the cell that generally is used for expression of gene encoding protein and adapted to expression of the antibody gene is preferred. For example, bacteria (Escherichia coli, etc.), actinomyces, yeasts, insect cell (SF9, etc.), mammalian cell (COS-1, CHO, myeloma cell, etc.) are included.

To industrially produce a recombinant antibody, generally a recombinant animal cell line that stably and highly expresses the antibody, for example, a CHO cell line may be utilized. For preparation of such recombinant cell lines, cloning, gene amplification for high expression and screening, a publicly known method can be used (see, e.g., Omasa T.: J. Biosci. Bioeng., 94, 600-605, 2002; and others). Further, two kinds of promoters can be used for establishment of animal cell line for high expression. Herein, by using combination of different promoter activities such as high activity and low activity, preferably utilizing a promoter having a relatively weak activity as an expression promoter for selection marker, a clone having an high expression ability can be efficiently (or selectively) acquired. Examples of preferred combination of promoters and specific methods are indicated in EXAMPLES.

As the constant region of the human antibody to use for preparing a recombinant human antibody, any human antibody constant region, for example, Cγ1 and Cγ4 for the human antibody heavy chain constant region, and Cκ for the human antibody light chain constant region can be used.

Among the antibodies of the present invention, the antibody comprising an amino acid sequence derived from human includes, as well as an antibody that is naturally occurring in the human body, a combinatorial library consisting of variable heavy chain and variable light chain, for example, a phage library for human antibody, and an antibody obtained from human antibody producing-transgenic animal, etc. The phage library for human antibody is a library, wherein an active fragment of antibody such as Fab and single-stranded antibody is made express on the phage surface by inserting a gene encoding the antibody prepared from human B cell into phage gene. The antibody of the present invention can be obtained by screening these libraries. These and other methods are well-known by persons skilled in the art (Huse et al., Science 246: 1275-1281 (1989); Winter and Harris, Immunol. Today 14: 243-246 (1993); Ward et al., Nature 341: 544-546 (1989); Harlow and Lane, supra, 1988); Hilyard et al., Protein Engineering: A practical approach (IRL Press 1992); Borrabeck, Antibody Engineering, 2nd Ed. (Oxford University Press 1995)). From the library, using an binding activity to the substrate immobilized an antigen as an index, a phage expressing an active fragment of the antibody having the desired antigen binding activity can be recovered. The active fragment of the antibody can further be converted into a human antibody molecule consisting of two intact H-chains and two intact L-chains by genetic engineering techniques.

The present invention includes, in addition to the antibody consisting of two heavy chains and two light chains, the active-fragment of the antibody of the present invention. For example, the active fragment of the antibody includes Fab (fragment of antigen binding), Fab′, F(ab′)2. The substance that the active fragment of the antibody is linked by linker and so on includes, for example, single-stranded antibody (single chain Fv: scFv) and disulfide stabilized Fv: dsFv). The peptide that contains the active fragment of the antibody includes, for example, a peptide containing CDR. These can be manufactured by the method of processing the antibody of the present invention with the suitable protease or the publicly known methods such as the recombinant DNA techniques.

Fab of the present invention can be obtained by treating the anti-GPVI antibody of the present invention with pepsin, the proteolytic enzyme in case of IgM, or by processing it with the protease papain in case of IgG. Alternatively, Fab can be produced by inserting DNA encoding Fab of the antibody into prokaryotic or eukaryotic expression vector, introducing the vector into procaryotes or eukaryotes, and expressing the same.

F(ab′)2 of the present invention can be obtained by treating the anti-GPVI antibody of the present invention with pepsin, the proteolytic enzyme. Alternatively, it can be prepared by linking the following Fab′ with the thioether bond or the disulfide bond.

Fab′ of the present invention can be obtained by treating F(ab′)2 that specifically reacts to GPVI with the reducing agent, dithiothreitol.

As VH and VL that are comprised in scFv of the present invention, those derived from either the antibody or the human antibody, which the hybridoma of the present invention produces, can be used. The scFv of the present invention can be manufactured by obtaining cDNA encoding VH and VL of the anti-GPVI antibody of the present invention, constructing the DNA encoding scFv, inserting the DNA into prokaryotic expression vector or eukaryotic expression vector, and introducing the vector into prokaryotes or eukaryotes to express the vector.

The term “dsFv” means an antibody that two polypeptides, wherein one amino acid residue in each of VH and VL is substituted with cysteine residue, bind each other via disulfide bond between said cysteine residues. The amino acid residue to be substituted for the cysteine residue can be selected based on the three-dimensional structure prediction of the antibody according to the method shown by Reiter et al. [Protein Engineering, 7, 697 (1994)]. As for VH and VL that is contained in dsFv of the present invention, the ones, which are derived from the antibody of either the first or second embodiment of the present invention may be used.

The dsFv of the present invention can be manufactured by obtaining cDNA encoding VH and VL of the anti-GPVI antibody of the present invention, constructing DNA encoding dsFv, inserting the DNA into prokaryotic or eukaryotic expression vector, and introducing the expression vector into prokaryotes or eukaryotes to express the same.

The peptide containing CDR is constituted by including at least one region or more of the H-chain CDR or of L-chain CDR. Plural CDRs can be bound directly or through the appropriate peptide linker. The peptide that contains CDR of the present invention can be manufactured by obtaining cDNA encoding VH and VL of the anti-GPVI antibody of the present invention, constructing DNA encoding CDR, inserting the DNA into prokaryotic or eukaryotic expression vector, and introducing the expression vector into prokaryotes or eukaryotes to express the same. Alternatively, the peptide that contains CDR can be manufactured by the chemical synthesis such as the Fmoc method (the fluorenylmethyloxycarbonyl method), the tBoc method (t-butyloxycarbonyl method), or the like.

The antibody of the present invention, its active fragment or derivatives thereof include, for example, an antibody that is produced by hybridoma, an antibody that is produced by the cell transformed with EBV, a recombinant antibody expressed from cDNA, or an antibody, wherein a radioisotope, protein, peptide or low molecular, etc. is chemically conjugated or through genetic engineering fused with the active fragment of the antibody. For example, an antibody bound to polyethyleneglycol etc. is highly useful in respect of stability, and is one of preferred examples. To the N-terminal or the C-terminal end of H-chain or L-chain in the anti-GPVI antibody of the present invention or an active fragment thereof, an appropriate residue or a side chain in the antibody or an active fragment of the antibody, and further a sugar chain in the antibody or an active fragment of the antibody, a radioisotope, protein, peptide or low molecular weight compound, etc. can be conjugated by a chemical method [The introduction to antibody engineering (Koutai kougaku nyuumon) (written by Osamu Kanemitsu, 1994, Chijin Shokan)].

Hybridoma means a cell that produces the monoclonal antibody having the desired antigen specificity, wherein it is obtained by fusing a lymphocyte with the myeloma cell derived from human, mouse, rat and so on, and can be prepared by publicly known methods.

When preparing a monoclonal antibody, in consideration of the compatibility with myeloma cell used for the cell fusion, selection is preferably performed. As for the myeloma cell, publicly known various cells are usable. These include SKO-007 from human, SHM-D33, which is a human-mouse heterozygous myeloma, P3, P3U1, SP2/O, NS-1 derived from mouse, and YB2/0 and Y3-Ag1 through Ag3 from rat.

In the case of human antibody, a method of preparing hybridomas utilizing activation of lymphocytes by in vitro immunization and a method of preparing hybridomas using an animal, in which human antibody gene is recombined, in particular, a transgenic mouse such as KM mouse are included (WO 2002/070648 (Tokuhyou 2005-504507) and WO 2002/043478 (Tokuhyou 2004-515230)). When using a heterogeneous GPVI, for example, a protein, wherein a partial amino acid sequence of human GPVI is integrated into mouse GPVI, the above-mentioned transgenic animal such as a mouse is immunized, it will be considered that a human antibody that reacts with the integrated amino acid sequence from human, preferably epitope, but not the amino acid sequence from mouse on GPVI can be obtained efficiently. Therefore, the human antibody obtained by such method is useful as a human antibody having the features of the antibody of the first or the second embodiment, and said method is particularly useful. A cell used for preparation of hybridomas is not always limited. In the case of preparing human antibody, among plural cells for preparation of hybridomas, at least one is preferably a cell derived from human. As the cell derived from human, lymphocytes from peripheral blood, lymph nodes or spleen can be used, and especially, human lymphocytes, in which production of autoantibody is confirmed, is preferred.

Activation of lymphocyte can be according to the publicly known method. For example, preferred are a method of preparing hybridoma with the myeloma cell derived from the human B cell or the mouse myeloma cell by collecting B cell from peripheral blood or spleen of the human and stimulating an antigen with in vitro immunization, a method of fusing with the mouse myeloma cell by transforming with EBV, and a method of fusing by stimulating with mitogen such as PWM and activating B cell to polyclonal antibody (Immunological experiment procedures (Men-eki jikken sousa-hou) I and II, edited by Shunsuke Migita et al., Nankoudo).

The antigen used for immunizing an animal or for stimulation of the cell is not always limited. The animal, from which the protein as an antigen is originated, can be appropriately selected for any purpose of the antibody. The protein as an antigen may be naturally occurring product, genetically engineered product, chemically synthesized product, or fusion protein with other protein or peptide, and the like. For example, the platelet, the membrane of the platelet, purified GPVI, recombinant GPVI, and GPVI-Fc, preferably GPVI-Fc can be used. In addition, the peptide of the eighth embodiment of the present invention and the polypeptide of the ninth embodiment can appropriately be used as an antigen for immunization to prepare an anti-GPVI antibody.

Fusion of the activated lymphocytes with myeloma cells can be performed using the publicly known methods such as the method by Milstein et al. (Methods in Enzymol., volume 73, pages 3). The methods include, for example, the method using polyethylene glycol (PEG) as a fusing agent (Introduction to the monoclonal antibody experiment procedure (Tan-kuron koutai jikken sousahou nyuumon), written by Tamie Ando and Takeshi Chiba, Kodansha) or the electrofusion method. The mixing ratio of the immunocyte and the myeloma cell is not limited as long as it is the ratio that the cells can be fused. Preferably, 1/10 to equal amount of the myeloma cells to the activated lymphocytes may be used. In the cell fusion using PEG (average molecular weight: 1,000-4,000), PEG concentration is not always limited, but 50% is preferable. In addition, as the fusion efficiency accelerator, auxiliary substance such as dimethylsulfoxide (DMSO) may be added. The fusion is started by adding pre-warmed PEG solution at 37° C. to the mixed cells, and is terminated by adding medium after the reaction for 1-5 minutes.

The hybridoma, which was formed by this fusion is cultured for one (1) to ten (10) days in selection medium such as the medium containing hypoxanthine, thymidine and aminopterin (HAT medium) to isolate unfused cells. The obtained hybridoma is further selected based on the antibody to be produced. The selected hybridoma is isolated to a single clone by the publicly known limiting dilution. Thereby, a monoclonal antibody-producing hybridoma is established.

For a method of detecting the activity of the antibody that is produced by the hybridoma, the publicly known method can be used. Herein, the activity of the antibody is detected in the following two steps: the binding ability to GPVI antigen as the first step and the activity of inhibiting the binding of GPVI and collagen as the second step. Examples of the detection method for the first step include ELISA, Western blotting, radioimmunoassay and the like.

As the detection method for the second step, ELISA (inhibiting the binding), protein interaction analysis (BIACORE and so on), and a platelet aggregation suppression assay are given. In addition, the peptide of the eighth embodiment of the present invention and the polypeptide of the ninth embodiment can be used as an antigen for detecting the anti-GPVI antibody. The specific method is illustrated in EXAMPLES.

The established hybridoma can be cultured by publicly known method, and from the culture supernatant, a monoclonal antibody can be obtained.

The antibody can be purified using the publicly known purification means such as the salting-out method, gel filtration, ion exchange chromatography or affinity chromatography.

The concentration of the antibody can be measured by the publicly known quantification method of protein, e.g. the measurement of absorbance at 280 nm absorbance.

For a method of confirming the antigen binding property of the anti-GPVI antibody of the present invention or a method of detecting GPVI in the biological sample using the anti-GPVI antibody of the present invention, fluorescence antibody technique, enzyme linked immunosorbent assay (ELISA), radioimmunoassay (RIA), immunohistochemical method such as immunohistological staining and immunocytological staining (ABC method, the CSA method and so on), the Western blotting method, immuno-precipitation method, enzyme-linked immunoassay as described above, sandwich ELISA method [Tan-kuron koutai jikken manual (The Monoclonal Antibody Experiment Manual) (Kodansha Scientific, 1987), Zoku seikagaku jikken kouza (The Continued Biochemical Experiment Course 5) Menneki-seikagaku kenkyuuhou (Immunobiochemical Research Method) (TOKYO KAGAKU DOZIN, 1986)] can be used.

The antibody of the present invention is an antibody that specifically binds to human GPVI. The antibody of the present invention, the active fragment of the antibody, the modified antibody that binds to the chemicals or the composition that comprises these mixtures have a variety of uses including prevention, diagnosis and treatment of the human diseases, and detection of human GPVI in test sample, cells, tissues and the like.

Since the antibody of the present invention has a high specificity for binding to GPVI, and solely has little activity that enhances or induces human platelet activation and/or thrombocytopenia, in particular, it is useful for prevention and/or treatment of human diseases, for example, diseases caused by activation or aggregation of platelet, or vascular endothelial disorders or arteriosclerotic reaction. In addition, it can be used for prevention and/or treatment of diseases caused by thrombus or embolus such as thrombosis, embolism and the like. Examples of these diseases include venous thrombosis as well as arterial thrombosis, or cerebral infarction caused by atrial fibrillation.

Specific examples of human diseases or clinical conditions that are able to prevent and treat by the antibody of the present invention include myocardial infarction, vascular endothelial hypertrophy, restenosis of blood vessel, angina pectoris or myocardial infarction at the time of thrombolytic therapy, percutaneous transluminal coronary angioplasty (PTCA), stent operation, bypass surgery or artificial blood vessel operation, or thereafter; atrial fibrillation or atrial flutter, and thrombosis, embolism or cerebral infarction caused by these diseases; thromboangiitis obliterans; acute arterial occlusion; arteriosclerosis obliterans or deep venous thrombosis etc.; cerebral infarction (atheromatous thrombotic infarction, lacunar infarction, cardiogenic infarction); transient cerebral ischemic attack; cerebrovascular spasm after subarachnoid bleeding; pulmonary thrombosis; pulmonary embolism; vascular purpura; idiopathic thrombocytopenic purpura; thrombotic thrombocytopenic purpura; disseminated intravascular coagulation; prevention of blood coagulation at the time of extracorporeal circulation; systemic lupus erythematosus; multiple arteritis; antiphospholipid antibody syndrome; purpura nephritis; endothelial cell injury associated with diabetes mellitis; diabetic nephritis; diabetic retinopathy; nephritic embolism; complications associated with transplantation (veno-occlusive disease of liver, graft-versus-host disease); and so on.

The antibody of the present invention can be administered to the diseases to be objective for the aforementioned prevention and/treatment solely or in combination with other pharmacologically active ingredient. Examples of such pharmacologically active ingredient include publicly known thrombolytic agents such as tissue plasminogen activator (t-PA) and derivatives thereof (including mutant or so-called second generation); urokinase, streptokinase or publicly known platelet inhibitor (e.g. aspirin, ticlopidine, clopidogrel, thromboxane antagonist, thromboxane synthesis inhibitor, GPIIb/GPIIIa antagonist); publicly known anticoagulant (e.g. warfarin, heparin, low molecular weight heparin, pentasaccharide, thrombin inhibitor, FXa inhibitor, FVIIa inhibitor); and the like. Herein, the term “combination” includes the case where a combined formulation comprising both the antibody of the present invention and the pharmacologically active ingredient is administered and the case where the antibody of the present invention and the pharmacologically active ingredient are administered as an independent formulation at the same time or time difference, and for the dosage form, no object as long as both exist simultaneously in the blood of a patient.

A medicament comprising the antibody of the present invention and pharmaceutically acceptable composition as an effective ingredient can be prepared as tablet, injectable solution, powdered drug, suppository and the like using carrier for formulation, excipient and other additives used generally, and administered to human and other animals.

When applied to the human, route for administration includes oral, intravenous (bolus, continuous drip, intermittent drip), subcutaneous, intramuscular, intra-articular, transdermal, and transnasal administration. In general, oral administration or intravenous administration may be used. Clinical dose of the antibody of the present invention to the human may appropriately be determined in consideration of conditions, body weight, ages, sex, etc. of the patient to be administered. In general, to adult, by intravenous administration, dose of 1-10000 mg, preferably 10-1000 mg per day may be used, and the amount can be administered at one time or within several times. Since dosage varies depending on various conditions, there is a case where the amount less than the above-mentioned range is effectively administered.

Herein, the antibody of the present invention includes various antibodies having different mechanisms meanwhile sharing recognition of GPVI in common. For example, since in the antibody that directly inhibits the binding of GPVI to collagen, or that suppresses activation and/or aggregation of platelet by cleaving GPVI, a relatively immediate effect can be expected, there have possibilities that the antibody is useful at least in the acute period of the disease (for example, in the time of myocardial infarction or PTCA implementation, or right before or after the events). In such a case, preferably, to make the antibody of the present invention bind to most of GPVI on the surface of platelet in the blood, relatively massive antibodies can be administered, e.g. by a single or divided intravenous injection or an intravenous drip. Also, in the antibody, in which GPVI is incorporated internally, a continuous effect, but not an immediate effect can be expected considering a life-time of human platelet in the blood (about 9-10 days) and a half-life period of human antibody in the blood (in the case of IgG, several weeks). Thus, for example, there have a possibility that it is useful in chronic stage of diseases (several days to several months after development of myocardial infarction or PTCA implementation). In such a case, the antibody, whose amount is necessitated for depleting GPVI on the platelet surface to the extent of inhibiting a reactivity of platelet in the blood against collagen partially, preferably fully, can be administered at intervals of relatively long duration, e.g. from several days to several weeks per cycle, for example, by single dose or divisional intravenous injection, or the intravenous drip infusion. Therefore, in the preferred embodiment, the antibody of the present invention may possess these effects in parallel. In addition, a treatment, wherein multiple anti-GPVI antibodies that respective effects can be expected are combined, may be performed.

A composition for parenteral administration generally includes a solution of the immunoglobulin dissolved into the acceptable carrier, preferably aqueous carrier or its mixture. Various aqueous carriers such as water, buffer solution, a phosphate buffered saline (PBS), 0.4% of saline, 0.3% of glycine, human albumin solution and the like can be used. These solutions are aseptic, and generally, microparticle material does not exist in these solutions. These compositions can be sterilized by the conventional and well-known method of sterilization. To approximate physiological conditions, the compositions may be included, on demand, pharmacologically acceptable auxiliary substance such as pH adjusting and buffering agent, toxicity regulating agent, and the like, specifically sodium acetate, sodium chloride, potassium chloride, calcium chloride and sodium lactate. The concentration of the antibody in these formulations can vary extensively, i.e. change from less than about 0.005% by weight (usually, at least about 1% by weight) to the large quantity, i.e. 15% or 20% by weight, and may be selected according to the selected and specific mode of administration, mainly based on the volume of the solution, the viscosity and the like.

An actual method of preparing a parenteral administration composition is publicly known or obvious for persons skilled in the art, and further is described in detail in Remington's Pharmaceutical Sciences (15th Edition, Mack Publishing Company, Easton, Pa., 1980), wherein the reference is incorporated by reference in its entirety. A composition suitable for wash (lavage) or the other route is selected according to the intended specific use. Some pharmaceutical compositions can include an anti-GPVI antibody and the other treatment agent which is regularly used in the disease. In any cases, the bolus administration and the continuous administration can be applied. In addition, an effective amount for prevention or treatment is arbitrarily determined depending on object diseases, clinical condition and the condition of the patient and the like.

