Antibody of human origin specifically binding to activated state of platelet integrin receptor GPIIb/IIIa

This application is a National Stage of International Application PCT/EP02/11154, filed Oct. 4, 2002, published Apr. 17, 2003 under PCT Article 21(2) in English; which claims the priority of EP 01123851.6 filed Oct. 5, 2001.

FIELD OF THE INVENTION

The present invention is directed to an antibody for inhibiting platelet aggregation, and a method for identifying and/or isolating such an antibody. Furthermore, the present invention concerns the DNA coding for this antibody and a pharmaceutical or diagnostic preparation containing the antibody or its coding DNA.

BACKGROUND OF THE INVENTION

Platelets or thrombocytes play a crucial role in the field of thrombosis, myocardial infarction and unstable angina: The platelet integrin receptor GPIIb/IIIa is of particular importance since it mediates platelet aggregation by binding of the bivalent plasma molecule fibrinogen. This receptor has at least two conformational states: 1) A non-activated state, which is the default state on unstimulated platelets. In this non-activated state, the receptor demonstrates a very low affinity for its ligands and is not capable of inducing platelet aggregation. 2) An activated state which is present after platelet activation, e.g. by thrombin. In this activated state GPIIb/IIIa has undergone a conformational change, which leads to high affinity binding of fibrinogen (Shatill et al., J. Biol. Chem. 1985: 260(20): 11107-11114).

Consequently the therapeutic blockade of GPIIb/IIIa is a very effective anti-platelet strategy, since it affects the final common endpoint of the platelet activation cascade. During the last years a great variety of GPIIb/IIIa-blockers have been developed. These are either chimeric mouse/human Fab-fragments of a GPIIb/IIIa-blocking monoclonal antibody (Abciximab) (Coller B., et al., J. Clin. Invest. 1983, 72: 325-338), cyclic peptides (Eptifibatide) or polycyclic synthetic peptidomimetics (e.g. Tirofiban) (Bhatt D L and Topol E J. JAMA. 2000; 284(12):1549-58; Topol E J, et al., Lancet. 1999; 353(9148):227-31). This therapy has been proved to be effective but there still retain some problems in this context:

- especially under the therapy with Abciximab, an increased prevalence of severe thrombocytopenia is present (˜1%) (Dasgupta H., et al., Am Heart J. 2000; 140(2):206-11).
- due to the expensive production the costs of the therapy are considerably high, especially for Abciximab. (Hillegass W B, et al., Pharmacoeconomics. 2001; 19(1):41-55).
- there is an increase in bleeding complications which are especially important when GPIIb/IIIa-blockers are combined with thrombolysis.
- synthetic GPIIb/IIIa-blockers which are administered orally brought disappointing results, due to their pharmacokinetic properties, particularly a rather low affinity for the receptor. (Chew D P. et al., Circulation. 2001, 103(2):201-206).
- there is evidence that GPIIb/IIIa-blocker, especially the low molecular agents, interact with the receptor after binding. This might result in a paradoxical intrinsic activating effect (Peter K., et al., Blood. 1998; 92(9):3240-)
- reversibility of the effect of Abciximab is very slow (>12 hours)
- approx. 6% of the patients treated with Abciximab develop anti-human-chimeric antibodies (AHAC); 11% in case of patients treated repeatedly (Gawaz M., Therapie bei koronarer Herzerkrankung. Stuttgart, New York: Thieme, 1999).

All GPIIb/IIIa-blockers, currently used, are binding to the activated and non-activated receptor with similar affinity. An activation specific inhibitor might offer several advantages. For example platelet adhesion would still be intact which should result in a reduction of bleeding events. Moreover interactions with the non-activated receptor would be prevented. It would be desirable to develop a smaller GPIIb/IIIa-blocking agent with an affinity similar to an antibody, which should demonstrate better pharmacokinetic properties.

Another application for an activation specific antibody would be the detection of activated platelets, which is very useful in a variety of research and diagnostic-settings.

SUMMARY OF THE INVENTION

It is the object of the present invention to find an antibody with such improved properties, as well as to provide methods for identifying such an antibody.

This object is solved by providing the antibody according to independent claim 1. Further advantageous features, embodiments and aspects of the present invention will become more readily understandable when looking at the further independent and dependent claims, the description and the drawings.

Accordingly, the invention is directed to an antibody of human origin for inhibiting platelet aggregation, characterised in that it is effective by substantially exclusive binding to the activated state of platelet integrin receptor GPIIb/IIIa.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows a FACS analysis of a clone expressing an antibody fragment according to a first embodiment of the invention.

FIG. 2
a shows the nucleic acid sequence (SEQ ID NO: 1) of clone MB9 coding for a scFV antibody according to the present invention. FIG. 2b shows the amino acid sequence of MB9 (SEQ ID NO: 159).

FIG. 3 shows the sequence of C9 scFv (SEQ ID NO: 2) and E4 scFv (SEQ ID NO: 3).

FIG. 4 shows oligonucleotides (SEQ ID NOs: 5-22) used for the construction of the human scFv based synthetic library. BbsI restriction enzyme recognition sites are indicated in bold style, cut sites are underlined

FIG. 5 shows a schematic representation of annealing positions of oligonucleotides used for the construction of pEXHAM4/C9 and pEXHAM4/E4. Genes of the scFv's C9 and E4 cloned in pEXHAM1 are shown as boxes. Black painted areas represent CDR regions; Oligonucleotides are represented by arrows and identified by numbers (c.f. FIG. 4). BpiI restriction endonuclease recognition sites are indicated.

FIG. 6 shows vector maps of pEXHAM4/C9 and pEXHAM4/E4

FIG. 7 shows vector maps of pEXHAM7/C9 and pEXHAM7/E4

FIG. 8 lists oligonucleotides (SEQ ID Nos: 23-77) used as primers in 1. PCR for amplification of human heavy and light chain variable regions

FIG. 9 lists oligonucleotides (SEQ ID Nos: 78-132) used as primers in 2 PCR for introduction of restriction endonuclease recognition sequences (Marked in bold style)

FIG. 10 shows the FACS analysis of clones SA8, SA10 and SA11. Binding of indicated scFv's to activated (black curve) and non-activated (grey curve) thrombocytes.

FIG. 11 shows the entire nucleotide sequence (SEQ ID NO: 134) concerning the vector map pEXHAM4/E4.

FIG. 12 shows the entire nucleotide sequence (SEQ ID NO: 133) concerning the vector map pEXHAM4/C9.

