COLLAGEN PEPTIDES, METHODS AND USES

The present invention relates to peptides, in particular peptides based on collagen sequences, useful in modulating platelet and other cell function, including aggregation and activation, also to methods of use of the peptides and compositions comprising them.

More specifically, the present invention is based on the inventors' findings in relation to collagenous peptides whose activity is sufficient without polymerisation or cross-linking to induce platelet aggregation.

Human collagens comprise a family of 28 or more proteins each of which is characterised by the presence of a triple-helical structure, formed as three separate left-handed helical protein strands, sometimes different gene products, that wind around one another to form a right-handed superhelix. This defining conformation is facilitated by a repetitive [glycine-x-x ' J _n structure, where n may be as large as 350, so that a triple helical domain (col domain) may be in excess of 1000 amino acids in length. In nature, the amino acids x and x' are quite often proline and hydroxyproline, which occur in quite high proportion within col domains and are essential for the formation of the triple-helical structure which is known as the polyproline helix. The collagens are widely distributed within the vertebrate organism where they perform an essential structural role. Examples of this are provided by the most abundant fibrillar collagens, types I, II and III, which occur in skin, bone, cartilage, tendon and in the vitreous humour of the eye. More subtle roles are played by the more complex non- fibrillar collagens, such as types IV and VI, which form two- and three-dimensional networks, supporting the interstitial

tissues of the body and being the fundamental component of the basement membranes to which epithelial and endothelial cell layers can attach.

Circulating platelets, also called thrombocytes, within the bloodstream are the crucial cellular components of blood which regulate the clotting process. In healthy, undamaged tissues, collagens which support the blood vessel wall and surrounding tissue are concealed by endothelial cell layers and cannot come into contact with the circulating platelets. However, should the endothelial cell layer be removed either in disease or upon tissue injury, then collagens are revealed which can interact with the cellular components of the blood as well as with proteins in blood plasma. The platelet surface contains a series of proteins known as receptors which sense the presence in the extracellular medium of specific molecules, including hormones, cytokines and other species. Collagen can bind directly to several such receptors on the platelet surface, notably integrin α2βl and Glycoprotein VI (GpVI). Indirect interactions can also occur, as when von Willebrand Factor (VWF) binds to the platelet surface through the Glycoprotein Ib/V/IX complex, interacting directly with Gplbα, and also to specific sites within the collagens. Thus a series of platelet receptors contribute to the interaction of the platelet with the collagens. See Farndale et al., reference 1, for a recent review.

The platelet regulates the blood clotting process in a complex series of interacting processes. First, the platelet must be captured from the circulation, a process that may vary with the shear stress experienced at the site of injury. VWF is required for this initial interaction. Next, the platelet must be secured at the exposed collagen surface, and the integrin

α2βl is considered to be crucial for this process. Finally, the platelet must be activated, and GpVI is considered the primary activatory collagen receptor expressed on the platelet surface. There remains debate about the exact role of these individual interactions in the capture and activation of platelets. Once the platelet becomes activated, a new series of events is set in motion, with further activatory materials such as ADP and ATP being secreted as part of the platelet dense granules' content, and thromboxane A ₂ being generated from endogenous arachidonic acid. Three definable endpoints characterise the platelet activation process:

First, the activation of integrins may allow tighter binding of integrin α2βl to collagens but most importantly increases the affinity of the fibrinogen receptor integrin αllbβ3 which allows plasma fibrinogen molecules to cross-link two copies of αllbβ3 on the surfaces of adjacent platelets, the fundamental interaction in platelet aggregation, or thrombus formation.

Next, the platelet surface may become procoagulant, that is, the distribution of phospholipids between its inner and outer leaflets changes such that negatively charged phospholipids, phosphatidylserine and phosphatidylethanolamine, are present in greater quantities upon the outer surface of the platelet where they act as a catalytic surface for the coagulation cascade and the generation of thrombin. This thrombin causes the proteolytic cleavage of plasma fibrinogen such that fibrin is formed which polymerises and clots, trapping local red cells and leading to the occlusion of breaches in the damaged vessel wall. These processes are central to the normal haemostatic response, and are well-understood in the field.

Finally, the activated platelet secretes, primarily from its α- granules, bioactive materials such as platelet-derived growth factor which may be important in stimulating cells locally to repair the damage to the blood vessel wall.

Each of these processes may also be important in disease, where atherosclerosis, the formation of plaques which constrict the blood vessel, progresses to the point where weak, disorganised, lipid-rich tissue may be prone to fissuring or rupture which reveals underlying collagens and collagen fragments causing the activation of platelets and the clotting process. This set of pathological events is known as atherothrombosis, and is life- threatening when it occurs in vital blood vessels such as the coronary artery, causing heart attack, or in the cerebral vasculature where it may cause stroke. Detailed reviews of these various topics can be found in references 2-4.

The recognition of collagen by the platelet receptors described above depends upon its primary sequence. The collagens must contain specific combinations of amino acids for recognition to occur. However, denatured collagens (gelatins) exist as single random coil polypeptides, which, despite sequence being preserved, will not bind to these receptors (reference 5) . Thus, triple-helical conformation of the collagens is also an essential pre-requisite for their recognition by platelet and other collagen receptors, and the use of synthetic triple- helical peptides comprising specific recognition motifs has allowed receptor-binding properties of the collagens to be investigated in detail (references 6 and 7). This strategy has been applied successfully to both α2βl and GpVI .

For integrin α2βl, some recognition sequences within collagens are well-known, such as the GFOGER motif which occurs in

collagens I, II and IV and is a high-affinity ligand that binds well even when the integrin is in a resting state. Other similar sequences that also bind α2βl especially when the receptor is activated include GLOGER, GMOGER, GLSGER, GASGER and GAOGER (references 8-11) . Such sequences are quite widely- distributed within the triple-helical col domains, especially of the fibrillar collagens. Related sequences that also bind integrin include GROGER found in collagen III, unusual in that it contains a positively-charged arginine rather than a hydrophobic residue in the second position of the motif.

For GpVI, however, no clear definition exists. Certain sequences are known to interact with GpVI, and these include the triplet GPO. The affinity of such triple-helical motifs increases with the number of adjacent GPO triplets, such that [GPO] ₂ binds detectably to GpVI, and [GPO] ₄ binds almost as well as the longest such peptide hitherto described, the collagen- related peptide (CRP) which contains a [GPO] ₁₀ triple-helical core (reference 12) . CRP is widely used in the field as a research tool, being specific for GpVI (references 13 and 14). Although CRP is considered specific for GpVI, it is not known whether the GPO motifs which form about 10% of the fibrillar collagen sequence are uniquely responsible for interaction with GpVI, or whether other motifs within collagens which lack the GPO triplet can also be recognised by the receptor.

Triple-helical collagens can be extracted from tissues using dilute acid, e.g. acetic acid, or by pepsin digestion which removes the telopeptides at each end of the col domain of the fibrillar collagens, allowing ready disassembly of the collagen fibre. Such collagen preparations, being triple-helical but known as monomeric collagens, will interact with the platelet surface and support platelet adhesion, but do not activate

platelets (reference 5) . Similarly, triple-helical collagen fragments such as those produced by digestion of collagens with cyanogen bromide may support platelet and other cell adhesion, but again do not activate platelets. However, when collagen monomers are allowed to assemble as fibres, or when collagen fragments are chemically cross-linked, their platelet- activatory properties are restored (reference 15) . Thus, there is a requirement for polymeric higher-order structure over and above the need for triple-helical conformation if a collagenous material is to activate platelets.