The antibody of the present invention may be frozen or lyophilized for storage, and prior to use, can be reconstituted in the suitable carrier. This technique is known to be effective in the conventional immunoglobulin. Further the publicly known techniques for lyophilization and reconstitution can be used. It is recognized for persons skilled in the art that lyophilization and reconstitution bring about activity loss of the antibody at various degrees (e.g. in the conventional immunoglobulin, IgM antibody, larger activity loss than that of IgG antibody tends to occur), and that use level may have to be regulated for compensation of loss.

A method of detecting GPVI in test sample using the antibody of the present invention or an active fragment thereof comprises the process, in which the test sample is brought in contact with the antibody of the present invention or an active fragment thereof, and the process, in which GPVI in the test sample bound to the antibody of the present invention or an active fragment thereof is detected. The process for quantifying GPVI in the test sample may further be comprised. Using the method of detecting GPVI in test sample, diseases can be diagnosed. In particular, it is possible to use for diagnosing human diseases such as thrombotic, embolic or arteriosclerosis diseases.

Examples of the method of detecting GPVI in the test sample using the antibody of the present invention includes, but is limited to, the sandwich ELISA system, the inhibition-ELISA system, fluorescence antibody method, immunohistochemical staining method, radioisotope-labeled antibody method, Western blotting method, immunoprecipitation method and so on. As a target test sample, a biological sample is used, but is not limited. Examples of the sample include body fluid, tissues or cells from animal, particularly human, bacterial cells, and extracts thereof, culture supernatant, smears and sections, but preferred is platelet.

In addition, measurement of GPVI on the platelet can be applied to monitoring of therapy associated with GPVI, especially prediction or determination of the effect of anti-GPVI antibody, or prognostic determination using disappearance of GPVI on the platelet as an index.

By the following examples, the present invention will further be described in detail. However, the present invention should not be understood by the limitation of these examples.

A. Preparation of Human GPVI Extracellular Region-Mouse Fc Fusion Protein (hGPVI-mFc) (1) Construction of the Expression Plasmid for hGPVI-mFc Fusion Protein

Using mouse genomic DNA as a template, gene regions encoding respective domains of mouse immunoglobulin (mIgG2a) heavy chain constant region were amplified. That is, with the following primer pair, PCR reaction was performed. As a result, CH1 domain, hinge region, CH2 domain and CH3 domain were amplified with mIgG2a-a (SEQ ID NO: 152) and mIgG2a-c (SEQ ID NO: 154), mIgG2a-b (SEQ ID NO: 153) and mIgG2a-e (SEQ ID NO: 156), mIgG2a-d (SEQ ID NO: 155) and mIgG2a-g (SEQ ID NO: 158), and mIgG2a-f (SEQ ID NO: 157) and mIgG2a-h (SEQ ID NO: 159), respectively. Then these four amplified products were mixed, and PCR reaction using primers mIgG2a-a and mIgG2a-h was performed to obtain an amplified product, wherein respective domains were ligated (DNA fragment encoding heavy chain constant region (Cγ2a)). After the amplified product was cloned into pT7-BlueT vector, DNA fragment encoding mouse Fc region was excised with restriction enzymes Bam HI and Kpn I to obtain fragment A. On the other hand, from pCAGGS-GPVI-Fc plasmid, DNA fragment encoding an extracellular domain of human GPVI was excised with restriction enzymes Xba I and Bgl II to obtain fragment B. These fragments were ligated to the downstream of EF promoter in the expression vector pEF2cew, which was prepared by cleavage with Xba I and Kpn I, so that fragment A+fragment B was made. As a result, a plasmid (pTK-2249) that expresses hGPVI-mFc (SEQ ID NO: 222) was constructed.

(2) Expression and Purification of hGPVI-mFc Fusion Protein

Cos-1 cells were cultured by passage with Dulbecco's MEM medium supplemented with 10% fetal bovine serum. Transfection was performed by mixing an appropriate amount of pTK-2249 with transfection reagent (FuGENE6, Roche Diagnostics), falling the above mixture in drop to serum-free Dulbecco's MEM medium, and replacing this with culture fluid. Under condition of 5% CO2, the cells were cultured at 37° C. for three days. The culture supernatant was subjected to the purification through Protein A column (Prosep-A, MILLIPORE), and the purified preparation was used as an antigen for preparation of anti-GPVI antibody.

B. Construction and Expression of Expression Plasmid pTK-2233 for Human GPVI Extracellular Region-Human Fc Fusion Protein

Fragment J that is obtained by cleaving the plasmid pCAGGS-GPVI-Fc harboring a gene encoding human GPVI-hFc with restriction enzymes Xba I and Eco T221, and fragment K that is obtained by cleaving it with restriction enzymes Eco T221 and Bgl II was prepared respectively. By ligating the two fragments to the downstream of EF-1α promoter on the expression vector pEF2cew to make fragments J+K, a hGPVI-hFc expressing plasmid pTK-2233 was constructed. In addition, the expression and purification of hGPVI-hFc was performed in a similar manner of the case for hGPVI-mFc.

C. Construction of Expression Plasmid pTK-2440 for Mouse GPVI Extracellular Region-Human Fc Fusion Protein (mGPVI-hFc)

Using mouse genomic DNA as a template and primer pairs listed, PCR reaction was performed. As a result, the PCR amplified product ‘hi’ was obtained with mGPVI-h (SEQ ID NO: 162) and mGPVI-i (SEQ ID NO: 163); the PCR amplified product ‘jk’ with mGPVI-j (SEQ ID NO: 164) and mGPVI-k (SEQ ID NO: 165); the PCR amplified product ‘lm’ with mGPVI-l (SEQ ID NO: 166) and mGPVI-m (SEQ ID NO: 167); the PCR amplified product ‘no’ with mGPVI-n (SEQ ID NO: 168) and mGPVI-o (SEQ ID NO: 169); and the PCR amplified product ‘pc’ with mGPVI-p (SEQ ID NO: 170) and mGPVI-c (SEQ ID NO: 160). Using a mixture of these amplified products as a template and a mixture of primers mGPVI-e (SEQ ID NO: 161), mGPVI-q (SEQ ID NO: 171), mGPVI-r (SEQ ID NO: 172), mGPVI-s (SEQ ID NO: 173) and mGPVI-c, PCR reaction was performed to obtain an amplified product, wherein each fragment is ligated. This amplified product was cloned into pT7-BlueT vector and designated pTK-2437. Since in the gene region including mouse GPVI extracellular domain in pTK-2437, there have Nhe I restriction enzyme recognition site at 5′ side and Bam HI restriction enzyme recognition site at 3′ side, DNA fragment was prepared by cleavage with these enzymes. Then the fragment was inserted into pTK-2233, which is prepared by cleaving with restriction enzymes Xba I and Bam HI, to construct a mouse mGPVI-Fc expressing plasmid, pTK-2440.

Based on the information of human GPVI gene that is a known sequence, an appropriate primer pair was designed and prepared. Then using these primers derived from human GPVI, PCR with cynomolgus monkey genomic DNA as a template was performed to determine a part of gene sequence encoding cynomolgus monkey GPVI. Then, based on the obtained sequence, novel primer pair for cynomolgus monkey was designed and prepared. Using each primer pair and cynomolgus monkey genemic DNA as a template, PCR reaction was performed again. As a result, the amplified product ‘ab’ was obtained with macGPVI-a (SEQ ID NO: 174) and macGPVI-b (SEQ ID NO: 175); the amplified product ‘dc’ with hGPVI-d (SEQ ID NO: 180) and macGPVI-c (SEQ ID NO: 176); the amplified product ‘dh’ with macGPVI-d (SEQ ID NO: 177) and hGPVI-h (SEQ ID NO: 182) and the amplified product ‘gg’ with hGPVI-g (SEQ ID NO: 181) and macGPVI-g (SEQ ID NO: 178). Meanwhile, using pTK-2233 as a template and macGPVI-h (SEQ ID NO: 179) and IgG1-i (SEQ ID NO: 183), PCR reaction was performed to obtain the amplified product ‘hi’. Using a mixture of the above-mentioned five amplified products as a template, PCR reaction with macGPVI-a and IgG1-i was performed once again. The amplified product obtained by this procedure includes a chimeric GPVI gene, in which cynomolgus monkey GPVI D1 and D2 and human D3 are fused. After cleavage with restriction enzymes Nhe I and Bam HI, the amplified product was inserted into pTH-2233 prepared by cleavage with restriction enzymes Xba I and Bam HI to construct an expression plasmid, pTK-2462 for cynomolgus monkey D1D2-human D3 chimeric GPVI-human Fc fusion protein (GPVI-FFH-hFc; SEQ ID NO: 223).

To prepare a polyclonal antibody to human GPVI, rabbit was immunized. That is, 20 μg of hGPVI-mFc prepared in EXAMPLE 1A was diluted with 500 μl of saline and mixed with 500 μl of Freund complete adjuvant (DIFCO). Then, the mixture was subcutaneously administered to the dorsal of female New Zealand white rabbit (KITAYAMA LABES) whose weight was 2.1-2.2 kg. Two weeks later, once again, 20 μg of hGPVI-mFc was diluted with 500 μl of saline and mixed with 500 μl of Freund incomplete adjuvant (DIFCO). Then, the mixture was subcutaneously administered to the dorsal. One week after the completion of administration, the blood was collected from the ear vein, and according to the conventional method, anti-serum was isolated to purify an antibody. That is, to the anti-serum ammonium sulfate was added to make the final concentrations of saturated solution 33%. After stirring at 4° C. for one hour, the precipitate that appeared was isolated by centrifugation. Next, the precipitate was dissolved in the Dulbecco phosphate buffer (hereinafter, referred to as D-PBS) and dialyzed against D-PBS for overnight. After filtration of the dialysate, it was applied to Protein A column (Prosep A, Millipore) to obtain a purified antibody by eluting a bound IgG fraction with 0.1-M Glycine-hydrochloride buffer solution (pH3.0). The obtained eluate fraction was immediately neutralized in 1 M Tris-HCl buffer (pH7.0) and dialyzed against D-PBS. After dialysis, protein concentration was calculated from the absorbance at 280 nm (The absorbance coefficient: 0.714 mg/mL). Hereinafter, the obtained antibody is denoted as the anti-GPVI polyclonal antibody.

By mixing 20 μg of GPVI-mFc and Freund complete adjuvant (DIFCO) at equal volume, an antigen for administration was prepared. The antigen was administered twice to female ddY mouse (8-week age, SLC), and 3 days later, lymphocyte was isolated from the lymph node. The obtained lymphocyte was mixed with P3×63-Ag. 8. U1 (ATTC), and according to “The Introduction to Monoclonal Antibody Experimental Procedure” written by Tamie Andoh and Takeshi Chiba (Kodansha, p 83), using polyethylene glycol (PEG1500, Sigma) a cell fusion was performed. Hybridoma was selected by HAT medium and one week later, the screening of the hybridoma that produces a desired antibody was performed by two ways. That is, a method of using a binding activity to hGPVI-hFc immobilized to a plate as an index, and a method of using an inhibiting activity for binding of collagen with GPVI as an index were utilized.

The hGPVI-hFc prepared by EXAMPLE 1B was diluted with D-PBS to 2 μg/mL and added to immunoplate (Maxisorp NUNC) at 50 μL/well. After incubation at 37° C. for one hour, the well was washed five times with ion-exchanged water, and D-PBS (pH7.4) containing 2% StabilGuard (Surmodics) was added to each well in 100 μL for blocking. Next, culture supernatant was added to each well and incubated at 37° C. for one hour. After incubation, the well was washed three times with the saline containing 0.05% Tween20. Peroxidase-labeled anti-mouse immunoglobulin antibody (DAKO, P260) was 1000-fold diluted with D-PBS containing 10% rabbit serum and added to each well at 50 μL. After incubation at 37° C. for one hour, washing in a similar manner was performed and TMB solution (BioFix) was added to each well. After ten minutes at room temperature, the reaction was terminated with 0.5 M sulfuric acid solution. Subsequently, absorbance at 450 nm was measured with the plate spectrophotometer (Multiscan JX, Dainippon Pharmaceutical). As a result, the cell whose culture supernatant reacted to hGPVI-hFc was selected and a cloning with limiting dilution method (“The Introduction to Monoclonal Antibody Experimental Procedure” written by Tamie Andoh and Takeshi Chiba (Kodansha, p 83)) was performed. Eight days later, screening was done in a similar manner and the antibody which reacts to hGPVI-hFc was selected.

(2) Screening of Hybridoma Using an Inhibiting Activity for Binding of Collagen with GPVI as an Index

Collagen (Horn) was diluted with D-PBS to 10 μg/mL and added to immunoplate (Maxisorp, NUNC) at 50 μl/well. After incubation at 4° C. for overnight, the well was washed five times with ion-exchanged water and blocked with D-PBS containing 5% BSA. Next, culture supernatant was added to each well at 25 μL/well. Further, hGPVI-hFc prepared to 2 μg/mL with D-PBS was added to the well at 25 μL/well. After incubation at 37° C. for one hour, the well was washed three times with the saline containing 0.05% Tween20. Peroxidase-labeled anti-human IgG antibody (BioMeda) was 1000-fold diluted with D-PBS containing 10% goat serums and added to each well at 50 μL. After incubation at 37° C. for one hour, washing in a similar manner was performed and TMB solution (BioFix) was added to each well. After ten minutes at room temperature, the reaction was terminated with 0.5 M sulfuric acid solution. Subsequently, absorbance at 450 nm was measured with the plate spectrophotometer (Multiscan JX, Dainippon Pharmaceutical). The well, wherein the absorbance decreased by equal to or more than 50% in comparison with the well without the antibody, was selected as a hybridoma that produces an anti-GPVI antibody.

As a result, by performing the multiple cell fusion (the F number shows 1 batch), hybridoma having an inhibiting activity for the binding of collagen with GPVI was selected.

(3) Preparation of Antibody that is Produced by Hybridoma

Hybridoma that produces an anti-GPVI antibody was cultured in 10% FCS/RPMI-1640 medium (Sigma). After replacement of the medium with Hybridoma-SFM medium (Invitrogen), hybridoma was further cultured to produce the antibody. From the culture supernatant, using Protein A column (Prosep-rA, Millipore), the antibody was purified. That is, the obtained culture supernatant was adsorbed onto Protein A column (Prosep-A, Millipore) preliminarily equilibrated with D-PBS. After washing non-adsorbed protein with D-PBS, adsorbed fraction was eluted with 25 mM Glycine-hydrochloride buffer (pH3.0). Thereafter, the fraction was dialyzed against saline. Concentration of the obtained antibody was calculated from absorbance at 280 nm using an absorption coefficient (E1%:1.4). Hereinafter, the obtained antibody is referred to as the anti-GPVI monoclonal antibody.

In order to classify the antibody obtained in EXAMPLE 2 by the binding property to GPVI, i.e. the difference of the binding region, using F1199, F1201, F1202, F1210 and F1211 antibodies, respectively, competitive assay was performed. Firstly, according to the method described by Nakane et al. (J. Histochem. Cytochem., 22, 1084, 1974), each peroxidase-labeled anti-GPVI monoclonal antibody was prepared.

Then, using the peroxidase-labeled antibody, competitive assay for respective purified antibodies. That is, hGPVI-hFc was diluted with D-PBS to 2 μg/mL, and the diluted solution was added to immunoplate (Maxisorp, NUNC) at 50 μL/well. After reaction at 37° C. for one hour, the well was washed five times with ion-exchanged water, and D-PBS containing 2% StabilGuard was added to each well for blocking. Then, 25 μL each of the above-mentioned labeled antibody and the purified antibody were added to the well and incubated at 37° C. for one hour. After washing five times with saline containing 0.05% Tween20, color development was done by TMB solution (BioFix). After reacting at room temperature for 10 minutes, the reaction was terminated with 0.5 M sulfuric acid solution. With plate spectrophotometer (Multiscan JX, Dainippon Pharmaceutical Co. Ltd.), absorbance at 450 nm was measured.

The absorbance without a purified antibody was made 100%, and the inhibiting activity to each labelled antibody was calculated to classify each purified antibody. As a result, the one which competes with labelled antibodies (i), (ii), (v) and (viii) was designated group a. The one which competes with labelled antibodies (i), (v) and (vii) was designated group b. The one which competes with labelled antibodies (i), (ii), (vii) and (viii) was designated group c. Also, the one which competes with labelled antibodies (iii) and (iv) was designated group d. The one which competes with labelled antibodies (iii), (iv) and (vi) was designated group e. The one which competes with labelled antibodies (iii) and (vi) was designated group f. Also, the one whose regularity cannot be seen was designated groups g and h. Since the groups g and h did not compete with each other, they formed another group. Therefore, it can be confirmed that the anti-GPVI monoclonal antibody prepared recognizes at least eight regions of the surface of GPVI.

Among the prepared anti-human GPVI monoclonal antibody, an antibody that binds to monkey GPVI and is applicable for ex vivo test was classified according to the method for the group classification performed in EXAMPLE 3. That is, using each labeled antibody prepared in EXAMPLE 3, in a similar method to that of EXAMPLE 3, the competitive assay was done to classify each antibody. As shown in Table 1, just like EXAMPLE 3, the antibody was classified into seven groups of the antibody that recognizes at least seven regions.

TABLE 1 Group a F1232-10-1 Group b F1232-16-1 Group c F1232-39-3 F1232-21-1 F1232-13-3 Group d F1232-7-1 F1232-9-1 F1232-11-1 F1232-19-1 F1232-37-2 F1201-18 Group e F1232-17-1 F1232-18-3 Group f F1232-8-3 F1232-14-2 F1232-15-1 F1232-24-1 F1232-38-1 F1232-45-1 F1232-29-2 Group g F1201-20 F1232-27-1 Group h F1232-43-3

Dissociation constant for the anti-GPVI monoclonal antibody prepared in EXAMPLE 2 was measured using protein interaction analyzer BIACORE3000 (BIACORE). That is, to sensor chip, CM5 chip (BIACORE), GPVI substitution mutant (hGPVIHHH-hFc and FFH-hFc) that is prepared in EXAMPLE 1 was linked according to the manual. Then each antibody was diluted with HBS-EP buffer solution (BIACORE) to prepare a series of diluted solution from 1.25 to 40 nM, and analyzed with BIACORE3000. Every each antibody bound chip, the chip was regenerated with glycine buffer solution at pH1.5. The obtained result was analyzed with Bivalent analyte, the evaluation software (BIACORE) to calculate dissociation constant. The result was shown in Table 2. It was shown that each anti-GPVI monoclonal antibody has a sufficient affinity to GPVI-HHH-hFc.

TABLE 2 Dissociation Constant (KD) M Classification Antibody Human GPVI Monkey GPVI Group a F1232-10-2 8.33 × 10−10 1.09 × 10−9 Group c F1232-21-1 1.39 × 10−9 3.56 × 10−9 Group d F1232-7-1 3.47 × 10−8 3.43 × 10−7 F1232-19-1 1.75 × 10−9 2.35 × 10−9 F1232-37-2 4.03 × 10−10 1.00 × 10−9 F1201-18 1.16 × 10−9 6.19 × 10−10 Group e F1232-17-1 5.22 × 10−10 1.77 × 10−10 F1232-18-3 1.65 × 10−9 3.65 × 10−7 Group f F1232-14-2 1.56 × 10−8 2.14 × 10−8 F1232-24-1  2.5 × 10−10  7.1 × 10−10 Group g F1201-20 1.50 × 10−8 7.55 × 10−9 Group h F1232-43-3 1.36 × 10−9 1.69 × 10−9

In order to prepare sandwich EIA system, the anti-GPVI polyclonal antibody obtained in EXAMPLE 2 was labeled with peroxidase just like EXAMPLE 3. Then a plate immobilized with anti-GPVI polyclonal antibody was prepared. That is, the antibody was diluted with D-PBS to 10 μg/mL and 50 μL of the antibody was added to each well of immunoplate (Maxisorp, NUNC) to incubate at 45° C. for 30 minutes. Then, the well was washed five times with ion-exchanged water and to each well, 100 μL of D-PBS containing 2% StabilGuard (Surmodics) was added for blocking. As a standard sample, the purified GPVI-hFc was diluted with 0.1% BSA/D-PBS to 0.75 ng/ml, 1.5 ng/ml, 3.1 ng/ml, 6.25 ng/ml, 12.5 ng/ml, 25 ng/ml and 50 ng/ml. As a blank, D-PBS containing 0.1% BSA was used. At first, a blocking agent on the plate was discarded and 50 μL of the prepared standard sample and the blank was dispensed to each well for incubation at 25° C. for overnight. The plate was washed three times with saline containing 0.05% Tween20. Subsequently, 50 μL of the peroxidase-labeled anti-GPVI polyclonal antibody that was diluted with D-PBS containing 10% rabbit serum and 0.1% Tween20 to 10 μg/ml was added to the well to incubate at 37° C. Similarly, after washing five times, TMB solution (BioFix) was added to each well to incubate at room temperature for 20 minutes, and the reaction was terminated with 0.5 M sulfuric acid solution. With the plate spectrophotometer (Multiscan JX, Dainippon Pharmaceutical), an absorbance at 450 nm was measured to prepare standard curve.