FIG. 13 shows the entire nucleotide sequence (SEQ ID NO: 136) concerning the vector map pEXHAM7/E4.

FIG. 14 shows the entire nucleotide sequence (SEQ ID NO: 135) concerning the vector map pEXHAM7/C9

FIG. 15
a shows aggregometry with MB9 scFv. Inhibition of platelet aggregation by addition of MB9 scFv in different concentrations. Aggregation was monitored by the increase of light transmission.

FIG. 15
b shows the comparison of Reopro and MB9 binding to activated throbocytes by aggregometry and FACS analysis

FIG. 16 shows the schematic representation of annealing positions of oligonucleotides used for the construction of pEXHAM9/C9 and pEXHAM9/E4. Genes of the scFV's C9 and E4 cloned in pEXHAM1 are shown in boxes. Black boxes represent CDR regions; oligonucleotides are represented by arrows and identified by numbers (c.f. FIG. 17). BpiI restriction endonuclease recognition sited are indicated.

FIG. 17 shows oligonucleotides (SEQ ID Nos: 137-157) used for the construction of the human scFv based synthetic VL library. BbsI restriction enzyme recognition sited are indicated bold, cut sites are underlined.

FIG. 18 shows the FACS analysis of affinity maturated scFv SCE5 and SCE18. Binding of scFv preparations of SA2 (original cloned) and light chain shuffled derivatives SCE5 and SCE18 to non activated (filled black curve) and activated (grey curve) thrombocytes. Measurement was done with whole blood in presence of fibrinogen. “Control” demonstrates no binding of secondary anti-H is antibody.

FIG. 19 shows the amino acid sequences (SEQ ID NOs: 160 and 161) of original MB9 scFv linker (italic) connecting VH and VL-domaines (underlined) and shortened linker (italic) for diabody formation.

FIG. 20 shows the localisation and features of heavy and light chain fragments of MB9 Fab in pREFAB9.

FIG. 21 shows the binding of MB9 Fab (dark grey curve) to activated but not to non-activated human thrombocytes (mean fluorescence intensities are indicated).

FIG. 22 shows the sequence (SEQ ID NO: 158) of pREFAB9/MB9 plasmid, and MB9 scFv translation recombinant human antibody fragment (SEQ ID NO: 159).

DETAILED DESCRIPTION OF THE INVENTION

The terms thrombocyte and platelet are used synonymously in this specification. The general term “platelet integrin receptor” means “platelet integrin receptor GPIIb/IIIa”.

According to the present invention, the antibody binds to the platelet integrin receptor GPIIb/IIIa (alpha IIb/beta 3) and inhibits the binding of the natural ligand fibrogen. As detailed above, this receptor is characterised by inducing the aggregation process when fibrinogen binds to it. Through blocking this receptor, crosslinking is impossible.

Due to the more selective effects obtainable, the antibody does “substantially exclusively bind” to the activated conformation of the platelet integrin receptor. This means that its binding affinity to the activated conformation of the platelet integrin receptor is much greater than its respective affinity for binding to the inactive conformation of the platelet integrin receptor. At best, the agent is substantially unable to bind to the nonactivated conformation of the integrin receptor.

In the present specification, the term “antibody” means immunoglobulins of human origin. The immunoglobulin may be also a fragment of human immunoglobulins comprising the variable domains of the heavy and light chain. The fragment may be a single chain antibody fragment (scFv), Fab or recombinant constructs and derivatives thereof. It may be monovalent, bivalent or multivalent.

It can contain modifications to its amino acid sequence when compared to genuine antibodies and exhibit a modified domain structure. It must however, still be able to adopt the typical domain configuration found in native antibodies, as well as an amino acid sequence, which is able to bind to targets (antigens) with high specificity. Typical examples of antibodies derivatives are antibodies coupled to other polypeptides, rearranged antibody domains or fragments of antibodies. The antibody may also comprise at least one further compound, e.g. a protein domain, said protein domain being linked by covalent or non-covalent bonds. The linkage can be based on genetic fusion according to the methods known in the art. The additional domain present in the fusion protein comprising the antibody employed in accordance with the invention may preferably be linked by a flexible linker, advantageously a peptide linker, wherein said peptide linker comprises plural, hydrophilic, peptide-bonded amino acids of a length sufficient to span the distance between the C-terminal end of the further protein domain and the N-terminal end of the antibody or vice versa. The above mentioned fusion protein may further comprise a cleavable linker or cleavage site for proteinases. Thus, e.g., the antibody might be linked to an effector molecule having a conformation suitable for biological activity or selective binding to a solid support, a biologically active substance (e.g. a cytokine or growth hormone), a chemical agent, a peptide, a protein or a drug.

The antibody of the present invention is of human origin. This is a particularly important feature of the invention, since it opens the use of such antibodies to a therapy in human patients without the risk of adverse immune reactions against other “foreign” antibody types. In particular, the overall structure/sequence and the constant regions of the used antibody are of human origin. The source of the human antibody may be a phagedisplay library of natural or modified human antibody fragments, screened for antibodies with affinity for thrombocytes.

Preferably, the antibody is a single chain antibody where a VH domain is linked to a VL domain. The term “linked” means preferably a peptide bond. Such a single chain antibody is preferably a recombinant scFv antibody. Methods for producing such a single chain antibody with the above mentioned properties or of DNA sequences coding for such an antibody, its expression in suitable hosts and its recovery and purification are described for examples in WO-A-89/09622, WO-A-89/01783, EP-A-0 239 400, WO90/07861 and Colcher et al., Cancer Research 49 (1989), p. 1732-1745. The scFv employed may be a recombinant construct of single chain antibody fragment(s), if such rearrangements or changes to the sequence are necessary in order to obtain the desired product. The person skilled in the art knows methods how to modify the respective immunoglobuline domains, e.g. via amino acid deletion, insertions, substitutions and/or recombinations. Methods for introducing such modifications in the coding sequence for the immunoglobuline chain are known to person skilled in the art (e.g. Sambrook et al., Molecular Cloning—A Laboratory Manual, Cold Spring Harbor (1989), N.Y.) On the other hand, the single chain antibody fragment may for example be derived from a human IgM or IgG antibody. Alternatively, recombinant BsAb or diabodies (containing two scFv fragments preferably linked via a peptide linker) can be formed. It will be also advantageously to construct tandem diabodies by homodimerisation of single chain fragments comprising four antibody variable domains (VH and VL) of two different specificities.