This illuminates the underlying mechanism by which collagen activates the platelet, which is presumed to require the ability to induce receptor clustering on the surface of the platelet as well as mere receptor occupancy. In this respect, collagen differs from many other signalling molecules, such as ADP or adrenaline, where the ligand binds to its receptor and transmits a conformational change in the receptor across the cell membrane which can be interpreted as a signal by the intracellular metabolic machinery.

The need for cross-linking is readily seen with CRP. The original reports used two strategies to introduce polymeric structure, either the inclusion at each end of the peptide of lysine residues, whose free ε-amino groups were cross-linked using glutaraldehyde, or by the inclusion of cysteine residues, which could be cross-linked to the free N-terminus of the peptides using the bi-functional reagent, SPDP (reference 16) . The monomeric, triple-helical CRP did not induce platelet aggregation, whereas the cross-linked peptides were extremely potent, inducing full platelet aggregation at as low a concentration as 50ng/ml. Monomeric CRP has been used successfully as an antagonist of GpVI, blocking platelet

activation by collagen, supporting the concept that receptor clustering by a polymeric ligand is needed for full and efficient platelet aggregation.

In contrast, some limited activatory processes, e.g. the onset of the phosphorylation of protein tyrosine residues, have been observed using non-cross-linked CRP (references 12 and 17) . At high dose, about 100 μg/ml, i.e. 1000 times higher dose than necessary for the cross-linked peptide, platelet aggregation was reported, although in this laboratory, such activity was absent. It may be deduced that the recognition by GpVI of CRP does not require its polymerisation, although as emphasised above, its activatory properties are critically dependent on polymeric structure, with the monomeric peptide being below the threshold of activity needed to induce full platelet activation .

The binding site for collagen on α2βl resides within the I- domain of its a. subunit, and has been well-characterised by site-directed mutagenesis and by co-crystallisation with a triple-helical peptide (references 9 and 18). Similarly, site- directed mutagenesis of recombinant GpVI has shown that collagen binds to the hinge region of GpVI, the apex between its two immunoglobulin folds (reference 19) . VWF bound to GpIb/V/IX, forms an important collagen-binding complex that acts as a collagen receptor, and each of its Al and A3 domains has been implicated in the recognition of collagen, although the role of Al is primarily to bind GpIb. Site-directed mutagenesis has revealed a set of residues within A3 that are important in binding collagen (reference 20) . The corresponding motif within collagen III has recently been identified (Lisman et al, Blood Online) .

The present invention provides collagenous peptides whose activity is sufficient without polymerisation to induce platelet aggregation.

5 Brief Description of the Figures

Figure IA shows the aggregation response of platelets to the CRPs indicated. Aggregation was measured as described in Materials and Methods using the Chronolog instrument for up to 10 15 min. The maximum aggregation response to increasing concentrations of peptide is shown. Data are the mean of two separate experiments.

Figure IB shows extent of platelet aggregation and thus potency 15 of the peptides in inducing platelet aggregation in platelet rich plasma (PRP) . Experiment performed as for Figure IA.

Figure 1C shows similar results as Figure IB, using washed platelets .

20

Figure 2 shows the adhesion of platelets to 96-well plates coated with the indicated peptides, each used at a saturating concentration of 20μg/ml. Adhesion was measured after 1 hour incubation of platelets by the colorimetric method described in

25 Materials and Methods. Platelets were pre-incubated as indicated with the anti-GpVI scFvs at a concentration of lOμg/ml.

Figure 3 shows the inhibition of CRP24-induced aggregation 30 caused by pre-incubation of platelets with scFvs, 10B12 and 1C3. Aggregation was measured as in Figure 1, and antibodies were used at lOμg/ml as in Figure 2.

Figure 4 shows the differential response of GpVI haplotypes to increasing levels of CRP18, measured in aggregation assays using the BioData PAP4 analyser. Pooled results are shown from 68 aa donors, 28 ab donors and 3 bb donors, as mean response ± SD.

Figure 5 shows Western blots that were prepared as described in Materials and Methods and probed for phosphotyrosine . The peptide used is shown over each blot and its concentration (μg/ml) over each lane. The positions of molecular weight markers are shown to the right of each blot (kDa) . Data show one of three similar experiments using aa donor blood.

Figure 6 show Western blots that were prepared as in Figure 5 were quantitated by video densitometry. The 75 kDa band indicated in Figure 5 was measured, being one of the most responsive to CRP and collagen, and containing the kinase p72syk. Where indicated, platelets were pre-incubated with either scFv 10B12 or 1C3, and data shown are from one of two similar experiments.

Figure 7 shows the adhesion of platelets to 96-well plates coated with the indicated peptides, CRP18 or CRP-9GFO9, each used at a saturating concentration of 20μg/ml, or BSA as control. Adhesion was measured after 1 hour incubation of platelets by the colorimetric method described in Materials and Methods. Platelets were pre-incubated as indicated with the anti-GpVI scFvs at a concentration of lOμg/ml.

Figure 8 shows the adhesion of platelets to 96-well plates coated with CRP-9GFO9 or BSA as control, as in Figure 7. Adhesion was measured after 1 hour incubation of platelets by the colorimetric method described in Materials and Methods.

Platelets were pre-incubated as indicated with the anti-GpVI scFvs at a concentration of lOμg/ml, either alone or in combination with the anti-integrin α2 subunit monoclonal antibody, 6Fl, also used at lOμg/ml.

Figure 9 shows the differential response of GpVI haplotypes to increasing levels of CRP18 and CRP-9GFO9, measured in aggregation assays using the BioData PAP4 analyser as in Figure 4. Pooled results are shown from 42 aa donors and 13 ab donors, as mean response ± SD.

Figure 10 shows the ability of antibodies 6Fl, 10B12 and 1C3 to inhibit aggregation caused by increasing levels of CRP-9GFO9, measured in aggregation assays using the Chronolog aggregometer as in Figure 1. Data are pooled from two experiments each using aa donor platelets.

Figure 11A and Figure HB show the results of when washed platelet suspensions at 37°C were treated with CRP-9GFO9 for the indicated times, in the presence of anti-integrin alpha2 subunit antibody, 6Fl (Figure HA and Figure HB lane 8), or anti-glycoprotein VI scFv, 10B12 (Figure HB, lanes 1-8), as indicated. Blocking antibodies were used at 5 and lOμg/ml respectively. The reaction was terminated by the addition of an equal volume of Laemmli ' s buffer, and proteins were separated by SDS-polyacrylamide gel electrophoresis, then transferred to nitrocellulose membrane. Phosphotyrosine- containing proteins were detected using the monoclonal antibody 4G10 (Upstate Biotechnology, Inc) according to the supplier's instructions, and data were captured on X-ray film using enhanced chemiluminescence .

Figure 12A shows intracellular calcium levels at various time points in the following experiment. Washed platelets were loaded with the intracellular calcium indicator, Fura2, centrifuged and resuspended, as described in Materials and Methods. Peptides CRP-XL or CRP-24 were added to stirred cuvettes (2ml) at the time indicated by the arrow, and the intracellular calcium levels shown were calculated from the ratio of fluorescence at 500nm using 340nm and 380nm excitation wavelength, as described in Materials and Methods.

Figure 12B shows the results of determination of intracellular calcium levels as for Figure 12A in platelet suspensions treated with the indicated level of CRP-18, CRP-9GFO9, or CRP- XL.