By replacing a part of the amino acid sequence in human GPVI with the corresponding amino acid sequence of mouse GPVI and examining the change of the antibody reactivity, the recognition region of each anti-GPVI antibody was narrowed. That is, since an extracellular region of GPVI, the platelet membrane protein is composed of three domains, i.e. immunoglobulin-like regions 1 and 2 (sometimes referred to as domain 1 or D1, and domain 2 or D2) and mutin-like domain (sometimes referred to as domain 3 or D3) (FIG. 1), substitution mutants for each domain were prepared and a binding assay of anti-GPVI antibody for each substitution mutants was performed.

In addition, with the purpose of further refinement of the anti-GPVI antibody recognition region, among the proteins registered to the Protein Databank (PDB), based on the information of human NK cell activating receptor Nkp46, Ig-like transcript 2 (ILT2), whose homology of the amino acid sequence to GPVI immunoglobulin-like region is relatively high, a modeling for human GPVI was conducted to deduce the regions in which amino acid mutations can be introduced (FIG. 1). Then, expression plasmids for human GPVI mutants having amino acid substitutions in those loop regions with mouse GPVI were constructed, and the binding assay of the anti-GPVI antibodies to each mutants was performed. As a result, each domain and loop region of GPVI, which each anti-GPVI monoclonal antibody recognizes could be confirmed. In addition, in the specification, the domain or the amino acid substitution mutant of GPVI is sometimes merely referred to as GPVI substitution mutants.

TABLE 3 Amino acid sequence of each loop in domain 1 and domain 2 of soluble human GPVI Loop No. Sequence No. Amino acid sequence Loop 1 1 GPLPKP Loop 2 2 PSSLVPLEKP Loop 3 3 PPGVDL Loop 4 4 SSSRYQDQ Loop 5 5 PAMKRSLAGR Loop 6 6 QNGSLWSLPSDQ Loop 7 7 VFAKPS Loop 8 8 AQPGPAVSSGGD Loop 9 9 TRYGFDQ Loop 10 10 KEGDPA Loop 11 11 ERWYR Loop 12 12 ITVTAAHS Loop 13 13 FSSRDPYL Loop 14 14 ELVVTG

Using hGPVI-hFc (hereinafter referred to as GPVI-HHH-hFc; SEQ ID NO: 135) expressing plasmid (pTK2233) and mGPVI-hFc (hereinafter referred to as GPVI-MMM-hFc; SEQ ID NO: 136), which are prepared in EXAMPLE 1, as a template, an expressing plasmid for GPVI substitution mutant, wherein the domains of human GPVI and mouse GPVI are exchanged, was constructed.

That is, to construct an expression plasmid for GPVI substitution mutant comprising mouse GPVI domain 1 and human domains 2 and 3 (hereinafter referred to as GPVI-MHH-hFc; SEQ ID NO: 137), using pTK2440 as a template and sense primer 1 that is designed at the position upstream mouse GPVI sequence (Table 4; SEQ ID NO: 184) and anti-sense primer 3, wherein 15mer of the N-terminus of domain 2 in human GPVI is ligated to 15mer of the C-terminus of domain 1 in mouse GPVI (Table 4; SEQ ID NO: 186), PCR was performed to amplify DNA fragment of domain 1 in mouse GPVI (417 bp). Next, using pTK2233 as a template and anti-sense primer 2 that is designed at the position on hFc sequence (Table 4; SEQ ID NO: 185) and sense primer 4, wherein 15mer of the N-terminus of domain 2 in human GPVI is ligated to 15mer of the C-terminus of domain 1 in mouse GPVI (Table 4; SEQ ID NO: 187), PCR was performed to amplify DNA fragment (571 bp) comprising human GPVI domain 2, domain 3 and the N-terminus of hFc. Then, by performing PCR using the amplified two DNA fragments, sense primer 1, anti-sense primer 2, DNA fragment (973 bp), wherein the domain 1 of mouse GPVI, the domain 2 and domain 3 of human GPVI, and the N-terminal sequence of hFc, was amplified. After cutting this amplified DNA fragment with restriction enzymes XbaI and BamHI, the fragment was inserted into XbaI-BamHI site of pTK2233 to construct GPVI-MHH-hFc expressing plasmid.

With respect to the expression plasmids for GPVI substitution mutants comprising sequences of human domain 1, mouse domain 2 and human domain 3 in GPVI (hereinafter represented by GPVI-HMH-hFc; SEQ ID NO: 138) and GPVI substitution mutant comprising sequences of mouse domain 1 and domain 2, and human domain 3 in GPVI (hereinafter represented by GPVI-MMH-hFc; SEQ ID NO: 139), in a similar manner, using sense and anti-sense primers (Table 4; SEQ ID NOs: 191 and 190, respectively) that are prepared by ligating 15mer each of human and mouse GPVI sequences at the joint position of the domain in GPVI desired to replace, sense primer 1 and anti-sense primer 2, needed DNA fragment was amplified. After cleaving with restriction enzymes XbaI and BamHI, by inserting the fragment into the XbaI-BamHI site of pTK2233, the expression plasmid was constructed. In addition, the amino acid sequences of five GPVI substitution mutants were shown in SEQ ID NOs: 141-151.

(2) Construction of Expression Plasmids for GPVI Substitution Mutants Wherein a Loop Region of Human GPVI is Replaced with the Corresponding Amino Acid Sequence Derived from Mouse

Expression plasmids for GPVI substitution mutants, wherein a single loop region to be a recognition region of an anti-GPVI monoclonal antibody is replaced with the corresponding amino acid sequence of mouse GPVI, was constructed as follows. First, in the loop region L2 of human GPVI, the base sequence was replaced so that the amino acid sequence from human was substituted for that from mouse. Further, sense primer 10, wherein 11mer of human GPVI base sequence was ligated upstream from the replaced base (Table 4) and anti-sense primer 9, wherein 13mer of human GPVI base sequence was ligated downstream from the replaced base (Table 4) were prepared. Next, by performing PCR using pTK2233 as a template, and sense primer 1 and anti-sense primer 9, DNA fragment (215 bp) corresponding to the N-terminus of the human GPVI, wherein the loop region L2 is substituted for the sequence from mouse, was amplified. In a similar manner, using pTK2233 as a template, and sense primer 10 and anti-sense primer 2, PCR was performed to amplify DNA fragment (773 bp), wherein the N-terminal sequence of hFc is connected to the C-terminal region of human GPVI, in which the loop region L2 is substituted for the sequence from mouse. Then, by performing PCR using the amplified two DNA fragments, sense primer 1, and anti-sense primer 2, DNA fragment (958 bp), wherein the N-terminal sequence of hFc is connected to the C-terminal region of human GPVI, in which the loop region L2 is substituted for the sequence from mouse, was amplified. After cutting this amplified DNA fragment with restriction enzymes XbaI and BamHI, the fragment was inserted into XbaI-BamHI site of pTK2233 to construct an expressing plasmid for GPVI substitution mutant, wherein the region L2 of human GPVI is replaced with the amino acid sequence of mouse GPVI (hereinafter represented by hGPVI-mL2-hFc; SEQ ID NO: 46).

Expression plasmid for GPVI substitution mutants, wherein other loop regions are replaced, such as hGPVI-mL3-hFc, hGPVI-mL4-hFc, hGPVI-mL5-hFc, hGPVI-mL6-hFc, hGPVI-mL7-hFc, hGPVI-mL8-hFc, hGPVI-mL9-hFc, hGPVI-mL10-hFc, hGPVI-mL11-hFc, hGPVI-mL13-hFc, hGPVI-mL14-hFc, were constructed in a similar manner. The primer sequences that are used for construction of each GPVI substitution mutant-expressing plasmid were shown in Table 4. In addition, the amino acid sequences of the GPVI substitution mutants prepared, was shown by SEQ ID NOs: 47-57.

TABLE 4 Primer sets and PCR-amplified regions for the construction of expression plasmid of each substituted GPVI mutant PCR-amplified Anti-sense GPVI mutant region Sense primer primer GPVI-HHH-hFc (Seq. ID. 135) GPVI-MMM-hFc (Seq. ID. 136) GPVI-MHH-hFc Mouse domain 1 Seq. ID. 184 Seq. ID. 186 (Seq. ID. 137) Human domain 2, 3 Seq. ID. 187 Seq. ID. 185 GPVI-HMH-hFc Human domain 1 Seq. ID. 184 Seq. ID. 188 (Seq. ID. 138) Mouse domain 2 Seq. ID. 189 Seq. ID. 190 Human domain 3 Seq. ID. 191 Seq. ID. 185 GPVI-MMH-hFc Mouse domain1, 2 Seq. ID. 184 Seq. ID. 190 (Seq. ID. 139) Human domain 3 Seq. ID. 191 Seq. ID. 185 hGPVI-mL2-hFc N-terminus side Seq. ID. 184 Seq. ID. 192 (Seq. ID. 140) C-terminus side Seq. ID. 193 Seq. ID. 185 hGPVI-mL3-hFc N-terminus side Seq. ID. 184 Seq. ID. 194 (Seq. ID. 141) C-terminus side Seq. ID. 195 Seq. ID. 185 hGPVI-mL4-hFc N-terminus side Seq. ID. 184 Seq. ID. 196 (Seq. ID. 142) C-terminus side Seq. ID. 197 Seq. ID. 185 hGPVI-mL5-hFc N-terminus side Seq. ID. 184 Seq. ID. 198 (Seq. ID. 143) C-terminus side Seq. ID. 199 Seq. ID. 185 hGPVI-mL6-hFc N-terminus side Seq. ID. 184 Seq. ID. 200 (Seq. ID. 144) C-terminus side Seq. ID. 201 Seq. ID. 185 hGPVI-mL7-hFc N-terminus side Seq. ID. 184 Seq. ID. 202 (Seq. ID. 145) C-terminus side Seq. ID. 203 Seq. ID. 185 hGPVI-mL8-hFc N-terminus side Seq. ID. 184 Seq. ID. 204 (Seq. ID. 146) C-terminus side Seq. ID. 205 Seq. ID. 185 hGPVI-mL9-hFc N-terminus side Seq. ID. 184 Seq. ID. 206 (Seq. ID. 147) C-terminus side Seq. ID. 207 Seq. ID. 185 hGPVI-mL10-hFc N-terminus side Seq. ID. 184 Seq. ID. 208 (Seq. ID. 148) C-terminus side Seq. ID. 209 Seq. ID. 185 hGPVI-mL11-hFc N-terminus side Seq. ID. 184 Seq. ID. 210 (Seq. ID. 149) C-terminus side Seq. ID. 211 Seq. ID. 185 hGPVI-mL13-hFc N-terminus side Seq. ID. 184 Seq. ID. 212 (Seq. ID. 150) C-terminus side Seq. ID. 213 Seq. ID. 185 hGPVI-mL14-hFc N-terminus side Seq. ID. 184 Seq. ID. 214 (Seq. ID. 151) C-terminus side Seq. ID. 215 Seq. ID. 185

The GPVI substitution mutant-expressing plasmid constructed in EXAMPLE 7 (1) and (2), pTK2233 and pTK2440 were introduced into the COS-1 cell and expressed in a similar manner shown in EXAMPLE 1.

From culture supernatants obtained, the desired GPVI substitution mutants were purified by the Protein A column (Prosep-A, Millipore).

Purity of GPVI substitution mutants obtained was confirmed by SDS-PAGE under both reduction and non-reduction conditions, and silver staining.

(4) Binding Activity with Human-Mouse GPVI Domain Substitution mutants

By replacing the domain of human GPVI with the corresponding domain of mouse GPVI, the recognition region of the antibody was narrowed. That is, the binding activity of GPVI-HHH-hFc, GPVI-MHH-hFc, GPVI-HMH-hFc, GPVI-MMH-hFc and GPVI-MMM-hFc with each anti-GPVI monoclonal antibody was assayed.

First, rabbit anti-human IgG antibody (DAKO) was immobilized to the plate (Maxisorp, Nunc) at 5 μg/mL to plate, and the plate was blocked with 2% StabiliGuard (SurModics). Next, after removing blocking solution, the GPVI substitution mutant diluted with 0.1% BSA/PBS was added to react at 37° C. for one hour. After washing with 0.9% saline containing 0.05% Tween20, blocking with 1% human serum (Cosmo Bio) diluted with 0.1% BSA/PBS was done at 37° C. for one hour. Once again, after washing with 0.9% saline containing 0.05% Tween20, the peroxidase-labeled antibody prepared in EXAMPLE 3 was diluted to 0.5 μg/mL with 0.1% BSA/PBS, and added to the plate to incubate at 37° C. for one hour. Finally, after washing with 0.9% saline containing 0.05% Tween20, the reaction mixture was developed using H2O2/TMB solution. The reaction was terminated with 0.5 M sulfuric acid, and an absorbance at 450 nm of wavelength was measured.

As a result, in F1232-10-2, F1232-21-1, F1201-20, F1232-43-3 and F1232-14-2 antibodies, the binding activity to GPVI-MHH-hFc, GPVI-MMH-hFc and GPVI-MMM-hFc, wherein domain 1 was substituted for mouse GPVI, remarkably reduced. From the fact, it became clear that the recognition region of these antibodies existed in the domain 1 of GPVI. On the other hand, in F1232-7-1, F1232-19-1, F1232-37-2, F1232-17-1, F1232-18-3 and F1232-24-1 antibodies, the binding activity to GPVI-HMH-hFc, GPVI-MMH-hFc, GPVI-MMM-hFc, wherein domain 2 was substituted for mouse GPVI, declined remarkably. It was supposed that the recognition region of these antibodies existed in the domain 2 of GPVI.

The antigen recognition region of each anti-GPVI antibody was shown in Table 5.

(5) Binding Activity with GPVI Substitution Mutant Wherein the Loop Region in Human GPVI was Replaced with the Corresponding Amino Acid Sequence Derived from Mouse

The binding activity of GPVI-HHH-hFc, hGPVI-mL2-hFc, hGPVI-mL3-hFc, hGPVI-mL4-hFc, hGPVI-mL5-hFc, hGPVI-mL6-hFc, hGPVI-mL7-hFc, hGPVI-mL8-hFc, hGPVI-mL9-hFc, hGPVI-mL10-hFc, hGPVI-mL11-hFc, hGPVI-mL13-hFc and hGPVI-mL14-hFc with each anti-GPVI monoclonal antibody was measured.

By comparing the binding activity to GVI substitution mutant, wherein the loop region in human GPVI was replaced with the loop region from mouse, with that to human GPVI, the recognition region of the antibody was narrowed to the loop region replaced.

In a similar manner as described in EXAMPLE 7 (4), peroxidase-labeled antibody prepared in EXAMPLE 3 was diluted to a measurable concentration with 0.1% BSA/PBS in conformity with the affinity of each antibody to GPVI-HHH-hFc, and used for the measurement. As a result, in F1232-10-2, F1232-21-1 and F1232-43-3 antibodies, the binding activity to hGPVI-mL2-hFc remarkably decreased. In F1232-14-2 antibody, the activity to hGPVI-mL3-hFc and hGPVI-mL5-hFc declined. In F1201-20 antibody, the activity to hGPVI-mL4-hFc and hGPVI-mL5-hFc declined. Also, in F1232-7-1, F1232-37-2, F1232-17-1, F1232-18-3 and F1232-24-1 antibodies, the activity to hGPVI-mL9-hFc and in F1232-19-1 antibody, the activity to hGPVI-mL9-hFc and hGPVI-mL11-hFc reduced, respectively. The antigen recognition region of each anti-GPVI antibody was shown in Table 5.

TABLE 5 Recognition D1 D2 Clone No. domain L2 L3 L4 L5 L6 L7 L8 L9 L10 L11 L13 L14 F1232-21-1 D1  F1232-10-2 D1  F1201-20 D1   F1232-14-2 D1   F1232-43-3 D1  F1232-7-1 D2  F1232-19-1 D2   F1232-37-2 D2  F1232-18-3 D2  F1232-17-1 D2  F1232-24-1 D2  : means a case of results that the antibody binding with a loop-substituted hGPVI mutant is 30% or less of binding with hGPVI-hFc.

TABLE 6 Results of ex-vivo assay Depletion of platelet membrane protein Loss of platelet aggregability Clone No. Thrombocytopenia Platelet activation CD41a CD42a GPVI Collagen-induced ADP-induced F1232-7-1 + − − − ++ + − F1232-18-3 − − − − ++ ++ − F1232-24-1 + − − − ± ± − F1232-10-2 − − − − ++ ++ ± F1232-21-1 + − ± ± ++ ++ ± F1232-14-2 − − − − ± − − F1201-18 ± − − − + ++ − F1232-43-3 − − − − ++ ++ − F1201-20 − − − − + + − F1232-37-2 − − − − + ++ − F1232-17-1 ± − − − ++ ++ − F1232-19-1 ++ − ± ± ++ ++ − Thrombocytopenia: −; 20%>, ±; 20%-40%, +; 40%-60%, ++; 60%< Platelet activation: −; 2-fold, ±; 2-fold-5-fold, +; 5-fold-10-fold, ++; 10-fold< Depletion of platelet membrane protein (FACS): −; 30%>(1 mg/kg administration), ±; 30%-70%(1 mg/kg administration), +; 70%<(1 mg/kg administration), ++; 70%<(0.3 mg/kg administration) Loss of platelet aggregability: −; 30%>(1 mg/kg administration), ±; 30%-70%(1 mg/kg administration), +; 70%<(1 mg/kg administration), ++; 70%<(0.3 mg/kg administration)

Each anti-GPVI monoclonal antibody prepared in EXAMPLE 2 was intravenously administered to male cynomolgus monkey (about 6 kg) at intervals of 24 hours in the dose of 0.3 mg/kg and 1 mg/kg. Prior to administration, 24 hours and 48 hours after the first administration, blood was collected, and number of platelet, expression level of CD62P protein (marker for platelet activation), expression level of platelet membrane protein (GPIIb/IIIa; CD41a and GPVI; CD42a), expression level of platelet GPVI and platelet aggregability (the reactivity to collagen and ADP) were assayed.

Number of platelet of citrated blood was counted using Sysmex F-820. In the monkeys, to which clones F1232-18-3, F1232-10-2, F1232-14-2, F1232-43-3, F1201-20 and F1232-37-2 were administered respectively, no decrease in platelet number was observed (the decreasing rate was less than 20%; in Table 6, represented by symbol “−”). Further, in the monkeys, to which F1201-18 and F1232-17-1 were administered, downward trend of platelet number was observed (the decreasing rate was 20% to 40%; in Table 6, represented by symbol “±”). Meanwhile, in the monkeys, to which F1232-7-1, F1232-24-1 and F1232-21-1 were administered, apparent decreasing of platelet number was observed (the decreasing rate was 40% to 60%; in Table 6, represented by symbol “+”), especially the decreasing was remarkable in the case of F1232-19-1 (the decreasing rate was equal to or more than 60%; in Table 6, represented by symbol “++”).