Due to the huge variability of the antibody generation process in the course of an immune response, in general a large number of different sequences suitable for attacking a foreign antigen can be produced. It is clear to the skilled person that therefore, several embodiments of antibody sequences could be found for meeting the requirements of the present invention. As an example, which is tested and worked well, the antibody according to the invention may be characterized in that the fragment comprises an amino acid region, which comprises the translation product of the nucleic acid sequence of FIG. 2 (SEQ. No. 1). In a further preferred embodiment, it comprises the amino acid sequence as shown in FIG. 2 or it consists of the amino acid sequence of FIG. 2. In a further embodiment, the present invention provides nucleic acid molecules encoding a fragment, derivative or allelic variation of the above polypeptide, which have substantially the same properties as that of FIG. 2. The term “derivative” in this context means that the sequence of these molecules differ from the sequences of the nucleic acid molecules and/or of the amino acid sequence of FIG. 2 at one or several positions but have a high level of homology to these sequences. Homology hereby means a sequence identity of at least 60%, in particular an identity of at least 70 or 80%, preferably of more than 90% and particularly preferred of more than 95%. The deviations of the above-mentioned nucleic acid molecules or peptide molecules may have been produced by deletion, substitutions, insertions or recombination.

Another suitable example is a synthetic library of antibody sequences. The identified fragment comprises a heavy chain CDR3 domain which contains the sequence ELEAYCRGDCYPPYYG or a derivative thereof with comparable structure and properties. This sequence is found to be able of binding to the integrin receptor, may be because it can mimic the fibrinogen structure.

A further preferred embodiment concerns the DNA sequence coding for the single chain antibody. These DNA sequences can be inserted into a vector or expression vector. Thus, the present invention also relates to vectors and expression vectors containing these DNA sequences. The term “vector” means a plasmid (pUC18, pBR322, PBLUESCRIPT® (nucleic acid vectors for gene cloning, gene characterization, and gene transfer), etc.), a virus or any other suitable vehicle. In a preferred embodiment, the DNA sequences are functionally linked to regulatory elements, which allow their expression in procaryotic or eucaryotic host cells. Such vectors contain besides the regulatory elements (e.g. promoter) a replication origin and specific genes which allow the phenotypic selection of a transformed host cell. The regulatory elements for the expression in procaryotes (e.g. E. coli) are lac-, trp-promoter or T7 promoter, and for the expression in eucaryotes A0X1- or Gal promoter (for expression in yeasts) and CMV-, SV40-, RVS-40 promoter, CMV- or SV40 enhancer (for expression in animal cells). Further examples for promotors are metallothein I and polyhderin promoter. Suitable expression vectors for E. coli are pGEMEX, pUC derivatives, pEXHAM and pGEX-2T. Suitable promoters for the expression in yeast are pY100 and Ycpad1 and for the expression in mammal cells pMSXND, pKCR, pEFBOS, cDM8 and pCEV4.

General methods known in the art can be used for the construction of the expression vectors, which contain the DNA sequences of the present invention and suitable regulatory elements. Examples of these techniques are in-vitro recombination techniques, synthetic methods and in-vivo recombination techniques (c.f. Sambrook et al., Molecular Cloning—A Laboratory Manual, Cold Spring Harbor (1989), N.Y.). The DNA sequences according to the present invention may be also inserted into a vector in combination with DNA sequences coding for other proteins or peptides to be expressed as fusion proteins.

The present invention further concerns host cells containing these vectors. These host cells are e.g. bacteria (e.g. E. coli strains XL1blue, HB101, DH1, x1776, JM101, JM109, BL21 and SG13009), Yeasts (preferably S. cervisiae), insect cells (preferably sf9 cells) and animal cells (preferably mammal cells). Preferred mammal cells are myeloma cells, preferably mouse myeloma cells). Methods for transforming these host cells, methods for the phenotypic selection of transformants and for the expression of the DNA sequences according to the present invention by using the aforementioned vectors are known in the present technical field.

The present invention further relates to a method for the recombinant production of the (single chain) antibody by using the aforementioned expression vectors. This method comprises the cultivation of the aforementioned host cells under conditions which allow the expression (preferably stable expression) of the protein (or fusion protein) and the recovery of the protein from the culture or from the host cells. The person skilled in the art knows conditions how to cultivate transformed or transfected host cells. Suitable methods for the recombinant production of proteins are known (e.g. Holmgren, Annual Rev. Biochem. 54 (1985), 237; La Valle et al., Bio/Technology 11 (1993), 187, Wong, Curr. Opin. Biotech. 6 (1995), 517; Davies, Curr. Opin. Biotech 6 (1995), 543). Furthermore, suitable purification methods are known (e.g. preparative chromatographie, affinity chromatographie, HPLC etc.).

The invention is further directed to a process for identifying and/or isolating antibodies for inhibiting platelet aggregation by binding to the activated form of integrin receptor GPIIb/IIIa of blood thrombocytes.

Such process according to the invention comprises the following steps:

- providing a library of nucleic acids encoding for sequences of candidates;
- producing a phage library from said nucleic acids library;
- successively reacting said phage library with nonactive thrombocytes, active thrombocytes, other cells expressing nonactive integrin receptor molecules, and other cells expressing active integrin receptor molecules; and
- eluting phages bond to said thrombocytes or other cells expressing active integrin receptor molecules.

An important step of the inventive process is that the phage library is depleted of less suitable polypeptides, which either bind to nonactivated platelets, or to other components on the surface of activated platelets. Following each of the binding steps, a recovery of the selected phages should be performed, which can be done with known methods. Finally, those phages carrying polypeptides which specifically bind to the integrin receptor, are tested for their blocking activity.

The steps of selecting with other cells can be also omitted. By this modification, phages inhibiting platelet aggregation by other mechanisms may be detected.

The step of providing a library may comprise the following steps:

- isolating whole RNA from human donors;
- isolating mRNA contained in the whole RNA coding for antibody polypeptides;
- generation of cDNA; and
- deriving DNA molecules coding for single chain fragments from cDNA molecules coding for antibody polypeptides.

By this, a “natural library”, based on the antibody population of the donors, can be obtained

Alternatively, a synthetic library may be used, wherein the step of providing a library comprises the following steps:

- providing a nucleic acid containing a sequence for a single chain antibody fragment containing a heavy and a light variable domain; and
- introducing at least one randomised nucleotide sequence in a region of said single chain antibody fragment.

The region into which the at least one randomised nucleotide sequence is introduced, preferably is the CDR3 region of vH or vL such a scFv.

Said other cells may preferably be CHO cells, which are well known and may express the integrin receptor on their surface after having been transformed.