Figure 13 shows results of the following experiment. Platelet suspensions prepared as described in Materials and Methods were treated with CRP-18 or CRP-9GFO9 at 16μg/ml or CRP-XL at 0.5 or 10 μg/ml as indicated. Reaction was stopped by the addition of lysis buffer. Phospholipase C gamma2 was immunoprecipitated as described in Materials and Methods, and phosphorylation of its tyrosine residues was detected in western blots as in the experiments for which results are shown in Figure HA and Figure HB. The position of phospholipase C gamma2 is indicated by the arrow. Lane 1 contains the product of unstimulated platelets, lanes 2 to 5 the immunoprecipitate from peptide-treated platelets as indicated, lane 6, from a reaction from which the immunoprecipitating antibody was omitted, lane 7, an immunoprecipitate using a non-immune Ig fraction, lane 8, from an immunoprecipitate to which platelets were omitted, and lane 9, from a stimulated platelet lysate before immunoprecipitation .

W 2

12

In accordance with one aspect of the present invention there is provided a peptide containing repeats of Glycine-Proline- Hydroxyproline (GPO), wherein the peptide is of the structure:

(GPO) _(n/2 ) (GFOGER) (GPO) _(n/2) G or (GPO) _n G

wherein n is 10-32.

The value of "n" may be 11-32, 14-32 or 16-32.

The value of "n" is preferably at least 14, more preferably at least 16, and it may be 16-24. A preferred value for n is 18. N may be 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 or 32 in accordance with various embodiments of the present invention, provided that for peptides of the structure (GPO) _(n/2) (GFOGER) (GPO) _{n/2 )G n is an even number.

A preferred peptide according to an embodiment of the invention is of the structure:

(GPO) ₉ (GFOGER) ₉ G

Another preferred peptide according to an embodiment of the invention is of the structure:

(GPO) ₂₄ G

As noted, a peptide of the invention is preferably of the structure :

(GPO) _(n/2) (GFOGER) (GPO) _{n/2 ,G

or (GPO) _n G wherein n is 10-32.

However, peptides according to embodiments of the invention are also provided in which n is at least 6, e.g. 6, 7, 8 or 9.

Peptides according to embodiments of the invention are also provided in which n is greater than 32, especially where the peptides are provided by recombinant production of a proline precursor with conversion of proline residues to hydroxy-proline as discussed further herein. In further embodiments of the invention the upper limit of the range of n is greater than 32, e.g. 33-40, 33-50, 33-60.

As discussed further, a peptide of the invention may be provided within a fusion peptide or polypeptide, such that the peptide is fused to other amino acids. The overall length of a peptide fusion in accordance with embodiments of the invention may be or be about 100 amino acids, 120 amino acids, 130 amino acids, 140 amino acids, 150 amino acids or less than 120, 130, 140 or 150 amino acids.

Non-naturally occurring peptides and polypeptides fusions including such a peptide are also provided as aspects of the present invention, particularly wherein the peptide is fused to one or more non-collagen sequences .

Thus, for example, the invention also includes derivatives, fusions and conjugates of the peptides, including the peptide linked to a coupling partner, e.g. an effector molecule, a label, a marker, a drug, a toxin and/or a carrier or transport molecule, and/or a targeting molecule such as an antibody or binding fragment thereof or other ligand. Techniques for

coupling the peptides of the invention to both peptidyl and non-peptidyl coupling partners are well known in the art.

A further aspect of the invention provides a peptide or peptide timer, as disclosed, attached or coated on to a solid surface, such as an inert polymer, a 96-well plate, other device or apparatus used in a clinical or investigative context.

Preferably, a peptide according to aspects and embodiments of the invention is able to bind GPVI, especially when in a peptide trimer. Preferably, a peptide according to aspects and embodiments of the invention is able to induce platelet activation and/or aggregation, especially when in a peptide trimer.

Further peptides according to various aspects and embodiments of the present invention are able to activate platelets through the binding of GpVI and/or integrin α2βl, or other platelet receptor or combinations of these, by incorporating triple helical motifs recognised by more than one collagen receptor into the peptide. Thus, for example, activity of native collagen fibres may be reconstituted using a peptide of the invention.

An additional sequence or motif may be incorporate, such as any one or more of a recognition sequence for von Willebrand Factor (VWF), one or more fibrinogen fragments, other integrin ligands such as fibronectin and fragments thereof, RGD peptides and so on. Indeed, the invention provides for any desired additional peptide to be included in a fusion with a peptide of the invention, including non-triple helical extensions of the triple helix formed by trimerising of the peptides.

A peptide of the invention will form a trimer under appropriate conditions .

This provides for trimers of the peptide which are able to activate platelets and/or induce platelet aggregation.

Peptides of the invention need not include any cross-linking, e.g. hexanoic acid cross-linking for trimerization (such as the lysyl-lysyl amino hexanoate cross-linking) . Indeed, it is an advantage of embodiments of the present invention that the peptides form trimers without cross-linking, and trimers consisting of peptides according to embodiments of the invention are provided without cross-linking. Furthermore, the invention shows that trimers formed from peptides of the invention without cross-linking are active.

A peptide in accordance with an aspect of the present invention may include one or more heterologous amino acids joined to the peptide structure set out herein. By "heterologous" is meant not conforming with the formula of the peptides of the invention as set out herein, and not occurring in any collagen joined by a peptide bond without intervening amino acids to a peptide of the invention, that is to say usually a chain of amino acids which is not found naturally joined to any peptide of the invention at the position of fusion in the peptide of the invention. Usually where heterologous amino acids are included, the contiguous sequence of amino acids does not occur within collagen, and may be 5 or more, preferably 10 or more, more preferably 15 or more, 20 or more or 30 or more amino acids with a sequence which does not occur contiguously in collagen. Other potential fusion partners have been mentioned already above.

Peptide trimers according to different aspects of the present invention may be homotrimeric as described above, where all three chains have identical sequence, but may also be heterotrimeric in nature, that is at least one of the chains is different from the others, and two but not three chains may be identical in sequence.

In a further aspect the present invention provides peptide trimers including one or more peptides according to the present invention, preferably consisting of three peptides according to the invention.

A still further aspect provides a method of making a peptide trimer, the method including providing peptides of the invention and causing or allowing (under appropriate conditions) the peptides to associate to form a trimer.

Trimerization may be followed by isolation of trimers, e.g. for subsequent use and/or manipulation.

Peptides, particular in trimerized form, in accordance with aspects of the present invention, may be used in influencing cell adhesion to collagen and/or other cell types, particularly adhesion of platelets, smooth muscle cells, tumour cells or vascular endothelial cells. They may be used to affect activation of cells, such as platelets. This may be in a therapeutic context, e.g. to induce platelet activation and/or aggregation to prevent bleeding through injury, or in vitro for diagnostic purposes to identify dysfunction of the collagen receptor pathways. Peptides comprising different receptor- binding activity may discriminate between different platelet dysfunctions according to the receptor involved.