Citrated blood was centrifuged at 100×g at 25° C. for 20 minutes to prepare platelet rich plasma (PRP). After PRP was diluted with PBS containing 0.5% immobilized FBS and 2.5 mM EDTA (hereinafter referred to as FACS buffer) to make platelet number of PRP 1×108 cells/mL, using anti-human CD62P-PE (BD Biosciences Pharmingen), expression of CD62P protein on the platelet from monkey was analyzed by FACS. That is, anti-human CD62P-PE was added to PRP and the mixture was stood at room temperature for 30 minutes. After that, the platelet was washed with FACS buffer, and fluorescent intensity of the platelet was measured by flow cytometer CYTOMICS FC500 (BECKMAN COLETER). As a result, in any PRP prepared from the monkeys, to which anti-GPVI monoclonal antibody was administered, no increasing of expression of CD62P was observed in comparison with the level before administration (induction of expression was less than two-fold; in Table 6, represented by symbol “−”).

A measurement of the expression level of CD41a and CD42a proteins on monkey platelet was performed by FACS analysis using anti-human CD41a-FITC (BD Biosciences Pharmingen) and anti-human CD42a-PE (BD Biosciences Pharmingen) respectively after preparation of PRP in a similar manner to the case of CD62P. As a result, in the PRP prepared from the monkey, to which F1232-21-1 and F1232-19-1 have been administered, a slight decrease in CD41a and CD42a proteins was observed (the depletion rate after administration of 1 mg/kg was 30% to 70%; in Table 6, represented by symbol “±”). However, in the PRP prepared from the monkey, to which other anti-GPVI monoclonal antibodies have been administered, no effect on the expression of CD41a and CD42a proteins was observed (the depletion rate after administration of 1 mg/kg was less than 30%; in Table 6, represented by symbol “−”).

Confirmation of GPVI protein on monkey platelet was performed by FACS analysis using anti-GPVI polyclonal antibody labeled with fluorescent dye Af488 after preparation of PRP in a similar manner to the case of CD62P. As a result, in the PRP prepared from the monkey, to which F1201-18, F1201-20 and F1232-37-2 had been administered, depletion of GPVI protein was observed (the depletion rate after administration of 1 mg/kg was equal to or more than 70%; in Table 6, represented by symbol “+”). Particularly, in F1232-7-1, F1232-18-3, F1232-10-2, F1232-21-1, F1232-43-3, F1232-17-1 and F1232-19-1, depletion was remarkable (the depletion rate after administration of 0.3 mg/kg was equal to or more than 70%; in Table 6, represented by symbol “++”). Meanwhile, in the PRP prepared from the monkey, to which F1232-24-1 and F1232-14-2 have been administered, partial disappearance was observed (the depletion rate after administration of 1 mg/kg was 30% to 70%; in Table 6, represented by symbol “±”).

The platelet aggregability on collagen and ADP was measured using platelet aggregation analyzer (PA-200 Aggregation Analyzer, Kowa). Firstly, after dilution of PRP with saline to make platelet number 3×108 cells/mL, CaCl2 solution at the final concentration of 1 mM was added, and the mixture was incubated at 37° C. for 3 minutes. Further, collagen solution at the final concentration of 2 μg/ml or ADP solution at the final concentration of 10 μM was added for incubation at 37° C. for 12 minutes. The platelet aggregation rate was calculated by measuring light transmission with PA-200 Aggregation Analyzer (Kowa). As a result, in the PRP prepared from the monkey, to which F1232-7-1 and F1201-20 have been administered, decrease in collagen-responsive platelet aggregability was observed (the rate of loss after administration of 1 mg/kg was equal to or more than 70%; in Table 6, represented by symbol “+”). Particularly, in F1232-18-3, F1232-10-2, F1232-21-1, F1201-18, F1232-43-3, F1232-37-2, F1232-17-1 and F1232-19-1, decrease was remarkable (the decreasing rate after administration of 0.3 mg/kg was equal to or more than 70%; in Table 6, represented by symbol “++”). Further, in the PRP prepared from the monkey, to which F1232-24-1 has been administered, partial decrease was observed (the decreasing rate after administration of 1 mg/kg was 30% to 70%; in Table 6, represented by symbol “±”). On the other hand, in F1232-14-2, little effect on aggregability was detected (the rate of loss after administration of 1 mg/kg was less than 30%; in Table 6, represented by symbol “−”).

In addition, ADP-induced platelet aggregation, in PRP prepared from the monkey, to which F1232-10-2 or F1232-21-1 have been administered, was partially decreased (the rate of loss after administration of 1 mg/kg was 30% to 70%; in Table 6, represented by symbol “±”). However, in PRP prepared from the monkey, to which other anti-GPVI monoclonal antibodies have been administered, no effect on the platelet aggregation induced by ADP was observed (the rate of loss after administration of 1 mg/kg was less than 30%; in Table 6, represented by symbol “−”).

Example of determining the amino acid sequence of variable region of the anti-GPVI antibody (clone F1232-7-1) was illustrated below. Regarding other anti-GPVI antibodies, using a similar experimental procedure, the amino acid sequence of variable region was determined.

That is, from hybridoma that produces the intended anti-GPVI monoclonal antibody, total RNA was extracted using Rneasy Micro Kit (QIAGEN), and single-stranded cDNA was synthesized with SuperScript III First-Strand Synthesis System for RT-PCR kit (Invitrogen). By PCR using the obtained single-stranded cDNA as a template and with Mouse Ig-Primer Set (Novagen), variable region was amplified and the base sequence was determined. The sequence information on database was retrieved to obtain the sequence around initiation codon, and 5′-primer was designed once again (F1232-7-1 heavy chain 5′-primer: 1232H-b (SEQ ID NO: 216); light chain 5′-primer: 1232K-a (SEQ ID NO: 218)). On the other hand, with respect to 3′-sequence of variable region, primers, to which restriction enzyme recognition site (Nhe I recognition sequence for the heavy chain and Bsi WI recognition sequence for the light chain) is attached, were designed without changing the amino acid sequence (F1232-7-1 heavy chain 3′-primer: 1031H-b (SEQ ID NO: 217); light chain 3′-primer: mIgK-BsiWI (SEQ ID NO: 219)). Herein, the restriction enzyme recognition site is able to ligate to human constant region. Using these primers, PCR was performed again to amplify heavy chain variable region and light chain variable region. Subsequently, using pT7BlueT vector (Novagen), the obtained amplified product was cloned. A plasmid harboring a gene fragment encoding heavy chain variable region and a plasmid harboring a gene fragment encoding light chain variable region were designated pTK-2464 and pTK-2466, respectively. Then, according to the conventional method, the base sequences of heavy chain variable region and light chain variable region were determined. The base sequences (SEQ ID NO: 87 for the heavy chain variable region and SEQ ID NO: 88 for the light chain variable region) and the amino acid sequences encoded by those (SEQ ID NO: 111 for the heavy chain variable region and SEQ ID NO: 112 for the light chain variable region) were shown in Table 7. Further, regarding clones other than F1232-7-1, the base sequences and the amino acid sequences were determined in a similar manner (Table 7). In addition, the sequence of CDR region in the amino acid sequence of variable region was shown in Tables 8 and 9.

TABLE 7 Nucleotide sequence Amino acid sequence Clone No. Heavy chain Light chain Heavy chain Light chain F1232-7-1 Seq. ID. 87 Seq. ID. 88 Seq. ID. 111 Seq. ID. 112 F1232-18-3 Seq. ID. 89 Seq. ID. 90 Seq. ID. 113 Seq. ID. 114 F1201-20 Seq. ID. 91 Seq. ID. 92 Seq. ID. 115 Seq. ID. 116 F1232-17-1 Seq. ID. 93 Seq. ID. 94 Seq. ID. 117 Seq. ID. 118 F1232-19-1 Seq. ID. 95 Seq. ID. 96 Seq. ID. 119 Seq. ID. 120 F1232-21-1 Seq. ID. 97 Seq. ID. 98 Seq. ID. 121 Seq. ID. 122 F1232-43-3 Seq. ID. 99 Seq. ID. 100 Seq. ID. 123 Seq. ID. 124 F1201-18 Seq. ID. 101 Seq. ID. 102 Seq. ID. 125 Seq. ID. 126 F1232-10-1 Seq. ID. 103 Seq. ID. 104 Seq. ID. 127 Seq. ID. 128 F1232-37-2 Seq. ID. 105 Seq. ID. 106 Seq. ID. 129 Seq. ID. 130 F1232-14-2 Seq. ID. 107 Seq. ID. 108 Seq. ID. 131 Seq. ID. 132 F1232-24-1 Seq. ID. 109 Seq. ID. 110 Seq. ID. 133 Seq. ID. 134

TABLE 8 Heavy chain Clone No. CDR1 CDR2 CDR3 F1232-7-1 DYAIH (15) VISIYYDDTNYNQKFKG (16) RRDSSGPYAMDY (17) F1232-18-3 SYAMS (18) TISSGGSYTYYPDSVKG (19) DYGYFDY (20) F1201-20 DYYVN (21) EIYPGSGNTYYNEKFKG (22) WRGNLYYDYDAETLFAY (23) F1232-17-1 SYGMS (24) TISSGGSYTYYPDNVKG (25) QVYYFGSCDY (26) F1232-19-1 SYGMS (27) TISSGGSYTYYPDSVKG (28) QVYYYGSSDY (29) F1232-21-1 YYLIE (30) VINPGSGVTNYNEKFKG (31) SIYYGTIDY (32) F1232-43-3 TYGIGVG (33) HIWWDDNKYYNTALKS (34) IHYYGSSLDY (35) F1201-18 SYAMS (36) IISSGGSYTYYPDSVKG (37) QGGGYFDY (38) F1232-10-1 TYGIGVG (39) HIWWNDDKYYNTALKS (40) VYYYGSSFDY (41) F1232-37-2 DYAIH (42) VISIYYDDTNYNQKFKG (43) RRDSSGPYAMDY (44) F1232-14-2 (45) (46) (47) F1232-24-1 DYAMH (48) VISTYYGDASYNQKFKG (49) SYDYDPWFAY (50) Number in parentheses following the sequence represents the sequence number.

TABLE 9 Light chain Clone No. CDR1 CDR2 CDR3 F1232-7-1 RASESVDSYGNSFMH (51) RASNLES (52) QQSNEDPYT (53) F1232-18-3 RASESVDNYGISFMH (54) RASNLES (55) QQSNTDPRT (56) F1201-20 RASQDISNYLN (57) ATSSLDS (58) LQYASYPYT (59) F1232-17-1 RASESVDNYGISFMH (60) RASNLES (61) QQSNKDPWT (62) F1232-19-1 RASESVDNYGISFMH (63) RASNLES (64) QQSNKDPWT (65) F1232-21-1 RASQDISNYLN (66) YTSRLHS (67) QQGSTLPYT (68) F1232-43-3 RASENIYSYLA (69) AATNLAD (70) QHFYGTPWT (71) F1201-18 RASESVDSYGNSFMH (72) RASNVES (73) QQSNEDPPT (74) F1232-10-1 RASENIFSILA (75) AATNLAD (76) QHFYGTPWT (77) F1232-37-2 RASESVDSYGNSFMH (78) RASNLES (79) QQSNEDPYT (80) F1232-14-2 (81) (82) (83) F1232-24-1 RASESVDNYGISFMH (84) RASNLES (85) QQSNKDPLT (86) Number in parentheses following the sequence represents the sequence number.

An antibody, wherein V region having an antigen binding activity is derived from hybridoma antibody, i.e. mouse antibody, and C region is derived from human, that is, a chimeric antibody was prepared.

Construction of expression plasmid for mouse-human chimeric anti-GPVI antibody (Clone No. F1232-7-1) was performed as follows.

First, by cleaving pTK-2464 with restriction enzymes EcoR I and Nhe I, a gene fragment C encoding heavy chain V region was prepared. On the other hand, by cleaving pTK-2232 (see, EXAMPLE 10 of WO 2005/7800) with restriction enzymes Eco 47111 and Bam HI, a gene fragment D encoding heavy chain constant region (Cγ4) was prepared. These fragments were ligated to the downstream of EF promoter in the expression vector pEF2cew preliminarily cleaved with Eco RI and Bam HI, to become fragment C+fagment D for construction of heavy chain-expressing plasmid (pTK-2468).

Next, by cleaving pTK-2466 with restriction enzymes EcoR I and BsiW I, DNA fragment E encoding the light chain variable region was prepared. On the other hand, using HeLa genomic DNA as a template and primers BsiWI-hIgK (SEQ ID NO: 220) and IgK-e (SEQ ID NO: 221), PCR was performed and human light chain constant region (Cκ) was cloned into pT-7BlueT (pT7-hIgK). To excise a human light chain constant region from pT7-hIgK, pT7-hIgK was cleaved with restriction enzymes Bsi WI and Bam HI to prepare DNA fragment F. These fragments were ligated to the downstream of EF promoter in the expression vector pEF2cew preliminarily cleaved with Eco RI and Bam HI, to become fragment E+fagment F for construction of light chain-expressing plasmid (pTK-2474).

Expression and purification of mouse-human chimeric antibody that is constructed were performed in a similar method as shown in EXAMPLE 1. That is, expression plasmids for heavy chain and light chain of the mouse-human chimeric antibody of intended clones were co-transfected to the COS-1 cell, and cultured at 37° C. for three days. After cultivation, the culture supernatant was purified by Protein A column. Purity of the obtained mouse-human chimeric antibody was confirmed by SDS-PAGE. Further, a binding ability to human GPVI and monkey GPVI was confirmed. First, GPVI substitution mutant, hGPVI-hFc or GPVI-FFH-hFc that were prepared in EXAMPLE 1 were diluted to 2 μg/mL with D-PBS and 50 μL/well of the solution was added to immunoplate (Maxisorp NUNC). Next, after incubation at 37° C. for one hour, the well was washed five times with ion-exchanged water, and 100 μL of D-PBS containing 2% StabilGuard (Surmodics) was added to each well for blocking. Then, the purified mouse-human chimeric antibody was added to each well and incubated at 37° C. for one hour. After incubation, the well was washed three times with saline containing 0.05% Tween20. Peroxidase-labeled anti-human light chain K antibody (DAKO) was 1000-fold diluted with D-PBS containing 10% rabbit serum and 50 μL was added to each well. After incubation at 37° C. for one hour, five times washing was performed in a similar manner. Subsequently, TMB solution (BioFix) was added to each well. After incubation at room temperature for 10 minutes, the reaction was terminated by adding 0.5 M sulfuric acid solution. With the plate spectrophotometer (Multi-Scan JX, Dainippon Pharmaceutical), absorbance at 450 nm was measured. As a result, it was confirmed that the prepared mouse-human chimeric antibody specifically binds to human and monkey GPVI.

To maintain high affinity in humanized antibody, framework residue is selected according to general method by Queen et al. (Proc. Natl. Acad. Sci. USA 86: 10029, 1989). As to a sequence from human, based on the database for K light chain and heavy chain sequences by Kabat et al. (Sequences of proteins of immunological interest, 5th ed., U.S. Department of Health and Human Services, 1991), a sequence having a high framework homology in mouse anti-GPVI monoclonal antibody (clone No. F1232-7-1) is selected. Moreover, by computer analysis, amino acids in the most suitable framework are altered. Specifically, using computer program ENCAD (Levitt, J. Mol. Boil. 168: 595 (1983)), and protein modeling tools such as Hommology (accelris) and FAMS (SGI), a molecular model for F1232-7-1 antibody variable region is constructed. CDR sequences of F1232-7-1 antibody may be transplanted into FR of the human Eu antibody molecular model obtained from the antibody database (Stephens et al., Immunology 85 (4), 668-674 (1995)). Through optimization or simulation such as molecular optimized calculation and molecular dynamical calculation, in FR region, in which a significant contact between CDR and FR is shown on the computer model, being different from the intrinsic human antibody model, substitution of amino acid derived from mouse antibody may be performed in the position where the contact between CDR and FR is expected to be improved by an amino acid substitution. Also, the amino acid residues in FR that infrequently appear at the position in the human antibody database are replaced with the amino acids consensus to human at those positions. Since the quality for the amino acid substitution must be confirmed by actual activity, several kinds of antibodies with the different type of the amino acid substitution may be prepared.

Based on the sequence selected in EXAMPLE 11A (1), a gene encoding the amino acid sequence comprising a signal peptide, a splice donor signal and a restriction site (e.g. Eco RI) is constructed. Several synthetic nucleotides (about 80 bases) based on the constructed gene were prepared to make them overlap. That is, the oligos are paired and annealed. Then they are extended by the Klenow fragment of DNA polymerase to obtain double-stranded fragments. After denaturing this fragment to become a single strand and annealing them, they are extended by the Klenow fragment of DNA polymerase to obtain double-stranded fragments encoding entire gene. The obtained fragment is amplified by PCR and purified. After cleavage with restriction enzymes such as Eco RI and Nhe I, the fragment is re-purified. The purified fragment is ligated to a gene fragment comprising CH1 exon to CH3 exon of human IgG4 constant region gene (Cγ4), wherein, for example, pTK-2232 is cleaved with Nhe I and Bam HI, and inserted into the downstream of EF promoter in expression vector pEF2cew, which is preliminarily cleaved with Eco RI and Bam HI to construct an expression plasmid for humanized heavy chain. Also, when the amino acid residues to be substituted are a little, it is possible to introduce the mutated amino acids into expression plasmid by site-directed mutagenesis. The sequence for light chain variable region can be constructed as described above. In this case, human Cκ region is excised from pT7-hIgK, ligated to the light-chain variable region sequence and incorporated downstream EF promoter of the expression vector pEF2cew.

To prepare a transformant which produces an antibody, a heavy chain plasmid and a light chain plasmid are linearlized with restriction enzymes and is introduced into mouse myeloma cell, Sp2-O-ag14 (ATCC CRL1581) using Genepulsar (BIORAD). For example, 20 μg of the linearlized DNA fragment is electroporated into 1×107 cells at capacitance of 360 V, 25 μD. Next, the cells are inoculated to 96-well plate and cultured for two days. After cultivation, to select the cell, in which the fragment of plasmid is incorporated, D-MEM (Sigma) containing 10% FCS, 1×HT (Invitrogen), 0.25 mg/ml Xanthine, 1 μg/ml Mycophenolic acid is added and is further cultivated for two weeks. After cultivation, by analyzing an antibody in the supernatant, the intended humanized F1232-7-1 antibody-producting cell line is selected. That is, the antibody in the supernatant is bound to the immobilized GPVI antigen and the bound antibody can be detected with peroxidase-labeled anti-human IgG4 antibody. The antibody producting cell line, wherein the binding has been detected, is cultured to become confluent in the medium containing 10% FCS, and the medium is replaced with serum-free medium (Hybridoma SFM, Invitrogen). Culture supernatant is recovered, and the antibody that is contained in the culture supernatant is bound to Protein A (Prosep-A, Millipore) to elute with 0.1 M Glycine-HCl (pH3.0). The purified antibody is dialyzed against PBS-(Sigma). Antibody concentration is calculated from absorbance at 280 nm (1 mg/mL of human antibody shows an absorbance of about 1.3).

The binding ability of humanized antibody to GPVI antigen of the parent mouse antibody can be measured with BIACORE system (BIACORE) and can be compared. That is, according to the manual of BIACORE3000, the purified hGPVI-mFc is immobilized to CM5 chip (BIACORE). Then a series of diluents of the antibody diluted with HBS-EP buffer (BIACORE) is prepared, and each sample may be injected. The obtained data are analyzed using program for analysis available from BIACORE (BIA Evaluation, BIACORE), and an affinity value (Kd) can be calculated.

In a humanized antibody, for maintaining CDR sequence to be grafted as an appropriate domain structure having an activity, a method of grafting the CDR sequence together with original FR region is also available. Whether any amino acid residue is involved in maintenance of the CDR domain structure is estimated from properties (hydrophobicity, hydrophilicity, acidity, basic, molecular size, etc.) of the amino acid in FR, and is presumable by modeling using computer. That is, modeling may be performed with the software QUANTA/CHARMm run on Silicon Graphics, or Modeler (Molecular Simulations). From the sequence of human antibody registered on Brookhaven Protein Data Bank (PDB), 3D structure of the antibody having a high homology to VH and VL regions of F1232-7 antibody is searched, and based on the result, the 3D structure of F1232-7 antibody is estimated. The amino acids in FR region, which binds via hydrogen bond to CDR of the heavy chain and the light chain on the deduced 3D structure (the first group), are selected, and further the amino acids in FR region, which binds via hydrogen bond to the first group, is selected (the second group). In a similar manner, the amino acids in FR region, which is estimated to bind to CDR by energy bond such as electrostatic interaction or van der Waal's force (the first group) and the amino acids in FR region, which further binds thereto (the second group) are selected. The amino acids in FR region thus selected, is grafted onto the sequence of human antibody together with the amino acid of CDR. However, when a sequence obtained by classification by Kabat et al. (Sequence of proteins of immunological interest, 5th ed., U.S. Department of Health and Human Services, 1991) or NCBI (National Center for Biotechnology Information), wherein the sequence does not exist in the amino acid sequence of variable region of human antibody, occurs, such sequence may not be grafted. Based on the thus obtained information, the sequence to be grafted to VH and VL of human antibody sequence can be determined, and the gene to be used for preparation of humanized antibody can be constructed.