The invention is further directed to the use of a pharmaceutical composition containing the antibody, DNA or expression vectors according to the present invention for blocking the platelet integrin receptor on thrombocytes.

The invention is still further directed to the use of the antibody, DNA or expression vector according to the invention for manufacturing a pharmaceutical composition.

The subject matter of the present invention is also of diagnostic interest. It may be used for determining the number of activated thrombocytes in relation to non-activated thrombocytes in a patient. It is particular useful for monitoring the (de)activation status if the patient is treated with thrombocyte aggregation inhibitors.

The pharmaceutical or diagnostic composition may contain additionally a pharmaceutically acceptable carrier. Suitable carriers are phosphate buffered saline solutions, Water, emulsions (e.g. water-in-oil emulsions), surfactants, sterile solutions etc. The administration of the pharmaceutical composition may be orally or parenterally (e.g. topically, intra-arterially, intramuscularly, subcutaneously, intramedullarly, intrathecally, intraventricularly, intraveneously, intraperitoneally or intranasally). The suitable dosage will be determined by the medical doctor and is dependent on various conditions (e.g. age, sex, weight of the patient, kind of illness and kind of administration, etc.).

The DNA sequences of the present invention may be also inserted into a vector suitable for gene therapy, e.g. under the control of a tissue-specific promoter. In a preferred embodiment the vector containing the DNA sequences is a virus (e.g. an adenovirus, vaccinia virus or adeno-associated virus). Preferred are retroviruses. Examples of suitable retroviruses are MoMuLV, HaMuSV, MuMTV, RSV or GaLV. For gene therapy purposes the DNA sequences according to the present invention may be also transported in form of colloidal dispersions to the target cells; In this connection also liposomes and lipoplexes are mentioned (Mannino et al., Biotechniques 6 (1988), 682).

Finally, the invention is directed to a method a treating a patient, comprising the following step:

administering a pharmaceutical composition according to the invention in a pharmaceutically effective dose to the patient.

In the following, examples for the production of human scFv antibodies specific for activated platelet integrin receptor GPIIb/IIIa will be given.

EXAMPLES

General Strategy

Phage libraries for the display of single chain antibody fragments (scFv) are generated from human IgM antibody genes. Alternatively, a synthetic library is generated by randomization of the CDR3 region of the heavy chain in two scFv master frameworks of human origin. Both libraries are subtracted for not activation specific binders by incubation on resting thrombocytes prior to using them for selection on activated platelets. To focus the selection onto the GPIIbIIIa receptor, additional rounds of selection are done on in vitro cultivated cells expressing recombinant GPIIbIIIa receptor. Following the selection, scFv clones are analysed for binding to activated thrombocytes and competition of fibronogen binding by FACS analysis.

Example 1
Production of the Human scFv Antibody Fragment MB9

RNA and cDNA Preparation

Total RNA is isolated from spleen samples of six human donors and peripheral blood lymphocytes (PBL) of five healthy human donors (app. 1−5×10⁸PBLs each, RNeasy™ Midiprep.Kit, Qiagen). From total RNA poly A⁺-RNA is prepared (Oligotex mRNA Kit, Qiagen) and used for cDNA synthesis (SuperScript™ Preamplifications System, Gibco BRL/LIFE Technologies).

Amplification of Human Ig Variable Regions

Oligonucleotides used in PCR for amplification of variable regions of human immunoglobulin heavy and light chains those of FIG. 8. Heavy chains are amplified using a single IgM specific constant primer and one of a number of different primers (VH-1 to VH-7) specific for the variable region in separate PCR reactions. Accordingly lambda and kappa light chains are amplified using a single lambda or kappa specific constant primer and one out of a number of different variable primers (Vλ-1 to Vλ10 and Vκ-1 to 6). PCR is done in a volume of 50 μl using 0.5 μl cDNA, 1 unit VENT_R® exo⁻-DNA-polymerase (DNA polymerase genetically engineered to eliminate the 3′→5′ proofreading exonuclease activity associated with Vent® DNA Polymerase New England Biolabs) and 0.5 μM of each primer under following conditions: 3 min 95° C., 20×[30 sec 95° C., 1 min 55° C., 1 min 72° C.] 5 min 72° C. The products of the first PCR are purified using the PCR purification Kit (Qiagen) and used as templates for as second PCR using a corresponding set of oligonucleotide primers of FIG. 9 to introduce restriction sites for cloning. The second PCR is carried out separately for each primer set according to the first PCR but using 1 min 57° C. for annealing. Products of the second PCR of the heavy chain, the lambda light chain and the kappa light chain are pooled and purified via PCR-purification Kit (Qiagen).

Cloning of the scFv Phage Display Library

Heavy chain fragments are digested with NcoI and HindIII, light chain fragments with MluI and NotI (each New England Biolabs) according to the suppliers instructions and finally purified by gel extraction from 1% agarose gels using the Gel Extraction Kit (Qiagen). To create a sublibrary, the heavy chains are cloned first into the phagedisplay vector pEXHAM1 (FIG. 1) containing a stuffer scFv. Vector DNA is cut with NcoI and HindIII, purified via gel extraction and ligated separately with heavy chain fragments originating from different donors. Ligation is done in 20 μl volume using 50 ng vector, 9 ng heavy chain fragment and 1 unit T4 DNA-ligase (Roche) for three hours at room temperature. The ligation mixture is precipitated, resuspended in 10 μl water and mixed with 35 μl of electrocompetent E. coli XL1 blue cells (Stratagene) for electroporation according to the suppliers instructions. Transformed cells are plated on selective LB agarose plates containing 50 mM glucose, 100 μg/ml ampicillin and 20 μg/ml tetracyclin and incubated at 30° C. over night. The size of the sublibraries is in the range of 1.5×10⁶to 7.1×10⁷as determined by plating appropriate dilutions.

Bacterial clones are scraped from the plates and used for DNA-maxipreparation (Qiagen) to prepare the vector DNA for cloning of the complete libraries. Sublibrary DNA is cut with MluI and NotI, purified by gel extraction and ligated with lambda and kappa light chain fragments separately. Ligation is done in 20 μl volume using 1 μg vector DNA and a two fold molar excess of light chain DNA. After incubation with 1 unit T4 ligase (Roche) over night at 8° C., the ligation mixture is precipitated and redissolved in 2.5 μl Tris 10 mM, pH8.5. Of this 2 μl are used for transformation of 50 μl aliquots of electrocompetent XL1 blue cells. Cells are plated on selective agarplates and the number of transformants is determined by plating of appropriate dilutions as described above. The total size of all libraries generated from spleen and PBL RNA material is 1.75×10⁹.