Peptides may be generated wholly or partly by chemical synthesis. The compounds of the present invention can be readily prepared according to well-established, standard liquid or, preferably, solid-phase peptide synthesis methods, general descriptions of which are broadly available (see, for example, in J. M. Stewart and J. D. Young, Solid Phase Peptide Synthesis, 2nd edition, Pierce Chemical Company, Rockford, Illinois (1984), in M. Bodanzsky and A. Bodanzsky, The Practice of Peptide Synthesis, Springer Verlag, New York (1984); in J. H. Jones, The Chemical Synthesis of Peptides. Oxford University Press, Oxford 1991; in Applied Biosystems 430A Users Manual, ABI Inc., Foster City, California , in G. A. Grant, (Ed.) Synthetic Peptides, A User's Guide. W. H. Freeman & Co., New York 1992, E. Atherton and R. C. Sheppard, Solid Phase Peptide Synthesis, A Practical Approach. IRL Press 1989 and in G. B. Fields, (Ed.) Solid-Phase Peptide Synthesis (Methods in Enzymology Vol. 289). Academic Press, New York and London 1997), or they may be prepared in solution, by the liquid phase method or by any combination of solid-phase, liquid phase and solution chemistry, e.g. by first completing the respective peptide portion and then, if desired and appropriate, after removal of any protecting groups being present, by introduction of the residue X by reaction of the respective carbonic or sulfonic acid or a reactive derivative thereof.

Another convenient way of producing a peptidyl molecule according to the present invention (peptide or polypeptide) is to express nucleic acid encoding a precursor wherein proline appears in place of the desired hydroxyproline, by use of nucleic acid in an expression system. Production of GPO- containing peptides may be achieved for example by co- expression of an appropriate hydroxylase, as has been done with lysyl residues (Nokelainen et al. , 1998). For peptides

containing Pro residues to be post-translationally converted by hydroxylation to Hyp (O) , prolyl-hydroxylase may be co- expressed.

Myllyharju, J. et al. Biochem Soc trans 2000, 4 353-7 describes an efficient expression system for recombinant human collagens useful in practice of embodiments of the present invention. This system uses the methylotrophic yeast Pichia pastoris, with co-expression of the desired peptides chains with the alpha- and beta-subunits of prolyl 4-hydroxylase .

Accordingly the present invention also provides in various aspects nucleic acid encoding a proline precursor of a peptide or polypeptide of the invention.

Generally, nucleic acid according to the present invention is provided as an isolate, in isolated and/or purified form, except possibly one or more regulatory sequence (s) for expression. Nucleic acid in accordance with the present invention may be provided as part of a recombinant vector.

Nucleic acid sequences encoding a polypeptide or peptide precursor in accordance with the present invention can be readily prepared by the skilled person using the information and references contained herein and techniques known in the art (for example, see Molecular Cloning: a Laboratory Manual: 3rd edition, Sambrook and Russell, 2001, Cold Spring Harbor Laboratory Press) .

In order to obtain expression of the nucleic acid sequences, the sequences can be incorporated in a vector having one or more control sequences operably linked to the nucleic acid to control its expression. The vectors may include other sequences such as promoters or enhancers to drive the

expression of the inserted nucleic acid, nucleic acid sequences so that the polypeptide or peptide is produced as a fusion and/or nucleic acid encoding secretion signals so that the polypeptide produced in the host cell is secreted from the cell. Polypeptide can then be obtained by transforming the vectors into host cells in which the vector is functional, culturing the host cells so that the polypeptide is produced and recovering the polypeptide from the host cells or the surrounding medium. Prokaryotic and eukaryotic cells are used for this purpose in the art, including strains of E. coli, yeast, and eukaryotic cells such as COS or CHO cells.

Thus, the present invention also encompasses a method of making a polypeptide or peptide (as disclosed) , the method including expression from nucleic acid encoding the polypeptide or peptide (generally nucleic acid according to the invention) . This may conveniently be achieved by growing a host cell in culture, containing such a vector, under appropriate conditions which cause or allow expression of the polypeptide. Polypeptides and peptides may also be expressed in in vitro systems, such as reticulocyte lysate. Following production by recombinant expression, proline is converted to hydroxyproline . As noted, this may be achieved within the expression system by provision of a prolyl-hydroxylase, or by enzymatic treatment following production.

As noted, methods of making peptides by chemical synthesis are also encompassed by the present invention.

A further aspect of the present invention provides a host cell containing heterologous nucleic acid as disclosed herein. The nucleic acid of the invention may be integrated into the genome (e.g. chromosome) of the host cell. Integration may be

promoted by inclusion of sequences which promote recombination with the genome, in accordance with standard techniques. The nucleic acid may be on an extra-chromosomal vector within the cell, or otherwise identifiably heterologous or foreign to the cell.

A still further aspect provides a method which includes introducing the nucleic acid into a host cell . The introduction, which may (particularly for in vitro introduction) be generally referred to without limitation as "transformation", may employ any available technique. For eukaryotic cells, suitable techniques may include calcium phosphate transfection, DEAE-Dextran, electroporation, liposome-mediated transfection and transduction using retrovirus or other virus, e.g. vaccinia or, for insect cells, baculovirus . For bacterial cells, suitable techniques may include calcium chloride transformation, electroporation and transfection using bacteriophage. As an alternative, direct injection of the nucleic acid could be employed.

Marker genes such as antibiotic resistance or sensitivity genes may be used in identifying clones containing nucleic acid of interest, as is well known in the art.

The introduction may be followed by causing or allowing expression from the nucleic acid, e.g. by culturing host cells (which may include cells actually transformed although more likely the cells will be descendants of the transformed cells) under conditions for expression of the gene, so that the encoded polypeptide (or peptide) is produced. If the polypeptide is expressed coupled to an appropriate signal leader peptide it may be secreted from the cell into the culture medium. Following production by expression, a polypeptide or peptide may be isolated and/or purified from the host cell and/or culture medium, as the case may be, and subsequently used as desired, e.g. in the formulation of a composition which may include one or more additional components, such as a

pharmaceutical composition which includes one or more pharmaceutically acceptable excipients, vehicles or carriers (e.g. see below) .

The present invention extends in various aspects not only to peptides as disclosed, optionally coupled to other molecules, and peptide trimers, but also a pharmaceutical composition, medicament, drug or other composition comprising such a peptide or conjugate or peptide trimer, a method comprising administration of such a composition to a patient, e.g. for a therapeutic purpose, which may include preventative treatment, use of such a peptide, conjugate or peptide trimer in manufacture of a composition for administration, e.g. for a therapeutic purpose, and a method of making a pharmaceutical composition comprising admixing such a peptide, conjugate or peptide trimer with a pharmaceutically acceptable excipient, vehicle or carrier, and optionally other ingredients .

A pharmaceutically useful compound according to the present invention that is to be given to an individual, is preferably administered in a "prophylactically effective amount" or a "therapeutically effective amount" (as the case may be, although prophylaxis may be considered therapy) , this being sufficient to show benefit to the individual. The actual amount administered, and rate and time-course of administration, will depend on the nature and severity of what is being treated. Prescription of treatment, e.g. decisions on dosage etc, is within the responsibility of general practitioners and other medical doctors .

A composition may be administered alone or in combination with other treatments, either simultaneously or sequentially dependent upon the condition to be treated.

Pharmaceutical compositions according to the present invention, and for use in accordance with the present invention, may include, in addition to active ingredient, a pharmaceutically acceptable excipient, carrier, buffer, stabiliser or other materials well known to those skilled in the art. Such materials should be non-toxic and should not interfere with the efficacy of the active ingredient. The precise nature of the carrier or other material will depend on the route of administration, which may be oral, or by injection, e.g. cutaneous, subcutaneous or intravenous.

Pharmaceutical compositions for oral administration may be in tablet, capsule, powder or liquid form. A tablet may include a solid carrier such as gelatin or an adjuvant. Liquid pharmaceutical compositions generally include a liquid carrier such as water, petroleum, animal or vegetable oils, mineral oil or synthetic oil. Physiological saline solution, dextrose or other saccharide solution or glycols such as ethylene glycol, propylene glycol or polyethylene glycol may be included.