The constructed gene may be prepared by a method in combination with the kit available from Amersham (Oligonucleotide-directed in vitro mutagenesis system version 2) and PCR method, a method of amplifying in combination with several kinds of nucleotides for long chain synthesis, or a method of amplifying using a gene encoding VH or VL of the chimeric antibody as a template and several primers followed by obtaining an intact gene fragment with their amplified gene fragment as a template. By ligating the obtained amplified gene fragment to the fragment of constant region in the plasmid (pTK-2232 or pT7-IgK) described in EXAMPLE 11A (2) and incorporating it into the downstream of EF promoter in the expression vector pEF2cew, an expression vector for humanized antibody can be constructed. By introducing the prepared plasmid into a cell by the method of EXAMPLE 11A (2) and obtaining a transformant, a purified antibody can equally be prepared. Further, as well as EXAMPLE 11A (3), evaluation of the antibody may be performed.

The recognition domain of anti-GPVI antibody in the blood of a patient with GPVI deficiency (Tateo Sugiyama and five other members, Blood (USA), 1987, volume 69, No. 6, pp. 1712-1720) was analyzed using a variety of recombinant proteins. A patient-specific purified antibody and a recombinant human GPVI loop substitution mutant described in EXAMPLE 7 were admixed. After reacting on polypropylene plate at 37° C. for two hours, 50 μL/well of sample mixture was added to recombinant hGPVI-hFc solid phase plate, which was prepared as with the method in EXAMPLE 2, and the plate was incubated at 37° C. for one hour. Then peroxidase-labeled anti-human κ and λ-chain antibody (DAKO, P130, P129) was 1000 fold diluted with D-PBS (pH7.4) containing 10% rabbit serums and 50 μL of the diluent was added to each well. After incubation at 37° C. for one hour, an absorbance was measured as with the method in EXAMPLE 2. A measurement value was compared with an absorbance in the case where the recombinant human GPVI loop substitution mutant was not added. From the result with loop substitution mutants without absorbance reduction or with a little extent of absorbance reducing, the antigen recognition site of anti-GPVI antibody included in GPVI-deficient patient's autoantibody was deduced.

The result was shown in Table 10. In the case where an antibody (#930819) of patient at early stage after developing of ITP was used as a subject, the absorbance obtained from hGPVI-hFc-mL4 (a protein wherein loop 4 of recombinant human GPVI-hFc is replaced with mouse loop 4; hereinafter, described as the same), hGPVI-hFc-mL5, hGPVI-hFc-mL9 and hGPVI-hFc-mL13 was not reduced. That is, it was found that the anti-GPVI antibody derived from GPVI deficient patient as an experiment subject comprises an antibody that recognizes loop 4, 5, 9 or 13. Meanwhile, in the case where an antibody (#021004) of patient with a long period after developing of ITP was used as a subject, the absorbance for hGPVI-hFc-mL9 and hGPVI-hFc-mL13 was not reduced. That is, it was found that the anti-GPVI antibody derived from GPVI deficient patient as an experiment subject comprises an antibody that recognizes loop 9 or 13.

TABLE 10 Absorption assay of the GPVI-deficient patient's antibody by mouse loop substitution mutant Patient's antibody mL1 mL2 mL3 mL4 mL5 mL6 mL7 mL9 mL10 mL11 mL12 mL13 mL14 # 930819 − − − + + − − + − − − + − # 021004 − − − − − − − + − − − + − +: shows the case where no or little decrease in absorbance was observed when it was added as an absorption antigen. −: shows the case where decrease in absorbance to the same extent as the unsubstituted protein (hGPVI-hFc) was observed when it was added as an absorption antigen. 12-2 Competitive Test of the Antibody of GPVI-Deficient Patient with the Mouse Anti-Human GPVI Monoclonal Antibody

A competitive test of mouse anti-GPVI hybridoma antibody prepared in EXAMPLE 2 with anti-GPVI antibody included in the blood of GPVI-deficient patient was performed. That is, 50 μL/well of anti-GPVI antibody derived from a patient with GPVI deficiency was added to hGPVI-hFc-coated plate, which was described in 12-1, and the plate was incubated at 4° C. for overnight. Thereto peroxidase-labeled mouse anti-GPVI hybridoma antibody was added to make the absorbance 0.5 through 1.0. The mixture was reacted at 37° C. for 45 minutes, and the absorbance was measured as with the method in EXAMPLE 2. As a result, the mouse anti-human GPVI monoclonal antibodies F1199-6 and F1232-37-2, which were prepared by the same method as that of EXAMPLE 2, competed with the anti-GPVI antibody included in the blood of GPVI-deficient patient. The result is shown in FIG. 2.

By mixing 20 μg of purified hGPVI-mFc fusion protein, which was prepared in EXAMPLE 1, Alum (PIERCE), and oligo CpG, an antigen to be administered was prepared. To ddY mouse (female, 8-week old, SLC) the antigen was administered and further additional 20 μg of antigen was administered. After three days, a lymphocyte was isolated from lymph node or spleen. Then a cell fusion was done using the same method as that of EXAMPLE 2 to select a hybridoma.

After one week, hybridoma, which produces a desired antibody was screened using the binding property to hGPVI-hFc and hGPVI-mL9-hFc prepared in EXAMPLE 7 as an index. That is, the purified hGPVI-hFc or hGPVI-mL9-hFc, which was prepared in EXAMPLE 1, was diluted with D-PBS (pH7.4) to 1 μg/mL. And 50 μL/well of the solution was added to make hGPVI-hFc or hGPVI-mL9-hFc-coated plate using the same method as that of EXAMPLE 2. Next, a culture supernatant was added to each well, and after incubation at room temperature for one hour, by the same method as that of EXAMPLE 2, using peroxidase-labeled anti-mouse immunoglobulin antibody (DAKO, P260) a reaction was performed to measure an absorbance. As a result, the cell producing an antibody that binds to the purified hGPVI-hFc (absorbance: equal to or more than one), but not hGPVI-mL9-hFc (abdorbance: equal to or less than 0.5) was selected and cloned by the method described in EXAMPLE 2. Eight through ten days later, a screening was done as with the similar way to obtain a hybridoma producing an L9-specific mouse anti-human GPVI antibody. As with EXAMPLE 2B (3), the obtained hybridoma was cultured and a monoclonal antibody was purified. The sub-type of each antibody was determined using IsoStrip Mouse Monoclonal antibody Isotyping Kit (Roche).

The purified hGPVI-mFc fusion protein or the rat GPVI-mFc fusion protein, wherein L9 or L11 loop is replaced with the loop derived from human (ratGPVI-hL9/h11-mFc) (SEQ ID NO: 288) was used as an antigen to be administered. In addition, after synthesizing cDNA by reverse transcription using a rat bone marrow RNA as a template and oligo-dT primer, whole length gene of rat GPVI was cloned by PCR using the abovementioned cDNA as a template. Primer pair used for PCR is mGPVI-a: CCACATAGCTCAGGACTGGG (SEQ ID NO: 289) and mGPVI-d: CCAAGTTATTTCTAGGCCAGTGG (SEQ ID NO: 290). Equal amount of antigen to be administered, 20 μg, and Freund's complete adjuvant (DIFCO) was mixed and administered to Wistar rat (female, 8-week old, SLC). After two weeks, a lymphocyte was isolated from lymph node. After mixing with SP2/O-Ag14 (ATTC), a cell fusion was performed using the same method as that of EXAMPLE 2 to select hybridoma.

After one week, hybridoma, which produces a desired antibody was screened by the method described in the above 14-1. As a result, the cell producing an antibody that binds to the purified human GPVI-hFc (absorbance: equal to or more than one), but not the purified human GPVI-hFc, wherein L9 is replaced with mouse L9 (abdorbance: equal to or less than 0.5) was selected and cloned by the method described in EXAMPLE 2 to obtain a hybridoma producing rat anti-human GPVI antibody. As with EXAMPLE 2B (3), the obtained hybridoma was cultured and a monoclonal antibody was purified. The sub-type of each antibody was determined using Rat MonoAB ID/SP Kit (ZYMED).

To analyze a property for each antibody obtained in EXAMPLE 13, binding to human GPVI-hFc, competition with F1232-37-2 described in EXAMPLE 4 and specificity for L9 loop were investigated. That is, as for the binding to human GPVI, according to the method of EXAMPLE 2, a binding activity to immobilized antigen was measured. For the binding activity, the range of absorbance from 0.5 to 1.0 is represented by “+”, from 1.0 to 2.0 by “++”, the range more than 2.0 by “+++”. The competitive test with F1232-37-2 was performed as follows. That is, according to the method by Nakane et al. (J. Histochem. Cytochem., 22, 1084, 1974), peroxidase (TOYOBO)-labeled F1232-37-2 antibody was prepared and an antibody concentration was calculated from the amount of the antibody used.

Next, by using the peroxidase-labeled antibody, the competitive test for a variety of purified antibody was performed. That is, 25 μL of the above labeled antibody and 25 μL of each purified antibody were added to well of hGP-hFc-immobilized plate. After incubation at 37° C. for one hour, the well was washed five times with the saline containing 0.05% Tween20, and the reaction mixture was developed with TMB solution (BioFix). After incubation at room temperature for 10 minutes, the reaction was terminated with 0.5 M sulfuric acid solution and an absorbance at 450 nm was measured with the plate spectrophotometer (Multi-Scan JX, Dainippon Pharmaceutical). The result is shown in Table 11. For inhibiting activity, the case where more than 50% inhibition to the absorbance without inhibition antibody is represented by “+++”, the case of 30% to 50% inhibition by “++” and the case of 10% to 30% inhibition by “+”.

Specificity for L9 loop was measured according to the method described in EXAMPLE 13, 13-1. As a result, it is concluded that the case where the reactivity decline of more than 50% is exhibited is designated “+”, that is, L9 loop is recognized.

TABLE 11 Anti-human GPVI monoclonal antibodies prepared as the experimental example. 14 Anti-GPVI Inhibition of binding monoclonal Binding to Collagen and Competition with F1232-37-2 for antibodies Species hGPVI-Fc hGPVI-Fc hGPVI-hFc KD F1249-18-2 Mouse +++ +++ ++ 1.12E−10 F1245-7-1 Mouse ++ − + 3.19E−10 F1246-1-1 Mouse ++ − + 9.64E−08 F1249-5-1 Mouse ++ + ++ 3.11E−10 F1249-20-1 Mouse +++ − ++ 2.23E−10 F1249-24-1 Mouse +++ + +++ 8.95E−10 F1249-30-1 Mouse +++ − +++ 3.78E−08 F1245-5-1 Mouse ++ + ++ 1.56E−07 F1245-6-2 Mouse ++ + +++ 1.33E−08 F1249-3-2 Mouse + + ++ 1.73E−07 F1245-4-1 Mouse +++ + ++ 4.65E−10 F1249-22-1 Mouse + ++ + 2.25E−07 F1251-1-1 Mouse + + + 8.61E−07 F1257-3-1 rat +++ − − 3.25E−10

Inhibiting activity for collagen-GPVI binding of each antibody obtained in EXAMPLE 13 was studied by the same method as that of EXAMPLE 2. The case where more than 50% inhibition to the absorbance without inhibition antibody is represented by “+++”, the case of 30% to 50% inhibition by “++” and the case of 10% to 30% inhibition by “+” (Table 11).

Dissociation constant of each antibody obtained in EXAMPLE 14 was determined by the same method as that of EXAMPLE 5. The result was shown in Table 11 as the relative value to the dissociation-constant of F1232-37-2.

An amino acid sequence of anti-human GPVI antibody variable region prepared in EXAMPLE 13 was determined by the same method as that of EXAMPLE 9. For the clone, wherein the base sequence for antibody variable region was determined, base sequence in CDR and variable regions and deduced amino acid sequence were shown in Tables 12 and 13 and the sequence listing (see Table 14). From the result of analyzing the gene sequence of antibody variable region, it is assumed that the sequences of these antibodies are derived from several kinds of certain genes for antibody. Further, it is recognized that repertoire selection of the antibody that recognizes loop 9 of human GPVI is characterized.

TABLE 12 CDR sequence for anti-GPVI antibody heavy chain Heavy chain Clone No. CDR1 CDR2 CDR3 F1245-4-1 SYWMH MIHPNSDNTNYNEKFKS HYYDYVDY F1245-5-1 SYWMH MIHPNSGSTHYNEKFKS GGVTPVAY F1245-6-2 SYWMH MIHPNSGSTNYNEKFKS GGVTPVAY F1245-7-1 SYWMH MIHPNSGSTNYNEKFKS PVTAVVEYYFDY F1246-1-1 SYGMS TISNGGTYTYYPDSVKG LRDYYAMDY F1249-3-2 SYGMS TISSGGSYTYYSDSVKG DSGYFDY F1249-5-1 SYWMH MIHPNSDITNYNEKFKN LGDYYAMDY F1249-18-2 SYWMH MIHPNSDITNYNEKFKN SGDYYAMDY F1249-20-1 DYAMH VISTYYGDTSYNQKFKG AEDYDPWFAY F1249-22-1 SYWMQ EIDPSDSYTNYNQKFKG GAITTATLDY F1249-24-1 DYAMH VISTYYGDTSYNQKFKG AEDYDPWFAY F1249-30-1 DYAMH VISTYYGDTSYNQKFKG AEDYDPWFAY F1251-1-1 DYYIH YINPNSGYTNYNEKFKS CNSGYGDWFAY F1257-3-1 TSGMVVS AIDWDGDKYYNPSLKS TPYYGYKEAYYFD Y

TABLE 13 Amino acid sequence of the CDR for anti-GPVI antibody light chain Light chain Clone No. CDR1 CDR2 CDR3 F1245-4-1 KASQSVSNDVA YASNRYT QQDYSSPWT F1245-5-1 KASQSVSNDVA YASNRYT QQDYSSPWT F1245-6-2 KASQSVSNDVA YASNRYT QQDYSSPWT F1245-7-1 KASQSVSNDVA YASNRYT QQDYSSLT F1246-1-1 RASESVDSYGNSFMH RASNLES QQSNEDPWT F1249-3-2 RASESVDSYGNSFMH LASNLES QQSNNEDPRT F1249-5-1 KASQSVSNDVT YASNRYT QQDYSSPPT F1249-18-2 KASQSVSNDVA YASNRYT QQDYSSPPT F1249-20-1 RASESVDSYGNSFMY RASNLES QQSDEDPLT F1249-22-1 KSSQSLLNSNNQKNYLA FASTRES QQHYITPLT F1249-24-1 RASESVDSYGNSFMH RASNLES QQSDEDPLT F1249-30-1 RASESVDSYGNSFMH RASNLES QQSDEDPLT F1251-1-1 KASQNINKNLD YTNNLQT YQYNSGPGT F1257-3-1 KASRAIDDYLS DATSLAD LQSYSTPWT

TABLE 14 Nucleotide sequence and amino acid sequence of the variable region of anti-GPVI monoclonal antibody Nucleotide sequence Amino acid sequence Heavy Heavy Clone No. chain Light chain chain Light chain F1245-4-1 Seq. ID 224 Seq. ID 226 Seq. ID 225 Seq. ID 227 F1245-5-1 Seq. ID 228 Seq. ID 230 Seq. ID 229 Seq. ID 231 F1245-6-2 Seq. ID 232 Seq. ID 234 Seq. ID 233 Seq. ID 235 F1245-7-1 Seq. ID 236 Seq. ID 238 Seq. ID 237 Seq. ID 239 F1246-1-1 Seq. ID 240 Seq. ID 242 Seq, ID 241 Seq, ID 243 F1249-3-2 Seq. ID 244 Seq. ID 246 Seq. ID 245 Seq. ID 247 F1249-5-1 Seq. ID 248 Seq. ID 250 Seq. ID 249 Seq. ID 251 F1249-18-2 Seq. ID 252 Seq. ID 254 Seq. ID 253 Seq. ID 255 F1249-20-1 Seq. ID 256 Seq. ID 258 Seq. ID 257 Seq. ID 259 F1249-22-1 Seq. ID 260 Seq. ID 262 Seq. ID 261 Seq. ID 263 F1249-24-1 Seq. ID 264 Seq. ID 266 Seq. ID 265 Seq. ID 267 F1249-30-1 Seq. ID 268 Seq. ID 270 Seq. ID 269 Seq. ID 271 F1251-1-1 Seq. ID 272 Seq. ID 274 Seq. ID 273 Seq. ID 275 F1257-3-1 Seq. ID 276 Seq. ID 278 Seq. ID 277 Seq. ID 279 NEW-HAN Seq. ID 280 — Seq. ID 281 — Eu-HC Seq. ID 282 — Seq. ID 283 — REI-KA — Seq. ID 284 — Seq. ID 285 Eu-KA — Seq. ID 286 — Seq. ID 287

The expression plasmid prepared by the same method as that of EXAMPLE 10 was introduced into COS-1 cells with the method below to develop a transient expression of chimeric antibody. In addition, when expressing cF1232-18-3 (Mouse/human chimeric antibody of mouse monoclonal antibody F1232-18-3 is referred to as cF1232-18-3. Hereinafter, other mouse monoclonal antibodies are referred to in a similar manner), co-transfection of the heavy chain expression plasmid pTK-2471 and the light chain expression plasmid pTK-2475 was done. For cF1232-43-3, cF1232-10-1 or cF1232-37-2, pTK-2504 and pTK-2514, pTK-2509 and pTK-2517, or pTK-2510 and pTK-2511 were co-transfected, respectively.

COS-1 cells were inoculated to CellSTACK 10 Chamber (CORNING) at 2.1×106 cells/bench and cultivated at 37° C. for four days. After discarding culture fluid, cells were washed twice with D-MEM. Then the following mixture with FuGENE6/DNA/production medium was added with about 1.3 L/chamber. After mixing 2.12 ml of FuGENE6 (Roche Diagnostics), 530 μg each of heavy chain expression plasmid and light chain expression plasmids according to the protocol attached, the mixture was added to 1.3 L of Hybridoma-SFM (Invitrogen). After addition of mixture with FuGENE/DNA/production medium, cultivation was performed under the conditions of 37° C. for 3-4 days to collect the supernatant. Hybridoma-SFM medium, 1.3 L was newly added to CellSTACK 10 Chamber, and further cultivation was done for 3-4 days to collect the supernatant again. The mouse/human chimeric antibody expressed in the supernatant was purified as described below.

Purification procedure was done at 4° C. if not otherwise specified.

The cF1232-37-2/COS culture fluid prepared in 16-1 was clarified at room temperature using a capsule cartridge filter (Toyo Roshi Kaisha, Ltd.) having a one micrometer pore size as a prefilter and Fluorodyne filter having a 0.22 μm pore size and Fluorodyne filter (PALL) as a real filter to obtain a culture supernatant. The culture supernatant was loaded onto rpm Protein A Sepharose Fast Flow (Amersham Biosciences) previously equilibrated with PBS- (Sigma). After washing non-absorbed proteins with PBS-, proteins non-specifically bound was eluted with 100 mM phosphate buffer containing 1.5 M NaCl. Thereafter the antibody specifically bound was eluted with 100 mM glycine-HCl buffer (pH 3.0). The volume of eluate was measured and by adding a 1/10 volume of 1 M Tris-HCl (pH 7.0) pH was immediately neutralized. The preparation obtained was dialyzed against 0.9% NaCl aqueous solution to obtain a purified preparation. Using a similar manner, cF1232-43-3, cF1232-10-1 or cF1232-18-3 was purified.