Library Rescue

For phage display of scFv's, the library is inoculated in 250 ml aliquots of LB medium supplemented with 50 mM glucose, 100 μg/ml ampicillin and 20 μg/ml tetracyclin at a start OD600 of 0.025 ensuring that the number of cells exceeds the complexity by a factor of 10. Cells are incubated at 37° C. and 200 rpm until an OD600 of 0.2 and infected with M13K07 helperphages at a multiplicity of infection of 10. After one hour incubation at 37° C. cells are harvested by centrifugation, resuspended in 250 ml glucose free medium and incubated over night at 30° C. and 200 rpm. Phage are isolated by PEG precipitation (PEG6000 20%, NaCl 2.5M) and redissolved in phage dilution buffer (Tris 10 mM pH 7.5, NaCl 20 mM, EDTA 2 mM).

Screening the Library for scFv's Binding Activated Platelets

Depletion of the Library for scFV's Binding Non Activated Platelets:

5 ml of human, venous blood are collected in a S MONOVETTE® (Sarstedt, blood taking apparatus) containing 25 μl prostaglandine E20 (10 mM) and centrifuged at 110 g for 10 min. Of platelet rich plasma (upper phase), 1 ml is transferred into a fresh tube, mixed with 9 ml CGS-buffer (sodium citrate 10 mM, dextrose 30 mM, NaCl 120 mM) and centrifuged at 1000 g for 10 min. The pellet is resuspended in 4 ml tyrode buffer (NaCl (150 mM), NaHCO3 (12 mM), KCl, MgCl (2 mM each), glucose, BSA (1 mg/ml each), pH 7.4) containing 2% skimmed milk powder and incubated with 1.75×10¹²bacteriophages (1000× complexity) for 2 hours at room temperature. Platelets are centrifuged at 1000 g for 10 min, the supernatant removed and stored at 4° C.,

Binding onto Activated Platelets:

5 ml of human, venous blood are collected in a S-MONOVETTE® (Sarstedt) and centrifuged at 110 g for 10 min. Of platelet rich plasma (upper phase), 1 ml is transferred in a fresh tube, mixed with 9 ml CGS-buffer and centrifuged at 1000 g for 10 min. The pellet is resuspended with 4 ml depleted phage solution containing CaCl₂, MgCl₂(2 mM each), ADP (15 μM) and incubated at room temperature for 2 hours. Platelets are washed twice by centrifugation (1000 g, 10 min) and resuspended in 14 ml tyrode buffer.

Elution:

For elution of binding phage, the platelets are centrifuged (1000 g, 10 min), resuspended in 1 ml glycine buffer (0.1 M, pH 2.2) and incubated for 10 min at room temperature. After centrifugation (1000 g, 10 min) the supernatant is neutralized by addition of Tris (2 M, pH 8.0).

Reinfection:

Eluted phages are mixed with 10 ml of logarithmic growing E. coli XL1 blue cells and incubated at 37° C. for 30 min. After centrifugation (10 min, 6000 g), cells are resuspended in 400 μl LB_GATmedium (LB medium containing 50 mM glucose, 100 μg/ml ampicillin and 20 μg/ml tetracyclin), plated on LB_GATagarplates and incubated over night at 37° C.

Packaging:

Colonies are scraped from agar plates using two times 5 ml LB_GATmedium and used for inoculation of 20 ml LB_GATmedium at an OD600 of 0.1. Cells are incubated at 37° C. and 200 rpm for one hour and superinfected with about 1×10¹⁰M13K07 helperphages. After one hour at 37° C., cells are collected by centrifugation (5 min, 6000 g) resuspended in LB medium supplemented with ampicillin (100 μg/ml) and kanamycin (50 μg/ml) and incubated over night at 30° C. and 200 rpm. Phages are collected by PEG precipitation and resuspended in 1 ml phage dilution buffer (as described for library rescue).

Screening the Library for scFv's Binding Recombinant GPIIb/IIIa on CHO-Cells

Depletion of scFv's Binding Non Activated GPIIb/IIIa:

Chinese hamster ovary cells (CHO) expressing non activated GPIIb/IIIa receptor (A5 cells; Peter et al., Blood, Vol. 9, 1998, pp. 3240-3249) are trypsinated, centrifuged (10 min, 140 g) and resuspended at 5×10⁶cells/ml in tyrode buffer. About 10⁹packaged phage from the first round of selection are mixed with 4 ml cell suspension and incubated for one hour at room temperature. Cells are centrifuged for 20 min at 140 g and the supernatant cleared again by centrifugation (20 min, 3200 g).

Binding on Activated GPIIb/IIIa:

CHO cells presenting active GPIIb/IIIa (Cl3 cells, Peter K and O'Toole T E, J Exp Med. 1995, 181(1): 315-326) are harvested by trypsination, centrifuged and washed once using 1 ml tyrode buffer. 4×10⁶cells are incubated with 4 ml depleted phage solution for 30 min at room temperature.

Elution by Antibody Competition:

Cells are centrifuged for 10 min at 140 g, resuspended in 50 ml tyrode buffer, three times centrifuged for 20 min at 700 g and resuspended in 1 ml tyrode buffer and finally resuspended in 200 μl REOPRO® (pharmaceutical preparations for the prevention and treatment of diseases of the circulatory system) (2 mg/ml). After 20 min at room temperature, cells are removed by 10 min centrifugation at 13000 rpm in a benchtop centrifuge.

Acidic Elution:

Cells are centrifuged for 10 min at 140 g, resuspended in 50 ml modified tyrode buffer (tyrode buffer pH 6 adjusted with Hepes, containing CaCl₂, MgCl₂(2 mM each) and 1 mg/ml BSA), twice centrifuged for 20 min at 700 g and resuspended in 1 ml modified tyrode buffer and finally resuspended in 1 ml glycine (pH 2.2). After 15 min at room temperature the mixture is neutralized by addition of 100 μl Tris (2 M, pH 8) and cleared by centrifugation at 13000 rpm for 10 min in a benchtop centrifuge.

Reinfection and packaging: is done as described above.

Restriction Endonuclease Digestion Analysis of Selected Clones

DNA of clones from selection experiments are prepared using DNA spin columns following the recommendations of the manufacture (Quiagen). DNA is digested with BstNI (New England Biolabs) and analysed on a 1% agarose gel.