For intravenous, cutaneous or subcutaneous injection, or injection at the site of affliction, the active ingredient will be in the form of a parenterally acceptable aqueous solution which is pyrogen-free and has suitable pH, isotonicity and stability. Those of relevant skill in the art are well able to prepare suitable solutions using, for example, isotonic vehicles such as Sodium Chloride Injection, Ringer's Injection,

Lactated Ringer's Injection. Preservatives, stabilisers, buffers, antioxidants and/or other additives may be included, as required.

Liposomes, particularly cationic liposomes, may be used in carrier formulations .

Examples of techniques and protocols mentioned above can be found in Remington's Pharmaceutical Sciences, 16th edition, Osol, A. (ed) , 1980.

The agent may be administered in a localised manner to a desired site or may be delivered in a manner in which it targets particular cells or tissues.

A composition may be administered alone or in combination with other treatments, either simultaneously or sequentially dependent upon the condition to be treated.

A peptide, or a nucleic acid molecule encoding a peptide, may be provided in a kit, e.g. sealed in a suitable container which protects its contents from the external environment. Such a kit may include instructions for use .

A peptide or peptide trimer of the invention may be used in raising, generating and/or isolating antibody molecules that bind and are preferably specific for the peptide or peptide trimer. Suitable techniques are well established in the art, and include hybridoma technology, humanisation comprising CDR-grafting, phage display, ribosorne display, and transgenic mice able to produce human antibodies. A method of obtaining an antibody molecule against a peptide or peptide trimer in

accordance with the invention may comprise bringing the peptide or peptide trimer into contact with a population of antibody molecules and selecting one or more antibody molecules that bind and are preferably specific for the peptide or peptide trimer, optionally e.g. in the case of phage, display accompanied by nucleic acid encoding the antibody molecule which can be used to produce further quantities of the antibody molecule and derivatives thereof. Antibody molecules may be whole or complete antibodies, such as IgG, e.g. IgGl, IgG2, or IgG4, or

IgE, IgA, IgD or IgM, or antibody fragments such as scFv, Fab, F(ab')2, Fd and dAb .

The use of peptides of the invention allows the receptor- reactivity of the ligand to be controlled and different reactivities to be combined. Thus, peptides containing GpVI-recognition motifs may also comprise those for either integrin α2βl or VWF. By varying composition in this way the role of each receptor pathway may be elucidated. The inclusion of reactive groups at one end of the peptide would allow chemical coupling to inert carriers such that peptides might be delivered to pathological lesions such as chronic wounds or sites of acute traumatic injury without entry into the bloodstream.

Peptides of the invention, which can be said to be collagen-mimetic, may be useful as valuable reagents in a number of laboratory and clinical settings: • As a reagent for research into the activation and/or aggregation of platelets, exemplified by the development of CRP and by peptides containing the integrin-binding GER motifs .

• For the diagnosis of platelet disorders, which routinely uses collagen fibres extracted from animal tissues as a reagent in platelet aggregometry, turbidimetric platelet aggregometry or whole-blood aggregometry, or immobilised collagen preparations as in the Platelet Function Analyzer and other instruments .

• In the diagnosis of bleeding disorders, involvement or defects in specific receptors may be defined by the response of the patient platelets to receptor- specific CRPs.

• In monitoring platelet activity and the efficacy of treatment during anti-thrombotic therapy.

• As bioactive surface coatings, collagen-mimetic peptides may act to secure cell adhesion directly as well as to activate platelets locally, leading to the production and release of other bioactive molecules .

• Coupled to inert polymer supports, GpVI-, VWF- and/or integrin α2βl-reactive peptides may stimulate haemostasis in the circulation and serve as an adjunct or alternative to platelet transfusion in cases of platelet insufficiency that may result from auto-immune thrombocytopaenia or from therapeutic ablation of bone marrow as in cancer therapy, as well as from bleeding disorders from other causes such as Glanzmann's disease.

• Presented in a form that cannot enter the circulation, GpVI-, VWF- and/or integrin α2βl- reactive peptides or combinations of two or all three may stimulate haemostasis in acute trauma, e.g. after road traffic accident or battlefield

injury, being applied topically to wounds that would otherwise cause fatal blood loss.

• Similarly, immobilised collagen receptor-recognition motifs may serve to stimulate haemostasis in chronic wounds such as ulcers, where, first, cell attachment may be enhanced, and second, the release of activated platelet granule contents may stimulate the migration of cells from the bloodstream and from nearby damaged tissues that contribute to the healing process.

• In investigation or screening of test compounds that bind to platelets, to GpVI, VWF and/or integrin α2βl, or to the peptides or peptide trimers themselves .

A further aspect of the present invention provides a method of activating platelets, comprising treating platelets with a peptide trimer of the invention. This may be in vitro.

Activity of treated platelets, i.e. platelets following contact with a trimer of the invention, may be measured or determined, for example in the presence or absence of a factor or agent, test composition or substance of interest, employing suitable control experiments as expected in the art.

One further aspect of the invention provides a method of determining the effect of a factor on platelet activation and/or aggregation, the method comprising treating platelets with a peptide trimer of the invention and determining the effect of the factor on the platelet activation and/or aggregation. Platelet activation and/or aggregation may be determined in the presence or absence

of the factor or with the factor at different concentrations .

Different samples from different sources, e.g. different patients with or suspected of having a bleeding disorder, a disorder of platelet function or other disorder or disease, may be analysed and/or compared. Comparison may be made within or between GPVI genotype, i.e. aa, ab or bb.

In another aspect, the invention provides a method of investigating platelet activity or function or of diagnosing a dysfunction in platelet activity in a sample, the method comprising determining activation and/or aggregation of platelets in a sample treated with a peptide trimer of the invention.

Controls are employed as appropriate within the routine knowledge and expectation of those skilled in the art.

Further aspects and embodiments will be apparent to those skilled in the art in view of the present disclosure, including the following experimentation to illustrate embodiments of the invention and the accompanying figures. The term "comprise" as used herein has the meaning of "include", i.e. permitting the presence of one or more additional components. All documents mentioned anywhere herein are incorporated by reference.

All peptide structures and sequences are indicated using the standard amino acid single letter code. ^λλ O" is hydroxyproline .

All references cited in this specification are incorporated by reference.

EXPERIMENTATION

MATERIALS AND METHODS.

Peptides

Peptides were synthesised in an automated peptide synthesiser using routine Fmoc chemistry, as described in reference 11. Peptide sequences are identified in Table 1. The identity of peptides was confirmed using mass spectrometry, and their triple-helical status was proven using optical polarimetry. For mCRP and CRPlO, melting temperatures of 68°C and 76 ⁰ C. Longer peptides could not be determined, since their melting temperature was beyond the upper limit of the instrument. However, good optical rotation of the peptide solutions was obtained, indicating true assembly as triple-helices.

Dried peptides were stored frozen, then diluted to Img/ml in 0.01M acetic acid and the concentrate was stored at 4 ^P C until use. 200μl working dilutions were made in 0.01M acetic acid and stored at 4°C for up to 1 month.

Blood

Blood samples were taken by venepuncture from volunteer donors. lβml of blood was taken into 0.105M citrate tubes and mixed immediately. The blood was processed and tested within 2 hours of sampling and was centrifuged at 30Og for 12 min at room temperature. PRP was taken off into a 15ml Falcon tube and the platelets were counted on a Sysmex counter. The platelet count of the PRP was adjusted to 200 x 10 ⁹ /l.