In addition, among monoclonal antibodies cloned in EXAMPLE 13, as for F1249-18-2, F1245-7-1, F1246-1-1, F1249-24-1, F1245-4-1, F1249-22-1 and F1251-1-1, using the similar methods, preparation of chimeric antibody expression plasmid, expression of the antibody in COS-1 cells and purification of the antibody were performed.

Dissociation constant of each chimeric antibody prepared in 16-2 was measured using the protein interaction analyzer BIACORE 3000 (BIACORE) in a similar way to EXAMPLE 5. It was exhibited that the chimeric antibodies prepared have a well affinity to hGPVI-hFc.

To compare an antigen binding property of mouse/human chimeric antibody with that of the corresponding mouse monoclonal antibody, in the same method as that of EXAMPLE 14, peroxidase-labeled F1232-37-2, F1232-18-3, F1232-43-3 and F1232-10-1 were prepared. By the competitive method (the method described in EXAMPLE 14, 14-1), wherein to the binding reaction system with these labeled antibody and immobilized hGPVI-hFc, the corresponding non-labeled mouse/human chimeric antibody is added, a binding activity of mouse monoclonal antibody to human GPVI was compared with that of the corresponding mouse/human chimeric antibody. As a result, as shown in FIG. 4, there was no difference in both. Further, an inhibiting activity of each antibody for binding of GPVI to collagen was measured according to the method described in EXAMPLE 2 or EXAMPLE 14. From a result of this experiment, as shown in FIG. 5, the inhibiting activity for binding of GPVI to collagen by each antibody to be tested was confirmed. The inhibiting activity for binding of GPVI to collagen by the chimeric antibody was equivalent to that by mouse hybridoma antibody.

To the plate, wherein hGPVI-hFc has been immobilized at 4 μg/mL, a mixture of peroxidase-labeled F1232-37-2 or cF1232-37-2 by the method described in EXAMPLE 14, and hGPVI-hFc, mGPVI-hFc, hGPVI-mL3-hFc, wherein these were prepared in EXAMPLE 7, or hGPVI-K59E-hFc (a fusion protein consisting of one amino acid mutated human GPVI extracellular region, wherein lysine at 59 of human GPVI is substituted for glutamic acid, and human Fc), which has been prepared by the same method as that of EXAMPLE 1 and diluted with 0.1% BSA/PBS was added, and the plate was incubated at 37° C. for one hour. After incubation, the mixture was developed using TMB solution, and absorbance at 450 nm was measured.

The experiment revealed that the binding of peroxidase-labeled F1232-37-2 or cF1232-37-2 to hGPVI-hFc is inhibited by hGPVI-hFc, hGPVI-mL3-hFc and hGPVI-K59E-hFc, but not by mGPVI-hFc. For this result, F1232-37-2 was equivalent to cF1232-37-2. On the other hand, mouse monoclonal antibody F1199-6 prepared by the same method as that of EXAMPLE 2 indicated a different event. That is, the binding of peroxidase-labeled F1199-6 antibody to hGPVI-hFc was inhibited by hGPVI-hFc and hGPVI-mL3-hFc, but not by hGPVI-mFc and GPVI-K59E-hFc. The result was shown in FIG. 6.

To confirm an antigen recognition site for cF1232-37-2 antibody, a binding activity of GPVI-HHH-hFc, GPVI-MMM-hFc, hGPVI-mL2-hFc, hGPVI-mL3-hFc, hGPVI-mL4-hFc, hGPVI-mL5-hFc, hGPVI-mL6-hFc, hGPVI-mL7-hFc, hGPVI-mL8-hFc, hGPVI-mL9-hFc, hGPVI-mL10-hFc, hGPVI-mL11-hFc, hGPVI-mL13-hFc, and hGPVI-mL14-hFc, wherein these were prepared in EXAMPLE 7, to cF1232-37-2 antibody was measured.

That is, when by replacing a human GPVI loop region with a corresponding mouse loop region, the binding activity of the antibody reduces relative to the binding activity to human GPVI prior to replacement, it is speculated that the loop region replaced has an antibody's recognition region. Further, when in the substitution mutant in which a mouse GPVI loop region is replaced with a corresponding human loop region, the binding activity of the antibody is restored, it can be assumed that the loop region replaced is an antibody's recognition region. Accordingly, from both experiments, it is possible to identify the reaction region of the antibody.

The measurement method was according to the method of 17-3.

The results analyzed revealed, as shown in FIG. 7, in the case of cF1232-37-2 the activity significantly reduced when using hGPVI-mL9-hFc. From this result, it is speculated that cF1232-37-2 recognizes loop 9 region.

Citrate-added blood collected from normal human or cynomolgus monkey was provided to Sysmex F-820 to count platelets. Then, by centrifugation at 170×g at 25° C. for 15 minutes in the case of human, or at 115×g at 25° C. for 20 minutes in the case of monkey, platelet rich plasma (PRP) was obtained; then by 1300×g at 25° C. for 15 minutes, platelet poor plasma (PPP) was obtained.

Subsequently, the obtained PRP was diluted with PPP to make a concentration of platelet 3.33×108 cells/mL. The antibody, which final concentration was 1-10 μg/mL, was added thereto, and the mixture was incubated at 37° C. for 30 minutes. After incubation, para-formaldehyde (final concentration: 1%) was added for fixation at 4° C. for one hour. After washing the platelet with PBS containing 0.5% immobilized FBS (hereinafter referred to as FBS buffer), anti-human CD62P-PE (BECKMAN COULTER) was added thereto and stood at room temperature under the condition of light shielding for 30 minutes. After 30 minutes, the platelet was washed with FBS buffer. Then by measuring a fluorescent intensity of platelet with Flow cytometer CYTMICS FC500 (BECKMAN COULTER), CD62P expression was analyzed.

As shown in FIG. 8, F1199-6 antibody activated human platelet or cynomolgusmonkey platelet depending on the concentration, meanwhile platelet activating action could hardly be recognized in F1232-37-2 antibody. Further, the action of cF1232-37-2 on human or cynomolgus platelet was the same as the above. In addition, when as for the chimeric antibodies prepared in EXAMPLE 16, that is, cF1249-18-2, cF1245-7-1, cF1246-1-1, cF1249-24-1, cF1245-4-1, cF1249-22-1 and cF1251-1-1, their activating action on monkey platelet was studied, any activating action was not observed. In some clones, there were mouse non-chimeric antibodies which exhibited the activating action on monkey platelet. However, by chimerization the action disappeared. In addition, as for mouse monoclonal antibodies prepared in EXAMPLE 2, the same experiment was performed. As a result, no antibody possessed a human platelet activating function.

The PRP prepared by the method described in 19-1 was diluted with PPP to make a concentration of platelet 3.33×108 cells/mL. The antibody, which final concentration was 1-10 μg/mL, was added thereto, and the mixture was incubated at 37° C. for 5 minutes. Hereto CaCl2 solution was added at the final concentration of 1 mM, and during incubation at 37° C. for 12 minutes, a light transmission of PRP was measured with MCM Hema Tracer 313M (MCMEDICAL) to investigate a platelet aggregation.

As shown in FIG. 9, F1199-6 induced human platelet aggregation depending on the concentration; meanwhile, the platelet aggregation action by F1232-37-2 and cF1232-37-2 could hardly be detected.

The PRP prepared by the method described in 18-1 was diluted with PPP to make a concentration of platelet 3.33×108 cells/mL. The antibody, which final concentration was 1-10 μg/mL, was added thereto, and the mixture was incubated at 37° C. for 5 minutes. Hereto CaCl2 solution was added at the final concentration of 1 mM, and after additional incubation at 37° C. for three minutes, collagen solution was added at the final concentration of 1 μg/mL. During incubation at 37° C. for 12 minutes, a light transmission of PRP was measured with MCM Hema Tracer 313M (MCMEDICAL) to investigate a platelet aggregation induced by collagen.

As shown in FIG. 10, F1232-37-2 and cF1232-37-2 inhibited collagen-induced human platelet aggregation depending on the concentration.

The PRP prepared by the method described in 18-1 was diluted with PPP to make a concentration of platelet 3.33×108 cells/mL. The antibody, which final concentration was 1-10 μg/mL, was added thereto, and the mixture was incubated at 37° C. for 5 minutes. Hereto CaCl2 solution was added at the final concentration of 1 mM, and after additional incubation at 37° C. for three minutes, ADP solution was added at the final concentration of 5 μM. During incubation at 37° C. for 12 minutes, a light transmission of PRP was measured with MCM Hema Tracer 313M (MCMEDICAL) to investigate a platelet aggregation induced by ADP. The result is shown in FIG. 11.

ADP-induced human platelet aggregation inhibiting action was not detected in F1232-37-2 and cF1232-37-2.

By the sandwich EIA method, concentration of anti-GPVI, antibody was measured. That is, the sandwich EIA system using hGPVI-hFc having a sequence of human GPVI, which has been prepared as a immobilized protein by the same method as that of EXAMPLE 1, and Anti Human Kappa Light Chain HRP (DAKO) was constructed.

As a standard preparation, anti-GPVI antibody prepared in EXAMPLE 16 was used. The hGPVI-hFc was diluted to 4 μg/mL with PBS (pH 7.4) and 50 μL of the diluent was added to each well of NUNC-Immunoplate Maxisorp (NUNC). After incubation at 4° C. for overnight, the plate was washed five times with ice cold water, and 100 μL of PBS (pH 7.4) containing 2% StabilGuard (SurModics, Inc.) was added to each well for blocking. Then diluted samples for sample to be measured and standard preparation were prepared using PBS (pH 7.4) containing 0.1% BSA as a dilute solution. Blocking agent was discarded from the plate, 50 μL of the diluted sample was added to each well, and the sample was incubated at 37° C. for one hour. After completion of the incubation, each well was washed three times with 0.05% Tween20/0.9% NaCl solution. Next, 50 μL of the labeled antibody diluted with PBS (pH 7.4) containing 10% rabbit serum were added to each well and incubated at 37° C. for one hour. After completion of the incubation, each well was washed three times with 0.05% Tween20/0.9% NaCl solution. Further, 50 μL/well of tetramethylbenzidine solution (BioFX) was added. After incubation at room temperature for about ten minutes, the reaction was terminated with 50 μL of 1 mol/L hydrochloride solution, and an absorbance at 450 nm was measured with plate photometer.

In addition, when the concentration of GPVI antibody in monkey blood plasma was measured, PBS (pH 7.4) containing 10% monkey plasma and 10% rabbit plasma was used, and measurement was done in a similar manner.

At the start of test, measurement of body weight and blood drawing of the cynomolgus monkey to be tested prior to drug administration were performed. Blood drawing was done at 0.5 hours to two weeks after administration of anti-GPVI antibody. To evaluate each antibody, 1) platelet count, 2) expression of platelet membrane protein (GPIIb/GPIIIa (CD41a), GPIX (CD42a)), 3) expression of platelet GPVI, and 4) platelet aggregation (the reactivity to collagen and ADP) were determined.

That is, citrate-added blood collected from leg vein of cynomolgus monkey (male, about 5 kg) was provided to Sysmex F-820 to count platelets. Then, by centrifugation at 115×g at 25° C. for 15 minutes, platelet rich plasma (PRP) was obtained; then by 830×g at 25° C. for 20 minutes, platelet poor plasma (PPP) was obtained. Subsequently, the obtained PRP was diluted with PBS containing 0.5% immobilized FBS and 2.5 mM EDTA (hereinafter referred to as FACS buffer). The anti-human CD41a-FITC (BD Biosciences Pharmingen) for CD41a expression, anti-human CD42a-FITC (BD Biosciences Pharmingen) for CD42a expression, rabbit anti-GPVI-mFc polyclonal antibody and mouse anti-GPVI-mFc monoclonal antibody, wherein both were labeled with fluorescent dye Af488, for GPVI expression were respectively added thereto, and the mixture was stood at room temperature for 30 minutes under the condition of light shielding. After 30 minutes, the platelet was washed with FACS buffer. Then by measuring a fluorescent intensity of platelet with Flow cytometer CYTMICS FC500 (BECKMAN COULTER), expression of each membrane protein was analyzed. In addition, using the obtained PRP, by Western blotting, expression of each membrane protein was also analyzed. The method was as follows. Various PRPs obtained from the monkey ex vivo test were washed twice with 2.5 mM EDTA/PBS. The 1× Sample buffer (+β-mercaptoethanol, Protease inhibitor cocktail (Roche) and Phosphatase inhibitor cocktail (PIERCE) was added to the platelet to make 0.5×106 platelets/μL. The sample was heat-treated at 99° C. for five minutes. The treated sample was frozen by liquid nitrogen to store at −30° C. until use.

To analyze respective proteins in the monkey PRP by Western blotting, firstly protein isolation by SDS-PAGE was performed. Using a 5-20% gradient polyacrylamide gel (ATTO), the sample was loaded at 2.5×106 platelets/lane to run at 30 mA/gel. Blotting was in the usual manner, that is, by the semi-dry method, the proteins was transferred to a low fluorescent membrane (Immobilon-FL PVDF, MILLIPORE). After blotting, the membrane was lightly rinsed with 0.1% Tween 20/PBS (TPBS) and was blocked with BlockAce (Dainippon Pharmaceutical Co., Ltd.) at 4° C. for overnight. After blocking, a first antibody diluted with 10% BlockAce/TPBS was added thereto and incubated at room temperature for one hour with rotary shaking. The first antibody used was anti-GPIIIa antibody (Anti-Integrin β3, Santa Cruz), anti-GPVI antibody (a polyclonal antibody to human GPVI synthetic peptide as an antigen), and anti-GPIX antibody (Anti-CD42a, Santa Cruz). After incubation with the first antibody, the membrane was extensively rinsed with TPBS and washed with TPBS. After washing, a second antibody diluted with 10% BlockAce/TPBS was added thereto and incubated at room temperature for 30 minutes with rotary shaking. The second antibody used was Anti-Rabbit Igs HRP for GPIIIa and GPVI, and Anti-Goat Igs HRP for GPIX (both are from DakoCytomation). After incubation with the second antibody, the membrane was extensively washed with TPBS, and the protein was detected with ECL Plus (Amersham Biosciences). After incubation, luminescence was detected with Typhoon9410 (Amersham Biosciences). As the detection conditions, Fluorescence mode was used. In this mode, ECL Plus luminescent intermediate was detected. From the detected band, using the analysis software ImageQuant5.2, proteins expressed were quantified.

Meanwhile, the reactivity for collagen and ADP was studied as follows. Firstly, the PRP was diluted with PPP to make a concentration of platelet 3×108 cells/mL. Then CaCl2 solution was added at the final concentration of 1 mM, and the mixture was incubated at 37° C. for three minutes. Further, collagen solution at the final concentration of 2 g/mL or ADP solution at the final concentration of 20 μM was added. During incubation at 37° C. for 12 minutes, a light transmission of PRP was measured with MCM Hema Tracer 313M (MCMEDICAL) to investigate a platelet aggregation responsive to collagen or ADP.

When mouse monoclonal antibodies F1232-37-2 and F1199-6 were intravenously administered at 0.3 mg/kg, one day after administration, reduction of response of platelet to collagen as well as decrease in platelet GPVI level was observed. This action was continued for more than two days in mouse monoclonal antibody F1232-37-2.

In cynomolgus monkey administered with F1199-6, the reduction of platelet number was detected after administration. However, in cynomolgus monkey administered with F1237-2, no significant change in platelet number was observed.

The results are shown in FIG. 12.

20-2 Cynomolgus Monkey Ex Vivo Test for Mouse/Human Chimeric Anti-Human GPVI Antibody

The antibody cF1232-37-2 was intravenously administered to cynomolgus monkey (male) at 0.1, 0.3 and 1 mg/kg.

As shown in FIG. 13, in the animal administered with cF1232-37-2 at 0.1, 0.3 and 1 mg/kg, response of platelet to collagen rapidly reduced after administration. This action was continued for more than two days in the animal administered at 0.3 and 1 mg/kg. In addition, in the animal administered with cF1232-37-2 at 0.1 mg/kg, the response to collagen reduced over two hours after administration, and the action was continued for one day.

The antibody cF1232-37-2 was administered to cynomolgus monkey (male) every other day four times. Blood drawing was scheduled prior to administration, at the day after administration, and at the day after final administration through 17 days after. In addition, regarding the blood collected, 1) platelet count, 2) expression of platelet membrane proteins GPIIb/GPIIIa (CD41a), GPIX (CD42a)), and GPIIIa (CD61), 3) expression of platelet GPVI, and 4) reactivity to collagen-induced platelet aggregation and ADP-induced platelet aggregation were determined.

From the day after the first administration, decrease in the platelet GPVI level and reduction of response of platelet to collagen was observed. The reduction of response of platelet to collagen was continued for two days after the final administration in the animal at 0.1 mg/kg administration and for more than ten days after the final administration in the animal at 0.3 mg/kg administration. In FIG. 14, collagen-induced platelet aggregation of cynomolgus monkey and transition of platelet GPVI, GPIIIa and GPIX in 0.3 mg/kg repetitive administration test were shown. In addition, in the test, expression level of the platelet membrane proteins GPIX (CD42a) and GPIIIa (CD61) was not significantly changed.

The antibody cF1232-37-2 was subcutaneously administered to cynomolgus monkey (male) at 0.1, 0.3 and 1 mg/kg. Using the blood obtained from the successive blood drawing prior to and after administration as a sample, 1) platelet count, 2) expression of platelet membrane proteins (GPIIb/GPIIIa (CD41a), GPIX (CD42a)), and GPIIIa (CD61)), 3) expression of platelet GPVI, and 4) platelet aggregation (the reactivity to collagen and ADP) were determined. The results are shown in FIG. 15.

In the animal subcutaneously administrated at 1 mg/kg and 0.3 mg/kg, from three hours after the administration, reduction of response of platelet to collagen was observed. The day after administration, decrease in the platelet GPVI level was detected. The reduction of response of platelet to collagen was continued for more than one week after administration in the animal at 0.3 mg/kg administration and for more than two weeks in the animal at 1 mg/kg administration. Further, in the animal administered at 0.1 mg/kg, on the day of administration reduction of response to collagen was observed. In addition, expression level of the platelet membrane proteins (GPIIb/GPIIIa (CD41a), GPIX (CD42a), GPIIIa (CD61)) was not significantly changed.

To prepare mouse monoclonal antibody F1232-37-2 F(ab′)2, the mouse monoclonal antibody F1232-37-2 obtained in EXAMPLE 2 was treated with LysylEndopeptidase. That is, to the purified mouse monoclonal antibody F1232-37-2 1/10 volume of 1 M Tris buffer (pH 8.5) was added, and further Lysyl Endopeptidase (Wako) was added thereto at the ratio of 30:1 for antibody: enzyme (molar ratio) to incubate at 37° C. for three hours. At the end of incubation, TLCK (SIGMA) was added to the reaction mixture at the final concentration of 30 mM.

Subsequently, F(ab′)2 was purified. At first, to remove undigested antibody and Fc region, the antibody treated with LysylEndopeptidase was loaded to Prosep rA (Millipore). Further, for removal of LysylEndopeptidase, non-absorbed fraction on Prosep rA was loaded to Superdex 75 (Amersham) to obtain F1232-37-2 F(ab′)2. Then the obtained F(ab′)2 was dialyzed against saline (Ohtsuka) and evaluated for purity of the antibody by analysis on acrylamide gel. In addition, concentration of the antibody was determined by Bradford method using bovine IgG as a standard.

To prepare F1232-37-2 Fab, the purified mouse monoclonal antibody F1232-37-2 obtained in EXAMPLE 2 was treated with Papain (Wako). That is, the purified mouse monoclonal antibody F1232-37-2 was placed into 1 mM cysteine and 20 mM EDTA/D-PBS- (pH 7.4) buffer, and further Papain (Wako) was added thereto at the ratio of 30:1 for antibody:enzyme (weight ratio) to incubate at 25° C. for 16 hours. At the end of incubation, Iodoacetamide (Wako) was added to the reaction mixture at the final concentration of 30 mM.

