WO2007010240A2

WO2007010240A2 - Method for high-throughput screening

Info

Publication number: WO2007010240A2
Application number: PCT/GB2006/002664
Authority: WO
Inventors: Hans Deckmyn; Isabelle Salles; Alexandre Fontayne
Original assignee: European Cardiovascular Genetics Foundation
Priority date: 2005-07-18
Filing date: 2006-07-18
Publication date: 2007-01-25
Also published as: GB0514678D0; WO2007010240A3

Abstract

This invention relates to methods for assessing platelet activation. Such methods can be used to diagnose whether a subject has an increased risk of bleeding or thrombosis, to screen for compounds that are agonists or antagonists of platelet activation, or to screen for defective platelets. The invention provides a method for assessing platelet activation, which comprises: stimulating platelets in a platelet sample to cause activation of the platelets; capturing activated platelets from the platelet sample in preference to resting platelets; and detecting the captured activated platelets.

Description

Method for High-Throughput Screening

This invention relates to methods for assessing platelet activation. Such methods can be used to diagnose whether a subject has an increased risk of bleeding or thrombosis, to screen for compounds that are agonists or antagonists of platelet activation, or to screen for defective platelets.

Assessment of platelet activation is classically performed by platelet aggregometry. This technique was originally described by Born²². It closely mimics the physiological process in which a stimulus induces an activation cascade that finally results in GPIIb/llla mediated platelet aggregation. However, platelet aggregometry suffers from several serious drawbacks. Although results are obtained in real time, it takes around five minutes for full aggregation to occur. In view of the relatively short active ex vivo fife- time of platelets, this precludes study of large numbers of agonists or antagonists at different concentrations unless a large number of aggregometers is used in parallel. The results obtained are not always reproducible, and interpretation of the results is sometimes not straightforward. For example, it is not clear which parameter should be measured (such as lag time, slope, or maximal amplitude). Depending on which machine is used to perform the test, 250-1 OOOμl of platelet-rich plasma (PRP) is needed. Furthermore, whilst hypo-active platelets can be detected (for example, due to Glanzmann thrombasthenia, Bernard- Soulier syndrome, or platelet inhibitors) it has so far been particularly difficult to detect hyper-active platelets. In fact, it is not even clear at present whether such platelets exist.

Other parameters of platelet activation, such as secretion of ATP (detectable by luminescence) or appearance of platelet activation markers

(such as activated integrins, expression of granular proteins such as P- selectin (also known as CD62P or GMP-140), or lysosome integral membrane protein-CD63²³) which are detectable by fluorescence- activated eel! sorting (FACS) analysis, have been proposed as alternatives to aggregometry. However, such methods are time-consuming, vary in sensitivity, require advanced and expensive instrumentation, and are not high-throughput.

It is desired, therefore, to provide sensitive, high-throughput assays to study platelet function, and to be able to detect hyper- (and hypo-) reactive platelets.

According to the invention there is provided a method for assessing platelet activation, which comprises: stimulating platelets in a platelet sample to cause activation of the platelets; capturing activated platelets from the platelet sample in preference to resting platelets; and detecting the captured activated platelets.

It is preferred that the selectivity of capture of activated platelets to resting platelets is at least 2:1 activated platelets to resting platelets, preferably at least 3:1, or 10:1, more preferably at least 100:1,

Preferential capture of activated platelets may be achieved in any suitable manner. However, it is preferred that activated platelets are captured using a monoclonal antibody (or a fragment or derivative thereof) that binds selectively to activated platelets. Preferably the monoclonal antibody binds to an activated platelet marker, such as an activated integrin (for example, the activated form of GPHb/llla or GPIa/lla), or a granular protein (for example, CD62P or CD63).

Monoclonal antibodies that bind selectively to activated platelets are known. PAC-1 is a monoclonal antibody that is commercially available from BD Biosciences (San Diego, California: Cat. No. 340507). It recognises an epitope on the GPllb/IIIa complex of activated platelets at or near the fibrinogen binding site. The GPIIb/Jlla complex (also known as σllb/?3) is a platelet-specific member of the integrin family of heterodimeric adhesive protein receptors found on a variety of cell types. The GPIIb/llla complex is located on the surface membrane of resting platelets. Platelet activation induces a calcium-dependent conformational change in GPIIb/llla that exposes a ligand binding site. Binding of fibrinogen to the activated form of the GPIlb/flla receptor is required for platelet aggregation. PAC-1 binds only to activated platelets and appears to be specific for this recognition site within GPIIb/llla.