Aggregation

Aggregation testing was carried out on a BioData PAP-4 analyser, or a Chronolog 490 aggregometer as indicated in results. Each channel was blanked with 250μl of platelet poor plasma (PPP) before each run. 250μl of PRP was tested with 5μl of agonist with concentration which gave a final concentration between lOOμg/ml and 0.01μg/ml. The aggregation trace was observed for β minutes or until a plateau had been reached.

Washed platelets

Washed platelets were prepared from PRP as described in reference 21 except that each centrifugation step was performed at 2600 rpm for 8 min and the platelet pellet was resuspended in 5 ml of buffer.

Adhesion

Ninety-six-well plates (Immulon-2 from Thermo Life Sciences, Basingstoke, UK) were coated with 100 μl of peptide (at 20 μg/ml in 0.01 M acetic acid) per well at 20°C overnight.

Platelets were then resuspended to 1.25 x 10 ⁸ platelets/ml in adhesion buffer (0.05 M Tris-HCl, 0.14 M NaCl, 2 mM MgCl ₂ , 0.1% BSA, pH 7.4) and allowed to rest for one hour at 25°C. Rested platelets were incubated with the fibrinogen receptor antagonist GR144305F (2 μM) for 15 min, to prevent platelet-platelet interaction. The assay was then performed as described except that ligand-coated wells were blocked with 200 μl of blocking buffer for one hour and 100 μl of platelets were added to each well. Platelet phosphatase activity was used as an index of platelet number, determined by a colorimetric procedure. These methods are well-established in the

field, as illustrated for example by references 11 and 21.

Protein Tyrosine Phosphorylation Acid citrate dextrose (ACD) (39 mM citric acid, 75 mM tri-sodium citrate- 2H ₂ O, 135 mM D-glucose, pH 4.5) and loading buffer (145mM NaCl, 5 mM KCl, 10 mM D-glucose, 1 mM MgSO ₄ , 0.5 mM EGTA, 10 mM HEPES-KOH, pH 7.36) were pre-warmed to 30 ⁰ C. ACD (10% (v/v) ) and apyrase (0.25 ϋ/ml final concentration) were added to PRP, followed by centrifugation (2000 rpm, 15 min) . The platelet pellet was resuspended in 5 ml of loading buffer. ACD was added as before, followed by centrifugation (2000 rpm, 10 min) . Platelets were then resuspended to 1 x 10 ⁹ platelets/ml in loading buffer, and allowed to rest for one hour at 30°C. Platelet suspensions (50 μl) were stimulated at 30 ⁰ C with 5 μl of the indicated concentration of peptide agonist or the vehicle control (0.01 M acetic acid) for two minutes, or for times stated in the figures and text below. Reactions were stopped by adding an equal volume of Laemmli's buffer and boiling for five minutes (reference 22) .

Western blotting and densitometry Protein separation was carried out as described in reference 23 except for the following modifications. Membranes were blocked with 5% (w/v) BSA dissolved in Tris-buffered saline Tween (TBST), 20 mM Tris-HCl, 140 mM NaCl, 0.1% Tween-20, pH 7.6) for 2 h at room temperature. Membranes were probed with 4G10 (1:2000 in 5% BSA-TBST, 1 h, 25°C) , washed for 45 min in TBST, incubated with the secondary antibody (1:5000 in 1% BSA-TBST, 1 h, 25°C) and then washed in TBST as before and subjected to enhanced chemiluminescence (ECL) . The intensity of the 72-75 kDa

T/GB2006/002958

32 band, which includes p72 ^syk , known to be activated downstream of GpVI (reference 24) , was quantified densitometrically using a Leica Q550C image analyser.

Immunoprecipitation of phospholipase Cγ2

Washed platelets (0.5 ml at 10 ⁹ /ml) were prepared from pooled platelet concentrates obtained from National Blood Service, Long Road, Cambridge, and treated with ligand as for protein tyrosine phosphorylation, above. The sample was lysed with saline containing 1% NonidetP40, ImM sodium orthovanadate, and 25mM tris-HCl, pH 7.3 (final concentrations given) . Dynal protein G beads were pre- coated with rabbit polyclonal anti-phospholipase C γ2 (Q- 20, Santa Cruz) at 1 μg/ml, then 30 μl bead suspension was added to the platelet lysate. After stirring overnight at 4°C, then washing in the same saline, proteins were recovered from the beads by boiling with Laemmli's buffer. Proteins were separated by SDS-PAGE and detected using the anti-phosphotyrosine monoclonal antibdy, 4G10 in Western blots as above.

Intracellular calcium measurements .

Fura-2 ratiometric fluorescence measurements were conducted at 37 ⁰ C in a Cairn spectrofluorimeter system (Cairn Research Limited, Faversham, Kent, UK) and converted to [Ca ²⁺ ] ± as described elsewhere using a dissociation constant for Ca ²⁺ of 224 nM (Rolf et al, Thromb. Haemost 2002; 88: 495-502).

RESULTS

Platelet aggregometry was used to examine the ability of the peptides to promote platelet activation. The data shown in Figure IA and IB in platelet rich plasma (PRP) indicate that the potency of a peptide in inducing

platelet aggregation varies with the length of the peptide, and therefore with the number of GPO triplets that it contains. This holds true up to 24 GPO triplets (CRP-24) . The onset of aggregation occurred using peptides of lβ triplets or more, below which limit no aggregation was obtained in platelet-rich plasma using ligand concentrations of up to lOOμg/ml. Significantly, no aggregation was observed using mCRP or CRP-IO or CRP- 13 (n=4) at up to lOOμg/ml.

The biphasic nature of these curves, with sub-maximal aggregation induced by supra-maximal levels of the CRPs, reflects the likely mechanism of activation. Thus, Long CRPs can accumulate multiple copies of the activatory receptor along their length, and this receptor clustering results in intracellular signals. However, at high CRP levels, the excess of peptide over receptor results in each copy of the receptor being occupied, so that clustering is reduced, with correspondingly lower platelet response. The capacity of peptides to cause aggregation of washed platelet suspensions is reduced, in line with usual experience, reflecting the removal of the plasma protein fibrinogen that cross-links adjacent activated platelets through its receptor, integrin αllbβ3.

To identify whether the CRPs interacted with platelet GpVI, static adhesion assays were undertaken using washed platelets, and platelets were incubated with two single chain antibodies (scFvs), 10B12 and 1C3. These antibodies have been shown to interact with Glycoprotein VI. The binding of the first scFv, 10B12, has been mapped to the collagen-binding site where it is anti- functional and inhibits collagen-mediated platelet

activation (reference 19) . The second scFv, 1C3, does not compete with 10B12, and its capacity to inhibit platelet activation and adhesion is much lower than that of 10B12. Modest inhibition of platelet activation by 1C3 has been observed under physiological blood flow, for example (reference 25) .

The ability of the peptides to support platelet adhesion and of these antibodies to inhibit platelet adhesion to the various CRPs is shown in Figure 2. Adhesion increased with peptide length, such that CRPlO and CRP13 supported less platelet adhesion than CRP16, but all longer peptides showed similar levels of adhesion as CRPl 6. The adhesion to all peptides was completely blocked by 10B12, and substantially by 1C3. However, although 1C3 caused complete blockade of platelet adhesion to CRP18 and shorter peptides, some adhesion remained to the two longer CRPs, with about 25% residual adhesion to CRP24. These data show that the platelet interaction with all the CRPs tested is dependent upon GpVI, and indicate a complex binding process to the longer CRPs. It is significant that the monomeric mCRP, which contains cysteine residues designed to allow crosslinks to be introduced between triple-helices, supported greater adhesion than any of the other peptides tested here. This was fully inhibited by 10B12, but only partially by 1C3.