Subsequently, Fab was purified. At first, to remove undigested antibody and Fc region, the antibody treated with Papain was loaded to Prosep rA (Millipore). Further, for removal of Papain, non-absorbed fraction on Prosep rA was loaded to Superdex 75 (Amersham) to obtain F1232-37-2 Fab. Then the obtained Fab was dialyzed against saline (Ohtsuka) and evaluated for purity of the antibody by analysis on acrylamide gel. In addition, concentration of the antibody was determined by Bradford method using bovine IgG as a standard.

Dissociation constant for F1232-37-2 (whole antibody) prepared in EXAMPLE 2, and F(ab′)2 and Fab prepared in 22-1 and 21-2 was measured with Protein Interaction Analyzer BIACORE 3000 (BIACORE). That is, to sensorchip CM5 (BIACORE) hGPVI-hFc prepared in EXAMPLE 1 was bound according to the manual. Then a series of dilution from 0 to 800 nM of each antibody was prepared with HBS-EP buffer (BIACORE) and analyzed with BIACORE 3000. After completion of antibody binding, chip was regenerated with glycine buffer (pH 1.5) (BIACORE). Regarding the results of whole antibody and F(ab′)2, evaluation software Bivalent analyte (BIACORE) was used for analysis; 1:1 Binding for Fab analysis. Then the dissociation constant was calculated. As a result, dissociation constant of F1232-37-2 F(ab′)2 and Fab was about 0.7 and 0.6, respectively, when that of whole antibody defined as one. It was found that both have an equivalent affinity in comparison with whole antibody.

From cynomolgus monkey, by the method described in EXAMPLE 20, blood was collected. The PRP prepared was diluted with PPP to make a concentration of platelet 3.33×108 cells/mL. The F1232-37-2 Fab, of which final concentration was 1-100 μg/mL, was added thereto, and the mixture was incubated at 37° C. for five minutes. Hereto CaCl2 solution was added at the final concentration of 1 mM, and the mixture was incubated at 37° C. for three minutes. Hereto, collagen solution was added at final concentration of 2 μg/ml. Then, during incubation at 37° C. for 12 minutes, a light transmission of PRP was measured with MCM Hema Tracer 313M (MC MEDICAL) to investigate a platelet aggregation responsive to collagen. The results are shown in FIG. 16.

Collagen-induced platelet aggregation inhibiting action of F1232-37-2 Fab was weaker than that of cF1232-37-2.

F1232-37-2 F(ab′)2 prepared in EXAMPLE 21 was subcutaneously administered to cynomolgus monkey at 1 mg/kg. Using a blood from the successive blood drawing prior to and after administration as a sample, expression of platelet GPVI and platelet aggregation (the reactivity to collagen and ADP) were measured.

As shown in FIG. 17, in the animal administrated with F1232-37-2 F(ab′)2, from 12 hours after the administration, reduction of response to collagen was observed. The day after administration, decrease in the platelet GPVI expression level was detected. The reduction of response to collagen was continued for more than two days. In addition, expression level of the platelet membrane proteins (FcγRII (CD32), GPIX (CD42a), and GPIIIa (CD61)) was not significantly changed.

On the start of measurement of bleeding time, firstly weight of cynomolgus monkey was weighed. Then no abnormality concerning hematological parameter, hemostatic parameter and platelet function was confirmed. At 1 mg/kg, cF1232-37-2 was subcutaneously administered and the bleeding time was measured at three hours and 48 hours after administration.

An injection needle was inserted to flank tail vein of cynomolgus monkey (male, 2.5 to 5 kg), and then the measurement of bleeding time was started.

In the animal administered with cF1232-37-2, no significant prolongation of bleeding time in comparison with the time prior to administration was detected.

To enhance a expression efficiency, cF1232-37-2 heavy chain-expressing plasmid (pTK-2510) was digested with EcoRI and NcoI. To the site between promoter and initiation codon, a fragment having Kozak sequence (Kozak, M. et al., J. Mol. Biol., 196, 947-950, 1987) consisting of sense strand: 5′-AATTCGCCGCCACC-3′ (SEQ ID NO: 291) and antisense strand: 5′-CATGGTGGCGGCG-3′ (SEQ ID NO: 292) was inserted to construct pTK-2571. In a similar manner, Kozak sequence was inserted into cF1232-37-2 light chain-expressing plasmid (pTK-2511) to construct pTK-2572.

Next, by cleaving pTK-2572 with restriction enzymes SspI and Sse8387I and blunting termini of the obtained fragment, the light chain expressing unit (EF promoter, gene encoding antibody light chain and polyA signal sequence) was prepared. Meanwhile, by cleaving TK-2571 with restriction enzymes Sse8387I, which cannot digest a heavy chain expressing unit, blunting and inserting the light chain expressing unit and marker unit to the above site, both chain stably coexpressing plasmid, in which three units are present on one vector, was constructed. Herein, marker unit is a gene fragment wherein an appropriate promoter/enhancer and polyA signal sequence is added to mouse dihydrofolate reductase (mDHFR) for gene amplification. Using four kinds of synthetic promoter/enhancer sequence (derived from adenovirus promoter, thymidine kinase promoter, SV40 promoter/enhancer and SV40 promoter), four kinds of both chain stably co-expressing plasmids (pTK-2550, pTK-2575, pTK-2576 and pTK-2577, respectively) were constructed (see FIG. 18).

25-2 Preparation of Transformant of CHO Cells, in which Mouse/Human Chimeric F1232-37-2 is Expressed

To DHFR gene-deficient CHO cells, expression plasmid pTK-2577 constructed in EXAMPLE 25-1 was transfected to establish CHO transformant producing chimeric antibody. That is, CHO DXB11, which had been naturalized in EX-CELL 325 PF CHO (JRH Bioscience) containing HT media Supplement (50×) Hybri-Max (Sigma; used at the final concentration of 1×) and 200 mM L-glutamine (Sigma; used at the final concentration of 4 mM), was centrifuged on the day of transfection to inoculate into flask at the density of 8×106 cells/150 cm2 Roux. Using 125 μl of FuGENE6 (Roche Diagnostics), 12.5 μg of expression plasmid pTK-2577 was prepared in accordance with FuGENE6-attached protocol and introduced into the above CHO DXB11. Under the condition of 5% CO2, the cells were cultured at 37° C. for two days and harvested. The cells were washed twice with EX-CELL 325 PF CHO medium without HT and with 4 mM L-glutamine (hereinafter referred to as EX-CELL (HT-) and resuspended in EX-CELL (HT-). Then at the density of 12,500-50,000 cells/well the cells were inoculated to 96 well plate and continued to culture under the condition of 5% CO2 at 37° C. During the culture, every three or four days half of medium was replaced with fresh EX-CELL (HT-). After about one month cultivation, colony-developed cells in the well were transferred to a new plate, and cF1232-37-2 expressed by CHO cells as a host in culture supernatant was measured by EIA method described in EXAMPLE 19. The cells wherein the cF1232-37-2 expression had been found in the supernatant was obtained as a cE1232-37-2/CHO producing transformant.

In a similar manner, CHO DXB11 was transformed with cF1232-37-2 expressing plasmids pTK-2550, pTK-2575 and pTK-2576, and cF1232-37-2/CHO produced by the obtained recombinant cells was determined by EIA method described in EXAMPLE 19. As shown in Table 15, it was confirmed that the productivity of cF1232-37-2/CHO is different depending on a kind of mDHFR promoters used as a selection marker, and by utilizing a relatively weak promoter as an expression promoter of selection marker highly expressing clone is obtained.

TABLE 15 Result of transfection experiment using each cF1232-37-2/CHO expression plasmid Introduced Number of clones having productivity of 1.8 mg/L plasmid or more/Number of clones evaluated pTK-2550  0/552 pTK-2576  1/352 pTK-2577 40/352 pTK-2575 19/252

The chimeric antibody expressing CHO transformant obtained in 26-2 was selectively cultured in EX-CELL (HT-) containing methotrexate (hereinafter referred to as MTX) to select clones, in which the production level of chimeric antibody is increased, for gene amplification.

That is, the transformant obtained in EXAMPLE 25-2 was suspended in EX-CELL (HT-) containing 30 or 100 nM MTX and inoculated to 96-well plate. During the culture, every three or four days half of medium was replaced with fresh EX-CELL (HT-) containing 30 or 100 nM MTX. The culture was continued in 5% CO2 at 37° C. until colony was developed. Expression level of the obtained colony into the culture supernatant was confirmed by EIA method and the clone increasing the productivity was selected. As a result, the transformant wherein its productivity is increased at about two to ten fold was obtained. In addition, by repeting selective culture of the gene-amplified transformant in medium with three to tenfolded concentration of MTX, the clone, in which the productivity is further increased, can be obtained.

25-4 Production of cF1232-37-2 by cF1232-37-2 Expressing CHO Cells

The clone CHO-G32DS25H8 cells obtained in 25-3 was inoculated to EX-CELL (HT-) containing 100 nM MTX at the density of 1.5×105 cells/ml and cultured at 37° C. for seven days. The obtained culture supernatant was used for the following purification.

25-5 Purification of cF1232-37-2/CHO

The procedures described below were performed at 4° C. if not otherwise specified.

The CHO cell culture supernatant prepared in 25-4 was clarified at room temperature using a capsule cartridge filter (Toyo Roshi Kaisha, Ltd.) having a one micrometer pore size as a prefilter and Fluorodyne filter having a 0.22 μm pore size and Fluorodyne filter (PALL) as a real filter. The clarified culture supernatant was loaded onto Protein A (rmp Protein A Sepharose Fast Flow; GE Healthcare/Amersham Biosciences) previously equilibrated with PBS- (Sigma). After washing non-absorbed proteins with PBS-, proteins non-specifically bound was eluted with 10×PBS- (Sigma). Thereafter the antibody specifically bound to Protein A was eluted with 100 mM glycine-HCl buffer (pH 3.0). The volume of eluate was measured and by adding a 1/10 volume of 2 M Tris-HCl (pH 8.5), pH was immediately neutralized to obtain an eluate containing purified antibody. The eluate was concentrated by ultrafiltration with Amicon PM10 ultrafiltration disc (MILLIPORE), and dialyzed against saline (OHTSUKA) to obtain a final purified preparation.

25-6 The Reactivity for Antigen Binding of cF1232-37-2/CHO

In the same method as that of EXAMPLE 5, dissociation constant of cF1232-37-2 for hGPVI-hFc was measured. As a result, it was exhibited that cF1232-37-2/CHO has a sufficient affinity to hGPVI-hFc and is equivalent to that of cF1232-37-2 transiently expressed in COS cells and prepared (sometimes abbreviated as cF1232-37-2/COS).

25-7 Confirmation of the Reactivity for cF1232-37-2/CHO

To confirm the reactivity of cF1232-37-2/CHO, using binding ability to hGPVI-hFc as an index, the competitive test with cF1232-37-2 prepared by transient gene expression system with COS cells was performed. The measurement method was in accordance with the method described in 17-3 of EXAMPLE 17. Using hGPVI-hFc as an immobilized protein, a competitive activity to labeled cF1232-37-2/COS was confirmed when unlabeled cF1232-37-2/CHO or cF1232-37-2/COS was added at a variety of concentration. The results are shown in FIG. 19. It was confirmed that the reactivity of cF1232-37-2/CHO is equivalent to that of cF1232-37-2/COS.

Among four kinds of mouse anti-GPVI antibodies, as the first choice for designing and preparing reconstituted human antibody, F1232-37-2 could be selected. Design of humanized antibody was performed by replacing three CDR of respective variable regions of this antibody with CDRs of the known antibody derived from human hybridoma. As a heavy chain, NEW (Saul, F. et al., J. Biol. Chem. 253, 585-597) and Eu (Cunningham B. et al., Biochemistry 9, 3161) were selected; as a light chain (K chain), REI (Epp, O. et al., Eur. J. Biochem. 45, 513-524) and framework of Eu (hereinafter referred to as FR) was used as a receptor. As NEW-HA, Eu-HA, REI-KA and Eu-KA, the first amino acid sequences (version A; FIGS. 20 and 21) were designed, respectively. However, since Eu-HA possesses extremely rare sequences in FR3 and FR4, Eu-HC (version C) having a more general sequence was re-designed. Next, from these amino acid sequences, gene sequences enable to express the above sequences were formulated. Then an intact sequence was divided into several fragments, and DNA derived from each fragment was synthesized. By joining with each fragment using ligase, a gene fragment encoding variable region was obtained. This gene fragment was amplified with primers having appropriate restriction enzyme cleavage sites (see EXAMPLE 10 in priority application). As a result, it could be bound to human antibody constant region (IgG4 as a heavy chain, and K chain as a light chain) to obtain an intact gene encoding humanized antibody. Further, by integrating this gene downstream of EF promoter of expression vector pEF2cew, plasmids enable to express respective humanized antibodies were constructed. These plasmids were designated pTK-2560 (NEW-HA), pTK-2632 (Eu-HC), pTK-2561 (REI-KA) and pTK-2631 (Eu-KA), respectively.

These humanized antibody-expressing plasmids were expressed in combination with the chimeric antibody-expressing plasmid (heavy chain: pTK-2571; light chain: pTK-2572). Then it was investigated whether the antibodies produced maintain the binding property to antigen. As a result, in Eu-HC, REI-KA and Eu-KA, expression and the binding activity to antigen could be recognized in any combination of chimeric antibody or humanized antibody. On the other hand, NEW-HA shows neither expression nor the binding activity in any combination. Consequently, a number of mutants were prepared by introducing mutation into the amino acid sequence of each FR of NEW-HA (replacing human sequence with mouse sequence), and the binding activity to antigen was comparatively studied. As a result, in NEW-HAN, a version wherein four mutations is finally introduced into FR1, expression of humanized antibody molecule and restoration of the binding activity to antigen were confirmed.

Further, it was perceived that by the competitive assay using loop-substituted GPVI-Fc, each humanized antibody exhibited the same binding specificity as that of original chimeric antibody (FIG. 22).

To cynomolgus monkey (male) the chimeric antibody cF1249-24-1 or cF1249-22-1 prepared in EXAMPLE 16 was intravenously administered at 1 mg/kg. Using the method shown in EXAMPLE 20, various analyses were performed. As a result, after administration, both antibodies caused decrease in platelet GPVI level. At the same time, both reduced responsibility of platelet against collagen. As shown in FIG. 23, the effects were maintained over two days for cF1249-24-1 and for six hours for cF1249-22-1. Further, both antibodies had no effect on expression level of the platelet membrane protein (GPIIIa (CD61)).

Firstly, weight of cynomolgus monkey (male, 2.5 to 5 kg) was measured. Then no abnormality concerning hematological parameter, hemostatic parameter and platelet function was confirmed. At 1 mg/kg, cF1232-37-2 was subcutaneously administered. The bleeding time was measured prior to administration and at three hours and 48 hours after administration. For comparison, eptifibatide was intravenously administered at 0.03, 0.1 and 0.3 mg/kg, and the bleeding time was measured at five minutes after administration in a similar manner. In addition, a collagen-induced platelet aggregation at the time when the bleeding time was measured was assayed by the method described in EXAMPLE 20.

After inserting an injection needle to the vein of both sides of tail of the cynomolgus monkey, a chronograph was activated and at the same time the needle was pulled out. The blood from the vein was suck up to filter paper (Advantec, filter paper for quantification No. 2, 150 mm) every five seconds. This operation was repeated during the blood adhered to the filter paper, and by quintupling the number of the filter paper with the bloodstain, it made a bleeding-time (seconds).

As a result, in the group administered with eptifibatide, under the condition of inhibiting a collagen-induced platelet aggregation, a significant prolongation of the bleeding time was detected. However, in the group subcutaneously administered with 1 mg/kg of cF1232-37-2, under the condition of inhibiting a collagen-induced platelet aggregation, no significant prolongation of the bleeding time was detected in comparison with that prior to the administration (FIG. 24).

Using HeLa genomic DNA as a template, Cμ1 region was amplified with a primer pair (IgM-b and IgM-c); Cμ2 region with a primer pair (IgM-d and IgM-e); Cμ3 region with a primer pair (IgM-f and IgM-g); and Cμ4 region with a primer pair (IgM-h and IgM-j), respectively. Using a mixture of the amplified product as a template, PCR with a primer pair (Nae-IgM and IgM-j) was performed again. As a result, the gene fragment, wherein each region was connected, was amplified. The amplified product was cloned into pT7-Blue(T) vector (TA cloning). By analyzing the sequence, it was confirmed that it was the gene sequence coding for human μ-chain constant region. The plasmid was designated pT7-IgM(Nae I).

(SEQ ID NO: 293) IgM-b GGAGTGCATCCGCCCCAACCCTT (SEQ ID NO: 294) IgM-c GCAGCTCGGCAATCACTGGAAGAGGCACGT (SEQ ID NO: 295) IgM-d ACGTGCCTCTTCCAGTGATTGCCGAGCTGC (SEQ ID NO: 296) IgM-e TGGCTGTGTCTTGATCGGGGCCACACATGG (SEQ ID NO: 297) IgM-f CCATGTGTGGCCCCGATCAAGACACAGCCA (SEQ ID NO: 298) IgM-g TGTGCAGGGCCACCCCCTTGGGCCGGGAGA (SEQ ID NO: 299) IgM-h TCTCCCGGCCCAAGGGGGTGGCCCTGCACA (SEQ ID NO: 300) IgM-j GTTGACACGGTTAGTTTGCATGCA (SEQ ID NO: 295) Nae-IgM GCCGGCAGTGCATCCGCCCCAACC

Firstly, by cleaving pTK-2510 (the IgG4-type cF1232-37-2 heavy chain expressing plasmid described in EXAMPLE 16) with BamHI and Aor51HI, a gene fragment encoding γ4-chain region was removed to obtain a remaining segment of the vector (including EF-1a promoter for expression and variable region of cF1232-37-2). To the vector segment, the gene fragment encoding μ-chain constant region obtained by cleaving pT7-IgM(Nae I) with BamHI and NaeI was inserted to construct a chimeric IgM-type F1232-37-2 expressing plasmid (pTK-2820), in which γ4-chain of the constant region of cF1232-37-2 was replaced with μ-chain. In addition, in the similar way, from the IgG4-type cFI232-43-3 heavy chain expressing plasmid (pTK-2504), a chimeric IgM-type F1232-43-3 expressing plasmid (pTK-2822) was constructed.

Using HeLa genomic DNA as a template, the first exon fragment was amplified with a primer pair (IgJ-a and IgJ-d); the second exon fragment with a primer pair (IgJ-c and IgJ-f); the third exonfragment with a primer pair (IgJ-e and IgJ-h); and the fourth exon fragment with a primer pair (IgJ-g and IgJ-j), respectively. Using a mixture of the amplified product as a template, PCR with a primer pair (IgJ-b and IgJ-i) was performed again. As a result, the gene fragment, wherein each region was connected, was amplified. The amplified product was cloned into pT7-Blue(T) vector (TA cloning). By analyzing the sequence, it was confirmed that it was the gene sequence coding for human J-chain constant region. The plasmid was designated pT7-IgJ. By cleaving pT7-IgJ with XbaI and BamHI, a gene fragment encoding the J-chain was obtained. Then by inserting the fragment into the XbaI/BamHI site where is present downstream the EF-1α promoter of expression vector (pEF2cew), human J-chain expressing plasmid (pTK-2393) was constructed.