Preparation of Periplasmic Extracts

5 ml of LB_GATmedium are inoculated with 250 μl of an overnight culture and incubated at 37° C. and 180 rpm for 4 hours. Cells are harvested by centrifugation (5 min, 6000 g) resuspended in 5 ml LB medium containing ampicillin (10 μg/ml) and IPTG (100 μM) and incubated at 28° C. and 180 rpm over night. Cells are again harvested by centrifugation and resuspended in 500 μl shock solution (50 mM Tris HCl pH 8.0, 20% saccharose, 1 mM EDTA) and incubated at 8° C. for one hour. Cells are removed by centrifugation (10 min, 13000 rpm benchtop centrifuge) and the supernatant dialysed two times 3 hours against PBS at 4° C.

FACS-Analysis

FACS-analysis is done using a FACSCALIBUR™ (automated, multicolor flow cytometry system) device (Becton Dickinson).

Analysis of Activation Specificity:

Complete citrate blood (S MONOVETTE®, Sarstedt) is diluted 1/50 in 50 μl tyrode buffer with or without ADP (20 μM) and incubated for 20 min at room temperature with 10 μl of periplasmic scFv extracts. As secondary antibody FITC labelled anti-His-antibody (Dianova) is added, incubated for 20 min and fixed with CELLFIX™ (immunomagnetic tools, namely, magnetic nanoparticles and devices for cell separation, manipulation, collection and presentation for analysis, for use in laboratory research tests) (1×).

Analysis of Fibrinogen Competition:

Complete citrate blood (S MONOVETTE®, Sarstedt) is diluted 1/50 in 50 μl tyrode buffer with or without ADP (20 μM) and incubated for 20 min with FITC labeled anti-fibrinogen-antibody (WAK-Chemie Medical) in presence or absence of 20 μl of periplasmic scFv extracts before fixation with CELLFIX™ (1×, Becton Dickinson).

Aggregometry Measurements:

Aggregometry was performed using a Biodata PAP-AGGREGOMETER™ (an instrument for simultaneous performance of a routine aggregation profile) according to the manufacturer's recommendations. After incubation with the scFv for 10 min at 37° C. the aggregation was induced by addition of 20 μM.

Results

Selection of GPIIb/IIIa Binding scFv's

Human scFv phagedisplay libraries originating from spleen and PBL are screened for GPIIb/IIIa specific clones by selection on activated human platelets for one round. The library is depleted before on not activated platelets to remove not activation-specific binders. The second and third round of selection is done on CHO cells expressing recombinant, activated GPIIb/IIIa receptor after depletion on cells presenting a not activated variant. Elution is done either by acid or by competition with REOPRO®. After the third round of selection clones are randomly picked and analysed first for enrichment by BstN1 digestion and activation-specific binding to thrombocytes (Table 1). One clone, MB9, is found to be enriched using acidic as well as competitive elution to 10 of 80 clones and 10 of 60 clones respectively. MB9 is also strongly activation-specific in platelet binding and inhibits binding of fibrinogen to platelets as shown by FACS-analysis depicted in FIG. 1. Therein, the following is depicted: Left histogram: demonstrates binding of MB9 scFv to activated (black) but not to unactivated (grey) human thrombocytes.

Right histogram: Binding of fibrinogen to activated (black) but not to unactivated thrombocytes. Binding of fibrinogen to activated thrombocytes is inhibited in presence of MB9 scFv (filled bright grey curve).

Additionally MB9 competes with REOPRO® for binding. Other enriched clones like MA1 also show activation of specific binding but fail in inhibition of fibrinogen binding or are not strongly specific for activated thrombocytes like MA3 or MB1.

The DNA sequence of clone MB9 is given in SEQ ID No. 1 (FIG. 2). Restriction endonuclease recognition sequences flanking heavy and light chains (NcoI, HindIII and MluI, NotI respectively) are indicated.

A clone encoding MB9 has been deposited under DSM 14491 (XL1blue (pEXHAM4/MP9)) on Sep. 6, 2001 with the “Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, D-38124 Braunschweig” under the Budapest Treaty.

TABLE 1

Characterization of scFv clones enriched on activated GPIIb/IIIa.

Enrichment
Activation

compe-

Identical
specific
inhibition
tition

clones/
binding to
of
by

elution
analysed
human
fibrinogen
REO-

Clone
done by
clones
platelets
binding
PRO ®

MA1
Acid
20/80
++
−
−

MA2
Acid
10/80
++
++
+

MA3
Acid
24/80
+
−
−

MB1
Competition
21/60
+
+
+

MB9
Competition
10/60
++
++
+

Identical

to MA2

++: strongly positive;

+: positive;

−: negative

For MB9 scFv inhibition of aggregation of activated human platelets in presence of fibrinogen in a concentration dependent manner has been demonstrated by aggregometry (FIG. 15a). By this method half maximal inhibition of platelet aggregation was achieved with MB9 scFv (31 kDa) at 25 μg/ml (806 nM) and with REOPRO® (Fab, 50 kDa) at 2.7 μg/ml (54 nM), whereas half maximal inhibition of fibrinogen binding in FACS was reached at 1.1 μg/ml (35 nM) for MB9 and 0.75 μg/ml (15 nM) for REOPRO® (FIG. 15b).

Example 2
Production of the Synthetic Human Framework Based scFv Antibody Fragment

Origin of Human scFv Master Frameworks

For the generation of a synthetic library by randomization of the CDR3 region of the heavy chain two human master frameworks (C9 and E4, FIG. 3) are chosen because of their excellent production characteristics in E. coli cells. Both scFv's originate from a large human phage display antibody library (Little, M., et al., J. Immunol. Methods 1999, 231: 3-9) and specific for hepatitis B virus antigen (C9) and estradiol (E4) respectively.

Vector Construction for the Synthetic scFv Library

C9 and E4 scFv's are cloned in pEXHAM1 vector DNA replacing the stuffer scFv by standard recombinant cloning techniques using NcoI and NotI cloning sites.

To prepare a vector allowing the randomization of CDR3 of the heavy chain without changes of the original sequence this region is replaced by a stuffer DNA fragment containing restriction enzyme recognition sites of the type IIS enzyme BbsI (BpiI). Standard PCR reactions are set up using the oligonucleotid primers as shown in FIG. 4 to generate DNA fragments of the scFv regions 3′ and 5′ of the heavy chain CDR3 containing unique BpiI cloning sites as outlined in FIG. 5, which is a schematic representation of annealing positions of oligonucleotides used for the construction of pEXHAM4/C9 and pEXHAM4/E4. Genes of the scFv's C9 and E4 cloned in pEXHAM1 are shown as boxes. Black painted areas represent CDR regions; Oligonucleotides are represented by arrows and identified by numbers (cp. sequence definitions). BpiI restriction endonuclease recognition sites are indicated.