The ability of 10B12 and 1C3 to inhibit the CRP- stimulated aggregation process was tested. Sensitivity was observed to both antibodies of the aggregation induced by CRP24, with complete inhibition of aggregation caused by 10B12 at a supramaximal dose of CRP24 as shown in Figure 3. Similar data were obtained when CRP18 was

used as an agonist, although 1C3 was not tested (not shown) .

The five common single nucleotide polymorphisms (SNPs) in GpVI are linked, providing two common alleles, a and b, and the expression of GpVI on the platelet surface has been linked to these two common alleles (reference 26) . Haplotyped blood can be used to test further the dependence of CRP function on GpVI, since the response to collagen or CRP depends upon GpVI haplotype.

Blood donors were assigned to three groups according to their haplotype, aa, ab and bb . Aggregation curves were obtained using various doses of CRP18. These are shown in Figure 4. The sensitivity of platelets to the ligand varied with haplotype, the EC ₅₀ for CRP18 being about 2, 7 and >20μg/ml for aa, ab and bb platelets respectively. Little aggregation was observed in any of the bb platelet samples, although since the frequency of bb donors within the general population is around 2%, there are very few samples available to increase the size of the study.

To explore the activatory events underlying the aggregation process, we examined the ability of the CRP set to elicit protein tyrosine phosphorylation from human platelets, an early event that is known to mediate a series of intracellular signals. In Figure 5, the increase in protein tyrosine phosphorylation elicited by increasing levels of each of the peptides is shown. The experiments were run as two sets, with CRPlO, CRP18 and CRP24 run in parallel, and with CRP13, CRPlβ and CRP21 run in parallel. mCRP was included in the second group. In this process, proteins from activated platelets are separated by gel electrophoresis and transferred to a

nitrocellulose membrane. Proteins that contain phosphorylated tyrosine residues are located using a phosphotyrosine-specific antibody, detected using standard methodology. Within each set, protein tyrosine phosphorylation increased with peptide length. CRPlO was virtually without effect, with only minor increase in the intensity of a band at about 75kDa being observed at the highest peptide dose tested, 20μg/ml. Similar quantitative effect was observed using CRP18 at 0.5μg/ml or CRP24 at 0. lμg/ml . In the second set, the lowest effective doses were 5, 1 and 0.1 for CRP13, CRP16 and CRP21 respectively. mCRP caused detectable increase in tyrosine phosphorylation at lμg/ml. These data can be used to assemble a potency series, shown in Table 2, which also includes estimates of EC ₅₀ obtained by densitometric measurement of these Western blots. On review, some of the estimated figures have been revised, but the rank order remains the same.

The effect of the peptides in eliciting protein tyrosine phosphorylation from human platelets was abolished by 10B12 and substantially reduced by 1C3. Densitometric data are shown in Figure 6.

The facility with which peptides can be synthesised in the laboratory allows different receptor-recognition motifs to be incorporated into the same structure. We synthesised triple-helical peptides comprising the integrin-binding motif GFOGER within GPO triplets. The resultant peptide is known as CRP-9GFO9, and its GPO content is the same as CRP18. Comparison of the effectiveness of the two peptides was undertaken in terms of platelet adhesion and aggregation as described above.

Used as an adhesive substrate at the same coating density (10μg/ml) , CRP-9GFO9 was a more effective adhesive substrate for human platelets than CRP18, shown in Figure 7. The scFv 10B12 caused minor inhibition of adhesion to CRP-9GFO9, whereas 1C3 was without significant effect. As shown in Figure 2, adhesion to CRP18 was completely abolished by either of the scFvs.

To test the role of the integrin α2βl in the platelet adhesion to CRP-9GFO9, the blocking antibody 6Fl was used, which is an anti-functional monoclonal antibody specific for the integrin α2 subunit (reference 27) . No significant inhibition was observed using 6Fl alone, but in combination with the anti-GpVI scFvs, significant and substantial inhibition was observed, with the 6F1/10B12 combination being completely effective.

The two peptides were tested for potency in aggregation, using both the aa and ab haplotypes of GpVI. CRP-9GFO9 was about twice as potent as CRP18, as shown in Figure 9, the observed EC ₅₀ for the two peptides being 0.85 and 2.2μg/ml respectively in aa platelets and 2.5 and 6μg/ml in ab platelets respectively. Similar findings were observed using washed platelets (Figure 1C) . In the single bb donor available, increased potency of CRP-9GFO9 was also observed. However, there was no correlation between potency of the peptide and the α2βl genotype, which has also been correlated with receptor expression and reported to have a functional role (references 28 and 29) .

Platelet aggregation stimulated by CRP-9GFO9 was abolished by receptor blockade using the anti-GpVI scFvs 10B12 or 1C3 or by the monoclonal antibody 6Fl. The

peptide was used at up to just-maximal dose, and inhibition was complete by each of the antibodies, as shown in Figure 10.

CRP-9GFO9 elicited tyrosine phosphorylation from washed platelet suspensions, when applied at lβμg/ml for up to 5 min (Figure 11A) , with peak activity observed at about 1.5 min. This property was markedly reduced by the anti- integrin α2 subunit antibody, 6Fl, used at 5 μg/ml (Figure HA lane 8) (no antibody was present in the other lanes) , but depended mainly on Glycoprotein VI, being abolished by the anti-GpVI antibody, 10B12 (WO03/054020) , used at 10μg/ml (Figure HB, lanes 1 to 8 ) .

To explore the signalling properties of the Long CRPs further, calcium measurements were made in platelet suspensions loaded with the indicator Fura2 at 3μM, as described in Materials and Methods. Comparison was made between Long CRPS and CRP-XL, each used at 2μg/ml. The cross-linked peptide was more effective, causing a rise in intracellular calcium of about 2μM, compared with ~0.7μM elicited by CRP-24 (Figure 12A). Using doses of CRP-18 and CRP-9GFO9 (16μg/ml) that elicited rather greater protein tyrosine phosphorylation than CRP-XL used at 0.5μg/ml, the Long CRPs were less effective, causing smaller, but also less prolonged rises in calcium than CRP-XL (Figure 12B) .

The regulation of platelet intracellular calcium by collagen depends largely on phospholipase Cγ isoforms. The enzyme was immunoprecipitated from platelets activated with CRP-18 or CRP-9GFO9 at 16 μg/ml, using an anti-phospholipase Cγ2 antibody, then the immunoprecipitate was probed for tyrosine phosphorylation

in a Western blot. The result is shown in Figure 13. CRP-18 caused slight but detectable tyrosine phosphorylation of PLCγ2 (lane 2), whilst that caused by CRP-9GFO9 was much more marked (lane 3) , though much less than that caused by CRP-XL at either 0.5 or lOug/ml

(lanes 4 and 5) . This indicates differential ability of the cross-linked and long CRPs to activate specific pathways in platelets, such that the calcium pathway is more prominent when Glycoprotein VI has been activated by an undefined collagenous matrix (CRP-XL) than when the defined Long CRP has been used, despite both being potent stimulators of platelet aggregation. The effect of inclusion of the integrin recognition motif in CRP-9GFO9 is quite marked, with greater activity on PLCγ2 than CRP- 18 used at the same level.

DISCUSSION.

The data reported here show the ability of a set of monomeric peptides with the capacity to bind GpVI to induce full platelet aggregation.