AK-4 antibody is a purified mouse anti-human monoclonal antibody that is commercially available from BD Biosciences (San Diego, California: for example, Cat. No. 551345). It reacts with CD62P (also known as P- selectin), the 14OkDa membrane glycoprotein formerly known as platelet activation-dependent granule membrane protein (PADGEM), or GMP-140. CD62P is an integral part of the α-granule membrane of platelets and is rapidly translocated to the platelet plasma membrane upon activation.

Typically the monoclonal antibody (or fragment or derivative) is immobilised to a solid phase, such as a well of a microtitre plate.

The captured activated platelets may be detected in any suitable way, but they are preferably detected using a monoclonal antibody (or a fragment or derivative thereof) that binds selectively to platelets, but which does not differentiate between resting and activated platelets. Suitable monoclonal antibodies for this purpose are antibodies against GPIbα. Examples of commercially available monoclonal antibodies that bind specifically to platelets but which do not differentiate between resting and activated platelets are: SZ2 from lmmunotech (Marseille, France) or Beckman Coulter, AK2 from Cymbus Biotechnology (Hampshire, United Kingdom) or BioSource International, or HIP1 from Pharmingen (Becton Dickinson). The monoclonal antibody (or fragment or derivative) used to detect the captured activated platelets may be labelled with a detectable label, or it may be a primary antibody detectable by a labelled binding agent (which may be a secondary antibody).

Preferably the amount of captured activated platelets is quantitatively determined to assess platelet activation. A method of the invention may be a quantitative sandwich enzyme-linked immunosorbent assay (ELISA), radioimmunossay (RIA), immunoradiometric assay (IRMA), fluorescent immunoassay (FlA), chemiluminescent immunoassay (CLIA), or mirror resonance assay. Such methods allow high sensitivity assessments of platelet activation.

Methods of the invention are particularly advantageous because they can be used to determine simultaneously the effect of different concentrations of a platelet agonist on platelet activation. The platelet sample is split into a plurality of sample portions, and the platelets of each sample portion are stimulated with a different concentration of the same platelet agonist at the same time. A dose-response curve of platelet activation to concentration of platelet agonist can be generated.

In other embodiments of the invention the effect of a plurality of different agonists on platelet activation can be determined (preferably at the same time). The platelet sample may be stimulated with a plurality of different agonists (preferably at the same time), or the platelet sample may be split into a plurality of sample portions, and the platelets of each sample portion stimulated with a different agonist (preferably at the same time).

Methods of the invention may also be used to assess inhibition of platelet activation by contacting the platelets with an antagonist of platelet activation. According to such methods platelets of a platelet sample are stimulated to cause activation of the platelets, and the platelets are contacted with an antagonist of platelet activation. Activated platelets are then captured and detected. The platelets may be contacted with the antagonist before, during, or after stimulation.

In preferred embodiments, the platelet sample is split into a plurality of sample portions, and the platelets of each sample portion are contacted with the same concentration of the same agonist, and a different concentration of the same antagonist. A dose response curve of platelet activation to concentration of antagonist may then be generated.

In other embodiments of the invention the effect of a plurality of different antagonists on platelet activation can be determined (preferably at the same time). The platelet sample may be contacted with a plurality of different antagonists (preferably at the same time), or the platelet sample may be split into a plurality of sample portions, and the platelets of each sample portion contacted with a different antagonist (preferably at the same time).

Platelets may be stimulated in any suitable manner. Preferably, the platelets are contacted with a platelet agonist, for example ADP, TRAP, a TXA₂ analogue (such as U46619), collagen, or Collagen Related Peptide (CRP).

Any suitable antagonist of platelet activation may be used. An example is S-nitroso-N-acetyl penicillamine (SNAP).

Methods of the invention can be used to diagnose defects in platelet activation, or in inhibition of platelet activation.

Methods of the invention may be used to determine whether a subject has, or is at risk of having, an abnormal blood condition associated with an increased risk of bleeding. A method of the invention is carried out on a platelet sample obtained from the subject to determine whether activation of platelets in the sample is reduced compared to normal platelet activation.