The stuffer DNA fragment is generated directly by hybridisation of synthetic oligonucleotides. DNA-fragments are cut with BpiI and cloned in BpiI digested pEXHAM1 vector DNA to generate pEXHAM4/C9 and pEXHAM4/E4 (FIGS. 6, 11 and 12).

Direct use of pEXHAM4 vector DNA for cloning via BbsI necessitates the purification of two vector fragments, 3.8 and 0.5 kb in size. To avoid this both BbsI restriction sites outside the scFv sequence are removed in several steps without changing the protein sequence using mismatched oligonucleotides as primers for PCR or directly hybridised synthetic oligonucleotides to replace BbsI containing DNA-fragments by cloning via neighboring restriction sites. The final constructs is named pEXHAM7/C9 and pEXHAM7/E4 (FIGS. 7, 13 and 14) respectively.

Generation of the Synthetic, Human Framework Based scFv Library

To generate a library synthetic oligonucleotides encoding four to seven random amino acids by NNK codons (VHCDR3_—3.4/cut until VHCDR3_—3.7/cut; 1 μM each) are filled in separately using oligonucleotides VHCDR3_for/cut and VHCDR3_back/cut (0.2 μM) (FIG. 4) with 1 unit VENT® exo⁻DNA-polymerase (New England Biolabs) unter following PCR conditions: 2 min 94° C., 5×[1 min 94° C., 1 min 40° C., 1 min 72° C.] 10 min 72° C. in 100 μl Volume. PCR-products are purified using PCR purification Kit (Qiagen). ⅔ of the material is cut with 100 units BbsI for 6 hours and again purified via the mentioned kit. In case of VHCDR3_—3.4/cut and VHCDR3_—3.5/cut the vector DNA pEXHAM4/C9 and pEXHAM4/E4 is cut with BbsI (1 unit/μg in 6 hours) and both vector fragments (3.8 and 0.5 kb) are purified via gel elution from a 1 agarose gel (Gel Extraction Kit, Qiagen). For VHCDR3_—3.6/cut and VHCDR3_—3.7/cut pEXHAM6/C9 and pEXHAM6/E4 are used, therefore only one vector fragment had to be purified. Ligation is done in all cases at an equimolar ratio of all fragments. Afterwards the ligation mixture is precipitated, redissolved in Tris 10 mM, pH8.5 and used for transformation of XL1 blue cells essentially as described for example 1.

In addition to synthetic randomized DNA-fragments, CDR3 of the heavy chain is also replaced by natural CDR3 sequences amplified from the products of the first PCR of the natural library (see example 1) to focus on functional, in vivo used sequences for this region. Oligonucleotides used and described in FIG. 4 are designed to cover most of the human heavy chain CDR3 regions without modifying C9 or E4 framework sequences. PCR is done separately for each human VH PCR template using 1 unit VENT® exo⁻-DNA polymerase (New England Biolabs) and 0.2 μM primer in a volume of 100 μl under following conditions: 2 min 94° C., 30×[1 min 95° C., 1 min 50° C., 1 min 72° C.] 10 min 72° C. Oligo nucleotides #42, #43 and #44 are used as an equimolar mixture. PCR products are purified via PCR purification kit and material originating from spleen or PBL respectively is pooled. Restriction with BbsI, ligation with pEXHAM6/C9 and pEXHAM6/E4 respectively and transformation is done as described above.

The size of the whole synthetic library (synthetic and natural CDR3's cloned in C9 or E4 frameworks) in this example is 7.5×10⁸clones.

Library Rescue

Packaging of synthetic libraries is done as described for the natural library (example 1).

Screening of the Synthetic Library

Screening of the synthetic library is done exactly as described for the natural library (example 1) starting with 7.5×10¹¹bacteriophages (1000× complexity).

Results

The synthetic library derived from human scFv frameworks (C9 and E4) is screened for GPIIb/IIIa specific clones exactly as described in example 1. After the third round of selection clones are randomly picked and the DNA sequence of the VH-CDR3 regions was determined (c.f. Table 2)

TABLE 2

Analysis of the DNA-Sequence of VH-CDR3 of eleven GPIIb/IIIa selected

clones from the synthetic library (SEQ ID Nos: 162-169).

No of

Translation of VH-CDR3
identical
Oligonucleotide

clone
DNA
clones
used for CDR3

SA1
CAR RYRVG FDY
1
VHCDR3_3.5/cut

(SEQ ID NO: 162)

SA2
CAR GATYTSRSDVPDQTS FDY
2
VHCDR3_ev2/for/cut

(SEQ ID NO: 163)

SA3
CAR DDLAYCRGDCSGRFA FDI
2
VHCDR3_ev2/for/cut

(SEQ ID NO: 164)

SA4
CAR RFSISRA FDY
1
VHCDR3_3.7/cut

(SEQ ID NO: 165)

SA6
CAR RWGKARS FDY
1
VHCDR3_3.7/cut

(SEQ ID NO: 166)

SA8
CAK ELEAYCRGDCYPPYYG MDV
1
VHCDR3_ev3/for/cut

(SEQ ID NO: 167)

SA10
CAR DLFRGRGDYGDYG MDV
1
VHCDR3_ev2/for/cut

(SEQ ID NO: 168)

SA11
CAR TYYYDSRTDRRPPHA FDI
1
VHCDR3_ev3/for/cut

(SEQ ID NO: 169)

All of the clones use the E4 framework sequence. Three of the eleven analysed clones encode the amino acid sequence RGD (also present in fibrogen) within CDR3 (SA3, SA8 and SA10). In clones SA3 and SA8 the RGD motive is directly flanked by two cysteine residues that might stabilize the loop by disulfide bridges. Clone SA3 was found twice under eleven analysed clones and, therefore, has probably enriched by the screening procedure. The same is true for clone SA11. These scFv clones are similar to antibodies like PAC-1 that contain RGD-like sequences and inhibit fibrogen binding by blocking the activated receptor (Shatill et al., 1985). Only SA8, Sa10 and SA11 showed an activation specific binding to thrombocytes in the presence of fibrinogen (c.f. FIG. 10).

The selected clones probably interact exactly with the fibrinogen binding site of the GPIIb/IIa receptor but with an affinity similar or lower than fibrinogen. The affinity has been enhanced by mutation within the VH and/or the VL-domain of the scFv antibody fragment or the exchange of the whole VL domain (“chain shuffling).