(SEQ ID NO: 302) IgJ-a CACTCCTTATAGATCACACACCT (SEQ ID NO: 303) IgJ-b AAGTGAAGTCAAGATGAAGAACC (SEQ ID NO: 304) IgJ-c CTGTTCATGTGAAAGCCCAAGAAGATGAAA (SEQ ID NO: 305) IgJ-d TTTCATCTTCTTGGGCTTTCACATGAACAG (SEQ ID NO: 306) IgJ-e AAACATCCGAATTATTGTTCCTCTGAACAA (SEQ ID NO: 307) IgJ-f TTGTTCAGAGGAACAATAATTCGGATGTTT (SEQ ID NO: 308) IgJ-g CCATTTGTCTGACCTCTGTAAAAAATGTGA (SEQ ID NO: 309) IgJ-h TCACATTTTTTACAGAGGTCAGACAAATGG (SEQ ID NO: 310) IgJ-I TTAGTCAGGATAGCAGGCATCTG (SEQ ID NO: 311) IgJ-j AGAGCTATGCAGTCAGC

After mixing an appropriate amount of the expression plasmid prepared in 29-1 above with the transfection reagent (FuGENE6, Roche Diagnostics), the mixture was dropped into COS cell culture system to perform transfection. Then a chimeric antibody was transiently expressed. For expression of mouse/human chimeric IgM-type F1232-37-2 (hereinafter abbreviated as cF1232-37-2 (IgM)), a heavy chain-expressing plasmid pTK-2820, a light chain-expressing plasmid pTK-2474 and J-chain plasmid pTK-2393 were co-transfected. Also, for expression of mouse/human chimeric IgM-type F1232-43-3 (hereinafter abbreviated as cF1232-43-3 (IgM)), a heavy chain-expressing plasmid pTK-2822, a light chain-expressing plasmid pTK-2514 and J-chain plasmid pTK-2393 were co-transfected.

After transfection, the cells were cultivated for 3 days at 37° C. under 5% CO2-95% air. The culture supernatant was salted out by 60% ammonium sulfate and concentrated to purify it by an affinity chromatography with protein L column (ImmunoPure Immobilized ProteinL, PIERCE).

To each well of immunoplate immobilized with hGPVI-hFc (prepared by the method described in EXAMPLE 2 and the like), cF1232-37-2 (IgM) and cF1232-43-3 (IgM) was added and incubated at 37° C. for one hour. After washing the well with saline containing 0.05% Tween20, 50 μL of peroxidase-labeled anti-human μ-chain antibody (DAKO, P0322), which had been diluted by 2000-fold with 10% rabbit serum-containing D-PBS, was added to each well. After incubation at 37° C. for one hour, coloring reaction by the TMB substrate was done to measure an absorbance at 450 nm with plate spectro-photometer (Molecular Devices, Wako). As a result, it was confirmed that both cF1232-37-2 (IgM) and cF1232-43-3 (IgM) bind to hGPVI-hFc.

Cynomolgus monkey and human PRPs prepared by the method described in 18-1 of EXAMPLE 18 were diluted with FACS buffer to make the platelet concentration 3.75×108 cells/mL. To the diluted PRP, cF1232-37-2 (IgM) or cF1232-43-3 (IgM) prepared in the above 29-2 was added at the final concentration of 3-4 μg/mL, and stood at 25° C. for 30 minutes. After incubation, the platelet was washed with FACS buffer. Subsequently, after adding anti-human IgM-FITC (BD Biosciences Pharmingen) and standing at room temperature for 30 minutes, the platelet was washed with FACS buffer and the fluorescence intensity of the platelet was measured with flow cytometer CYTOMICS FC500 (BECKMAN COULTER). As a result, it was confirmed that both antibodies of cF1232-37-2 (IgM) and cF1232-43-3 (IgM) bind to the monkey and the human platelet.

By the method according to 18-1 of EXAMPLE 18, platelet rich plasma (PRP) was prepared from the monkey whole blood. To the prepared PRP, ACD-A (acid-citrate-dextrose) was added to adjust to pH6.5. After preparation, by centrifugation at 2000 rpm a platelet was precipitated. Then the platelet was washed with HEPES Buffer (adjusting to pH6.5 with ACD-A). After that, an appropriate volume of HEPES Buffer (pH7.4) was added to make the platelet suspend. With Sysmex the count of washed platelets prepared was measured.

The washed platelet was fractionated to microtube to make the platelet concentration 1.5×107 PLT/tube. Thereto 2 μL each of 10 mM CaCl2 and 10 mM MgCl2 was added and the volume was filled up to 18 μL with PBS. 5.0 μg/mL of Convulxin, cF1232-37-2, cF1249-22-1, cF1249-24-1, or cF1249-18-2 (each concentration 1 mg/mL) was added at 2 μL and mixed and the mixture was incubated at room temperature at room temperature for one hour. After incubation, by addition of 10 μL of 10 mM EDTA, the reaction was terminated. Then by centrifugation at 15000 rpm at room temperature for one minute, supernatant was isolated to precipitate. To each fraction, Sample Buffer was added, and the mixture was heated at 99° C. for 5 minutes. This sample was stored at −30° C., extracted in the use eve and used.

According to the method of EXAMPLE 20, by performing SDS-PAGE and Western blotting, GPVI in each sample was detected. As a result shown in FIG. 25, for the Convulxin addition, GPVI antigen was detected in the supernatant after incubation, but GPVI antigen was not detected by addition of cF1232-37-2, cF1249-22-1, cF1249-24-1 and the cF1249-18-2, and no shedding of the GPVI antigen was induced by the anti-GPVI antibodies tested.

To prepare a PEGylated F1232-37-2 antibody, a purified F1232-37-2 was reacted with linear PEG (NEKTAR) with 20 KD having succinimide group. That is, the purified F1232-37-2 was exchanged into PBS (pH7.4) buffer and PEG was added to make the ratio of antibody to PEG 1:10 (the molar-ratio) and incubated at 37° C. for one hour.

Next, PEGylated antibody was purified. In order to isolate unmodified antibody from PEGylated antibody, PEGylated antibody was loaded to anion exchange column Q Sepharose HP (Amersham). The purity was evaluated by analyzing the acrylamide gel (FIG. 26). Further, the concentration of the antibody was calculated with Bradford method (Bio-Rad) using Bovine IgG as a standard.

To prepare PEGylated F1232-37-2 Fab antibody, F1232-37-2 Fab antibody prepared in 21-2 of EXAMPLE 21 was used for PEGylation in the same method as the above-mentioned method.

Next, PEGylated Fab antibody was purified using Superdex 75 (Amersham). The purity was evaluated by analyzing the acrylamide gel (FIG. 26). Further, the concentration of the antibody was calculated with Bradford method (Bio-Rad) using Bovine IgG as a standard.

Binding activity to the antigen was detected by ELISA. To each well of immunoplate immobilized with hGPVI-hFc (prepared by the method described in EXAMPLE 2 and the like), F1232-37-2 antibody and F1232-37-2 PEGylated antibody, wherein both antibodies had been adjusted to the same concentration, was added and the plate was incubated at 37° C. for one hour. The well was washed with saline containing 0.05% Tween20. Then 50 μL of peroxidase-labeled anti-mouse immunoglobulin antibody (DAKO, P260), which had been diluted by 1000-fold with 10% rabbit serum-containing D-PBS, was added to each well. After incubation at 37° C. for one hour, coloring reaction by the TMB substrate was done to measure an absorbance at 450 nm with plate spectro-photometer (Molecular Devices, Wako). As a result, it was confirmed that both F1232-37-2 antibody and PEGylated F1232-37-2 antibody have almost equivalent binding activity.

It is thought that the sequences of the anti-GPV monoclonal antibody described in EXAMPLES 9 and 15 are derived from several kinds of certain antibody gene. In particular, it was recognized that the repertoire selection of the antibody that recognizes loop 9 of human GPVI is characteristic. An antibody gene is composed by the combination in the germ-line antibody gene segment (H, D, and J of the heavy chain, V and J of the light-chain) and moreover in many cases, is formed accompanied with the somatic mutation (see Immunoglobulin Genes 2nd eds. T. Honjo and F. W. Alt, Academic Press, 1995). Therefore, using the nucleic acid base sequences of heavy chain variable region and the nucleic acid base sequences of light-chain variable region of these antibodies as a query sequence, and database for Ig germ-line V gene of NCBI (National Center for Biotechnology Information) as a subject, Ig-BLAST search was done (http://www.ncbi.nlm.nih.gov/igblast/. As a result shown in Table 16, the variable region nucleic acid base sequence of each antibody showed a high identity % (Identity %) to the mouse germ-line antibody heavy-chain-gene H segment, D segment, J segment (VH, DH and JH segments), and an antibody light-chain gene V segment and J segment (VL and JL). In addition, in Table 16, three segments having highest score to each clone were disclosed. It is estimated that any segment is the one, which composes the germ-line antibody gene, from which each antibody gene is originated. However, among them, it is thought to have the highest possibility that each clone is derived from a gene composed of a combination of the first line segment.

The antibody of the present invention may be useful for prevention and/or treatment of especially human diseases such as diseases caused by activation or aggregation of platelet, or vascular endothelial disorders or arteriosclerotic reaction, and can be used for prevention and/or treatment of the disease caused by the thrombus or the embolism, for example, thrombosis, embolism, or arteriosclerotic diseases. Further, by the method of detecting GPVI in the test samples using the antibody of the present invention, diagnosis of diseases can be performed. Especially, it is possible to use for diagnosis of especially human diseases such as thrombus, embolism or arteriosclerotic diseases.

TABLE 16 Heavy V D Antibody VH family ID % idnt total ID % idnt total F1246-1-1 7183(VH5) 3:3.9 99 286 289 DSP2.7 90 11 12 7183(VH5) VH7183.7b 98.9 279 282 DSP2.5 90 11 12 7183(VH5) VH7183.27b 98.6 278 282 DFL16.2 100 8 8 F1249-3-2 7183(VH5) VH7183.27b 98 288 294 DSP2.9 100 8 8 7183(VH5) 3:3.9 97.9 285 291 DSP2.6 100 7 7 7183(VH5) VH7183.7b 97.6 287 294 DSP2.4 100 7 7 F1232-17-1 7183(VH5) 3:3.9 99 288 291 DFL16.1 86 20 23 7183(VH5) VH7183.7b 99.3 284 286 DFL16.1j 78 18 23 7183(VH5) VH7183.27b 99 283 286 F1232-18-3 7183(VH5) VH7183.9 99.7 289 290 DFL16.2 100 9 9 7183(VH5) VH37.1 97.9 284 290 DSP2.2 100 7 7 7183(VH5) VH76-1BG 99.6 265 266 DSP2.13 100 7 7 F1232-19-1 7183(VH5) 3:3.9 100 291 291 DFL16.1 95 21 23 7183(VH5) VH7183.7b 100 288 288 DFL16.1j 86 20 23 7183(VH5) VH7183.27b 99.7 287 288 DFL16.2 92 12 13 F1201-18 7183(VH5) VH7183.9 99 285 288 DSP2.9 100 7 7 7183(VH5) VH7183.3b 97.9 282 288 DSP2.13 100 7 7 7183(VH5) VH7183.14 97.6 282 289 F1245-4-1 J558(VH1) V186.2 94.5 276 292 DSP2.2 100 14 14 J558(VH1) V145 94.5 276 292 DSP2.6 92 13 14 J558(VH1) J558.33 94.2 275 292 DSP2.4 92 13 14 F1245-7-1 J558(VH1) V186.2 95.9 281 293 DFL16.1 93 14 15 J558(VH1) V145 95.9 281 293 DSP2.13 80 12 15 J558(VH1) J558.33 95.9 281 293 F1245-5-1 J558(VH1) V186.2 95.2 280 294 DSP2.9 100 7 7 J558(VH1) V145 95.2 280 294 J558(VH1) J558.33 95.2 280 294 F1245-6-2 J558(VH1) V186.2 95.6 281 294 DSP2.9 100 7 7 J558(VH1) V145 95.6 281 294 J558(VH1) J558.33 95.6 281 294 F1249-18-2 J558(VH1) V186.2 94.6 278 294 DFL16.1 83 10 12 J558(VH1) V145 94.6 278 294 DSP2.5 90 9 10 J558(VH1) J558.33 94.6 278 294 DSP2.7 90 9 10 F1249-5-1 J558(VH1) V186.2 94.2 277 294 DSP2.5 90 11 12 J558(VH1) V145 94.2 277 294 DSP2.6 100 9 9 J558(VH1) J558.33 94.2 277 294 DSP2.7 90 9 10 F1249-22-1 J558(VH1) V303 99.3 292 294 DFL16.2 100 14 14 J558(VH1) VH124 97.3 286 294 DFL16.1 100 10 10 J558(VH1) V304 96.1 271 282 DFL16.1j 100 9 9 F1249-20-1 J558(VH1) J558.n 98 288 294 DSP2.2 100 9 9 J558(VH1) J558.n1 97 225 232 DSP2.6 100 7 7 J558(VH1) J558.5 88.1 258 293 DSP2.4 100 7 7 F1249-24-1 J558(VH1) J558.n 98.3 289 294 DSP2.2 100 9 9 J558(VH1) J558.n1 97.4 226 232 DSP2.6 100 7 7 J558(VH1) J558.5 88.4 259 293 DSP2.4 100 7 7 F1249-30-1 J558(VH1) J558.n 97.6 287 294 DSP2.2 100 9 9 J558(VH1) J558.n1 96.6 224 232 DSP2.6 100 7 7 J558(VH1) J558.5 87.7 257 293 DSP2.4 100 7 7 F1232-24-1 J558(VH1) J558.n 98.3 286 291 DSP2.2 100 13 13 J558(VH1) J558.n1 97.4 226 232 DSP2.6 92 12 13 J558(VH1) J558.5 88.3 257 291 DSP2.4 92 12 13 F1232-7-1 J558(VH1) J558.n 94.8 276 291 DST4.2 92 12 13 J558(VH1) J558.n1 94.4 219 232 DST4-BALB/c 100 11 11 J558(VH1) J558.5 85.6 249 291 DST4-C57BL/6 90 10 11 F1232-37-2 J558(VH1) J558.n 94.5 275 291 DST4.2 92 12 13 J558(VH1) J558.n1 94 218 232 DST4-BALB/c 100 11 11 J558(VH1) J558.5 85.2 248 291 DST4-C57BL/6 90 10 11 F1232-21-1 J558(VH1) J558.13 98.6 286 290 DFL16.1 100 11 11 J558(VH1) J558.i 98.3 286 291 DFL16.2 100 9 9 J558(VH1) J558.14 97.3 283 291 DFL16.1j 100 8 8 F1201-20 J558(VH1) J558.23 98.6 281 285 DSP2.2 100 17 17 J558(VH1) J558.24 94.8 271 286 DSP2.4 94 16 17 J558(VH1) J558.22 93.4 270 289 DSP2.3 94 16 17 F1232-10-1 3609(VH8) CB17H-3 96.1 271 282 DFL16.1 100 19 19 3609(VH8) CB17H-10 95.7 270 282 DFL16.1j 100 12 12 3609(VH8) CB17H-1 93.3 263 282 F1232-43-3 3609(VH8) CB17H-3 96.5 272 282 DFL16.1 100 17 17 3609(VH8) CB17H-10 95.4 269 282 DFL16.2 100 11 11 3609(VH8) CB17H-1 93.6 264 282 Heavy Light J V J Antibody ID % idnt total ID % idnt total ID % idnt total F1246-1-1 JH4 100 53 53 21-5 100 296 296 JK1 100 38 38 JH2 85 33 39 21-10 98.3 290 295 JK2 89 31 35 21-4 91.2 270 296 JK5 89 23 26 F1249-3-2 JH2 100 45 45 21-10 100 298 298 JK1 100 37 37 JH4 85 33 39 21-5 98.3 292 297 JK2 89 31 35 JH3 100 14 14 21-4 90.9 269 296 JK5 89 23 26 F1232-17-1 JH2 100 39 39 21-5 98 290 296 JK1 100 38 38 JH4 85 33 39 21-10 96.3 284 295 JK2 89 31 35 JH3 100 14 14 21-4 91.2 270 296 JK5 89 23 26 F1232-18-3 JH2 100 45 45 21-5 97 289 298 JK1 97 36 37 JH4 85 33 39 21-10 95.3 283 297 JK2 91 32 35 JH3 100 14 14 21-2 90.9 261 287 JK5 89 23 26 F1232-19-1 JH2 100 39 39 21-5 98 290 296 JK1 100 38 38 JH4 85 33 39 21-10 96.3 284 295 JK2 89 31 35 JH3 100 14 14 21-4 91.2 270 296 JK5 89 23 26 F1201-18 JH2 100 45 45 21-5 99.7 297 298 JK5 100 35 35 JH4 85 33 39 21-10 98 291 297 JK2 100 14 14 JH3 100 14 14 21-4 90.9 270 297 JK1 89 23 26 F1245-4-1 JH2 100 38 38 19-32 100 282 282 JK1 100 38 38 JH4 84 32 38 19-17 88 242 275 JK2 89 31 35 JH3 100 14 14 19-13 88.3 204 231 JK5 89 23 26 F1245-7-1 JH2 100 45 45 19-32 100 283 283 JK4 97 36 37 JH4 85 33 39 19-17 88 242 275 JK2 100 13 13 JH3 100 14 14 19-13 88.3 204 231 JK3 100 12 12 F1245-5-1 JH3 100 41 41 19-32 100 284 284 JK1 100 38 38 JH2 100 14 14 19-17 88 242 275 JK2 89 31 35 19-13 88.3 204 231 JK5 89 23 26 F1245-6-2 JH3 100 41 41 19-32 99.6 283 284 JK1 100 38 38 JH2 100 14 14 19-17 87.6 241 275 JK2 89 31 35 19-13 87.9 203 231 JK5 89 23 26 F1249-18-2 JH4 100 53 53 19-32 100 285 285 JK1 100 36 36 JH2 85 33 39 19-17 88 242 275 JK2 89 31 35 19-13 88.3 204 231 JK5 89 23 26 F1249-5-1 JH4 100 51 51 19-32 99.6 284 285 JK1 100 36 36 JH2 85 33 39 19-17 87.6 241 275 JK2 89 31 35 19-13 87.9 203 231 JK5 89 23 26 F1249-22-1 JH2 100 41 41 8-24 99.3 300 302 JK5 97 37 38 JH4 85 33 39 8-30 92.5 273 295 JK2 89 25 28 JH3 100 14 14 8-28 90.5 258 285 JK1 89 24 27 F1249-20-1 JH3 98 46 47 21-5 98.3 291 296 JK5 100 38 38 JH2 100 14 14 21-10 96.6 285 295 JK2 89 24 27 JH4 100 11 11 21-4 89.9 266 296 JK1 89 23 26 F1249-24-1 JH3 100 47 47 21-5 99.7 295 296 JK5 100 38 38 JH2 100 14 14 21-10 98 289 295 JK2 89 24 27 21-4 90.9 269 296 JK1 89 23 26 F1249-30-1 JH3 100 47 47 21-5 99.7 295 296 JK5 100 38 38 JH2 100 14 14 21-10 98 289 295 JK2 89 24 27 21-4 90.9 269 296 JK1 89 23 26 F1232-24-1 JH3 100 47 47 21-5 98 290 296 JK5 100 38 38 JH2 100 14 14 21-10 96.3 284 295 JK2 89 24 27 21-4 91.2 270 296 JK1 89 23 26 F1232-7-1 JH4 100 49 49 21-5 100 296 296 JK2 100 38 38 JH2 85 33 39 21-10 98.3 290 295 JK1 89 31 35 21-4 91.2 270 296 JK4 92 22 24 F1232-37-2 JH4 100 49 49 21-5 100 296 296 JK2 100 38 38 JH2 85 33 39 21-10 98.3 290 295 JK1 89 31 35 21-4 91.2 270 296 JK4 92 22 24 F1232-21-1 JH2 100 41 41 ce9 97.9 278 284 JK2 100 38 38 JH4 85 33 39 cp9 94.4 268 284 JK1 89 31 35 JH3 100 14 14 cy9 84.5 191 226 JK4 92 22 24 F1201-20 JH3 100 43 43 cw9 95.8 272 284 JK2 100 38 38 JH2 100 14 14 bv9 96.1 269 280 JK1 89 31 35 cj9 93.2 261 280 JK4 92 22 24 F1232-10-1 JH2 100 42 42 12-46 96.1 273 284 JK1 100 38 38 JH4 85 33 39 12-44 95.8 272 284 JK2 89 31 35 JH3 100 14 14 12-41 94.2 258 274 JK5 89 23 26 F1232-43-3 JH2 100 41 41 12-44 96.8 275 284 JK1 100 38 38 JH4 85 33 39 12-46 96.5 274 284 JK2 89 31 35 JH3 100 14 14 12-41 95.3 261 274 JK5 89 23 26