The primary sequence of the peptides, the repeated GPO triplet, is identical with that widely used in the field to activate GpVI (references 7, 14, 17 and 30). In the present work, GpVI-dependence was shown in several ways.

Human platelets genotyped for the common alleles of GpVI were used. The homozygous donor with the aa haplotype expresses more copies of GpVI on the platelet surface, and such platelets are more responsive to stimulation by collagen or CRP than those of the bb haplotype (reference 26) . The dependence on haplotype of the response to CRP18 indicates the role of GpVI in its ability to bind and activate the platelet.

Secondly, the ability of two different anti-GpVI scFvs to inhibit platelet aggregation elicited by the CRPs, or to inhibit adhesion, or to inhibit platelet protein tyrosine phosphorylation further indicates that these peptides operate by binding and activating GpVI .

Previous evidence (references 12 and 17) has suggested that non-cross-linked peptides, based upon a [GPO] io core, can induce some platelet activation, measured as protein tyrosine phosphorylation, whilst in cross-linked form such peptides are potent platelet agonists (reference 16) . In the literature, chemical cross-linking exploits specific amino acids, either cysteine or lysine, incorporated at the point of synthesis into the peptide, allowing covalent coupling between triple-helices by the use of agents designed for the purpose, as described in reference 16. This conversion of a peptide which can be used as an antagonist (references 17 and 31) into a potent agonist provides insight into the mechanism by which signals are transmitted from such peptides to the interior of the platelet, and by inference into the mechanism by which native collagens signal. It is assumed that receptor clustering is needed, and that an activatory collagenous ligand must present a matrix of GpVI-recognition motifs suitably spaced to achieve the necessary degree of association of multiple copies of GpVI. Hitherto it has been assumed that a three- dimensional array, as in the highly-organised collagen fibre or in the randomly cross-linked CRPs just described, is needed for platelet activation.

The major advantage of the present peptides is that they can activate platelets without the need to introduce

chemical cross-links. The activation process reaches functional levels, i.e. will cause platelet aggregation, without such chemical strategies being employed. However, such utility is limited by the length of the peptide. Thus, as an agonist specific for GpVI, sixteen GPO repeats were needed for full activation to occur. To reach relatively high potency, EC _5O about lμg/ml or about 10OnM, eighteen GPO repeats are required.

It would be anticipated that scFv 10B12, which occludes the collagen-binding site on GpVI would be anti- functional in the various assays shown here. The capacity of scFv 1C3 to achieve the same end, although in some cases less completely, indicates that 1C3 can prevent the side-by-side association of GpVI in complex with the CRP that is needed for activation of platelets. Thus, an insufficient number of copies of the GpVI : 1C3 complex can associate with the CRP to generate an activatory signal.

In the present study we observed that mCRP, a peptide including cysteine residues at the N- and C-terminal regions which flank a [GPO] io core, exhibited greater activity as an adhesive substrate and in inducing tyrosine phosphorylation than its nearest equivalent, CRPlO. Although this mCRP was not cross-linked deliberately, it is likely that under ambient (oxidising) conditions some disulphide bridges form spontaneously that may introduce higher-order structure. This would be consistent with enhanced clustering of receptors needed for signalling, but the absence of aggregatory activity indicates that such properties remain at a low level. It is also plausible that cysteine residues may bind better

to the adhesive surface of the 96-well plate than peptides lacking such reactive sidechains.

Potency of such peptides can be further enhanced by incorporation of binding motifs for other receptors. An example is the GFOGER motif for integrin α2βl which, when placed between [GPO] ₉ flanking sequences as in CRP-9GFO9, increases potency of the peptide by a factor of two. Similar activity may well accrue from incorporating VWF- binding sequences along with either GpVI- or α2βl-binding motifs. Here we show that the inclusion of the GFOGER motif substantially enhances the ability of the peptide to activate phospholipase Cγ2, leading to a calcium signal .

Some potent ligands for GpVI other than collagen occur in nature. One such is the snake venom protein convulxin (reference 32) . This is a C-type lectin which exists as a tetramer (reference 33), and thus has the capacity to cluster at least four copies of GpVI. Given the possibility that GpVI may occur on the platelet surface in dimeric form (references 12 and 34), such ligands may offer the possibility of higher-order clustering of the receptor. However, they suffer from the disadvantages that would be expected in generating a snake venom protein as a biochemical reagent: purity may be limited, availability may be restricted by the supply of animals, and the production process may be expensive. Used in a clinical setting, such venom proteins would be immunogenic.

The novel CRP reagents presented here can be defined in terms of both their molecular structure and their concentration, which is not the case with collagenous

ligands which have been randomly chemically-cross-linked, or indeed with collagen fibres, which present a reactive surface rather than a soluble ligand to platelet and other cellular receptors.

TABLE 1

PEPTIDES USED IN THIS STUDY AND THEIR SEQUENCES

The single-letter code is used for the above peptides; O = hydroxyproline

TABLE 2

REFERENCES

I. Farndale et al. J Thrombos Haemost. 2004; 2:561-573.

2. Gresele et al . eds . Platelets in thrombotic and non- thrombotic disorders. Cambridge: Cambridge University Press; 2002

3. Michelson AD ed Platelets. Amsterdam: Academic Press; 2002

4. Arnout et al . eds. Thrombosis. Fundamental and clinical aspects. Leuven: Leuven University Press; 2003

5. Morton et al . Biochem J. 1994;299 ( Pt 3):791-797.

6. Barnes et al . Curr . Opin. Haematol. 1998; 5:314-320

7. Knight et al . Cardiovasc Res. 1999; 41:450-457.

8. Knight et al . J Biol Chem. 2000; 275:35-40.

9. Emsley et al . Cell. 2000; 101:47-56.

10. Xu et al. J Biol Chem. 2000; 275:38981-38989.

II. Siljander et al . J Biol Chem. 2004; 279:47783-47772

12. Farndale et al . Biochem Soc Symp. 2003; 70:81-94.

13. Gibbins et al . J Biol Chem. 1996; 271:18095-18099.

14. Kehrel et al . Blood. 1998; 91:491-499.

15. Zijenah et al. Thromb. Res. 1990; 59:553-566.

16. Morton et al. Biochem J. 1995; 306: 337-344.

17. Asselin et al . Biochem J. 1999; 339:413-418.

18. Smith et al. The Journal of Biological Chemistry. 2000; 275:4205-4209.

19. Smethurst et al. Blood. 2004; 103:903-911.

20. Romijn et al . J Biol Chem. 2001; 276:9985-9991.

21. Onley et al . J Biol Chem. 2000; 275:24560-24566.

22. Laemmli Nature. 1970; 227:680-685.

23. Sundaresan et al . FEBS Lett. 2002; 528:139-144.

24. Ichinohe et al . J Biol Chem. 1997; 272:63-68.

25. Siljander et al . Blood. 2004; 103:1333-1341.

26. Joutsi-Korhonen et al. Blood. 2003; 101:4372-4379

27. Coller et al . Blood. 1989; 74:182-192.

28. Kritzik et al . Blood. 1998; 92:382-388.

29. Carlsson et al. Blood. 1999; 93:3583-3586.

30. Gibbins et al. FEBS Lett. 1997; 413:255-259.

31. Penz et al . Faseb J. 2005; 19:898-909.

32. Polgar et al. J Biol Chem. 1997; 272:13576-13583

33. Batuwangala et al . Acta Crystallogr D Biol Crystallogr. 2004; 60:46-53.

34. Miura et al. J Biol Chem. 2002; 277:46197-46204.