Example 3
Improvement of GPIIb/IIIa Specific Synthetic scFv Fragments by Light Chain Shuffling

Vector Construction:

CDR3 sequences of the variable domain of the light chain of C9 and E4 scFv in pEXHAM1 were replaced by a synthetic stuffer introducing flanking BbsI restriction sites as outlined in FIG. 16. Standard PCR reactions were set up using the oligonucleotides shown in FIG. 17 to amplify the DNA-Fragments of the scFv regions in 3′ and 5′ direction of the light chain CDR3. PCR-fragments were purified cleaved with BbsI. The stuffer CDR3 fragment was generated directly by hybridisation of the indicated oligos. All three fragments were ligated with BbsI digested pEXHAM1 vector-DNA to generate pEXHAM9/C9 and pEXHAM9/E4 respectively. To remove additional BbsI sites the light chain fragments of pEXHAM9/C9 and pEXHAM9/E4 were recloned in pEXHAM6/C9 and pEXHAM6/E4 using MluI and NotI sites to generate pEXHAM10/C9 and pEXHAM10/E4. To facilitate the cloning of randomized VLCDR3 sequences, a 1.5 kb DNA fragment amplified from bacteriophage lambda DNA (bp 15629-17152) using standard PCR reactions and primer LAM1 and LAM2 (FIG. 17) was introduced via SacI between BbsI sites in the stuffer region of pEXHAM10 derivatives to create pEXHAM11(C9) and pEXHAM11/(E4).

Generation of a Synthetic, Human Framework scFv Based Light Chain Library

To generate a VL library, synthetic oligos encoding four to six random aminoacids by NNK-codons (C9VLCDR3_—4/cut until C9VLCDR3_—6/cut and E4VLCDR3_—4/cut until E4VLCDR3_—6/cut, FIG. 17) were filled in separately essentially as described in example 2 using oligos C9VLCDR3_for/cut and C9VLCDR3_back/cut or E4VLCDR3_for/cut and E4VLCDR3_back/cut (FIG. 17). PCR-products were cut with BbsI and ligated with BbsI digested vector DNA fragments of pEXHAM11/C9 and pEXHAM11/E4 respectively and used for transformation of E. coli XL1blue cells essentially as described for example 1.

In addition to the synthetic randomised DNA-fragments, CDR3 of the light chain was replaced also by natural VLCDR3-sequences. Vλ genes amplified in the first PCR on human cDNA from PBL and spleen as described in example 1 were used as a template to amplify VLCDR3 essentially as described in example 2 using oligos VLCDR3_ev/for/cut and VLCDR3_ev/back/cut (FIG. 17). PCR-products were cloned after BbsI digestion in pEXHAM11/C9 and pEXHAM11/E4 and used for the transformation of E. coli XL1blue as described above.

The overall size of the VL-library is 3.6×10⁷clones for the C9 framework and 4.7×10⁷for the E4 framework.

Generation of GPIIb/IIIa Specific scFv-Sublibaries by Chain Shuffling

ScFv clones SA2, SA3, SA8, SA10 and SA11 described in example 2 were selected for chain shuffling. VL genes of these clones were replaced via MluI and NotI sites by randomised VL genes from the E4 VL-library using standard cloning procedures. By repeated transformations of E. coli XL1blue separate sublibraries of 2.6×10⁷until 6.5×10⁷clones were generated.

Library Rescue

Packaging of the sublibraries was done as described in example 1.

Screening of GPIIb/IIIa Specific Sublibraries

Five rounds of panning were performed on activated GPIIb/IIIa-expressing CHO-cells essentially as described for example 1, but using only 105 CHO cells. During the first four rounds bound phages were eluted by low pH and in the fifth round by increasing concentrations of Eptifibatide (0.1-1000 μg/ml). Phages eluted with 100 μg/ml were further investigated.

Results

Five scFv clones isolated from the synthetic library and showing an RXD-motif in CDR3 of the heavy chain (SA2, SA3, SA8, SA10 and SA11, see example 2) were selected for affinity maturation by chain shuffling. For each clone the constant light chain domains were replaced by an E4 light chain library with randomised synthetic or natural CDR3 regions. These sublibraries were screened again on GPIIb/IIIa presenting CHO-cells using acidic elution for four rounds of panning. In the fifth round elution was done by increasing concentrations of Eptifibatide, a low molecular weight GPIIb/IIIa inhibitor (RGD-mimetic), to select clones with increased affinity. Single clones were analysed first by FACS for binding to activated platelets. Most of the clones showed increased but nonuniform binding to activated platelets. By subsequent DNA-sequencing no strong enrichment of single clones was observed. Two clones, SCE5 and SCE18, were identified that stained all activated platelets in an uniform manner (FIG. 18). Both clones are derived from the SA2 framework and have similar light chain CDR3 sequences (SCE5: CLLYYGGGQQGVFGGG, SEQ ID NO: 170; SCE18: CLLYYGGAWVFGGG, SEQ ID NO:171).

Example 4
Converting GPIIb/IIIa Specific scFv in Other Formats

MB9 scFv was converted into different recombinant antibody formats to generate variants possibly improved e.g. in terms of size, stability or affinity.

MB9 Diabody

To increase the size of MB9, a diabody can be generated for example by reducing the linker between VH and VL to generate a molecule that is not able to form a functional scFv but a non-covalently linked homodimer bearing two antigen binding sites. Such a diabody was generated by shortening the original linker to three amino acids (FIG. 19) using standard PCR methods. Absence of scFv monomers and presence of dimeric (and multimeric) forms has been demonstrated by size exclusion chromatography for this MB9 derivative as well as inhibition of fibrinogen binding to activated platelets.

MB9 Fab

To convert MB9 scFv into a Fab format the variable domains of the heavy and light chains were amplified separately by PCR and cloned into the Fab-Vector pREFAB9 (FIG. 20) in front of CH1 and Clambda respectively. The sequence is shown in FIG. 22. FACS-analysis of periplasmic preparations of the MB9 Fab demonstrated specific binding of MB9 Fab to activated human thrombocytes (FIG. 21).

Number	Name	Date	Kind
6265150	Terstappen et al.	Jul 2001	B1
6790938	Berchtold et al.	Sep 2004	B1
6955900	Barbas et al.	Oct 2005	B1
20030219434	Carter et al.	Nov 2003	A1

Antibody of human origin specifically binding to activated state of platelet integrin receptor GPIIb/IIIa

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Term Extension

Abstract

Description

Claims

Priority Claims (1)

PCT Information

US Referenced Citations (4)

Foreign Referenced Citations (1)

Related Publications (1)