Platelets in the sample may be stimulated with a platelet agonist. If platelet activation is reduced compared to normal platelet activation, the subject is diagnosed as having, or being at risk of having, an abnormal blood condition associated with an increased risk of bleeding. Alternatively, platelets in the sample may be stimulated with a platelet agonist, and contacted with a platelet antagonist (either before, during, or after stimulation). If platelet activation is reduced compared to normal platelet activation this indicates that the response of the platelets in the sample to the antagonist is greater than normal, and the subject is diagnosed as having, or being at risk of having, an abnormal blood condition associated with an increased risk of bleeding.

Subjects with a hypo-reactive functional platelet type are termed easy bruisers (EBs). Such individuals may be normal healthy individuals, but the EB phenotype may be of major clinical significance in situations where normal haemostasis is challenged. EB identification is also expected to be of clinical use in certain categories of patients.

Identification of the EB phenotype by a method of the invention is expected to be of value for the following categories of individuals:

i) Patients with a low platelet count (< 50 x 10⁹/L or even < 100 x 10⁹/L in exceptional circumstances, i.e. brain surgery or liver biopsy) because of, reduced generation by the bone marrow in cancer (i.e. by infiltration of the bone marrow with cancerous cells or iatrogenic), increased platelet destruction (i.e. in patients with antibody-mediated thrombocytopenia's), or increased consumption (i.e. in patients with disseminated intravascular coagulation) or Hypersplenism; ii) Patients undergoing major surgical procedures known for a high risk of significant blood loss (i.e. cardio-thoracic surgery, orthopaedic surgery, certainly the category of redo's; organ transplantation, especially liver, heart and lung and possibly kidney; surgery of trauma patients; brain surgery inclusive imaging; vessel surgery); iii) Patients undergoing biopsies of liver, kidney or brain or other vital organs; iv) Patients with inherited bleeding disorder such as haemophilia A or B, Von Will brands' disease and other more rare forms of inherited coagulopathies or with acquired coagulopathies (i.e. liver disease, viral hepatitis); v) Patients receiving blood thinning drugs (anti-coagulants, antiplatelet drugs, thrombolytic drugs) to reduce the risk of venous and/or arterial thrombosis. One of the major unwanted side effects of these drugs is bleeding. Cerebral bleeds or bleeds in other vital organs can cause life-long disability or even death; vi) Healthy individuals who give blood as a donor by routine donation or blood derivatives by apheresis technology. One of the most frequent complaints of donors is serious bruising post-donation at the site of venepuncture.

If a patient has an EB phenotype, special precautions can be taken. Some examples are:

i) In surgery: increase availability of donor blood, consider additional drugs (i.e. DDAVP, Tranexamic Acid) to reduce the risk of bleeding; ii) In patients with low platelet counts because of low endogenous production. Increase platelet transfusion trigger, i.e. from 1O x 10⁹/L to 20 x 10⁹/L Currently, donor platelets are generally given when the platelet count < 10 x 10⁹/L (higher if patient is unstable or in particular surgical and/or invasive diagnostic procedures). However, in EB patients a more conservative level of 20 x 10⁹/L or even 50 x 10⁹/L could be used. This may reduce the chance of serious bleeding without causing unacceptable increase in the demand of therapeutic donor platelet concentrates. Similarly, platelet levels required to cover invasive procedures could be tailored better to the patient by determining the platelet functional phenotype; iii) In patients with low platelets because of antibody-mediated destruction of platelets. Treatment of autoimmune mediated thrombocytopenia is based on a combination of the platelet count and clinical signs of bleeding. However, in patients with a very low platelet counts (<20 x 10⁹/L) treatment is given to reduce the risk of serious bleeding, even in the absence of bleeding. In these patients expensive therapies or therapies with major side effects are prescribed to reduce the relatively small risk of bleeding. Stratifying of these therapies on basis of the platelet functional phenotype may be possible; iv) The prophylactic treatment regime in haemophilia patients may be better tailored to the patient depending on platelet functional phenotype. In addition life-style advice could be modified depending on the platelet functional phenotype; v) Patients on blood thinning drugs (these include anti-coagulants, anti-platelet drugs or thrombolytic drugs, or novel anti-platelet drugs still under development and based on inhibitors of collagen mediated platelet activation, such as drugs directed against platelet GPVI (such as antibodies 10B12 described in Patent WO 03/054020 A2) or against any of the proteins in the signalling pathway downstream of GPVI (for example src family kinases such as fyn). Patients with an EB phenotype may require reduced dosing with these drugs. There are no effective antidotes for several of these drugs (i.e. Clopridogel, GPKb/llla antagonists, Tissue plasminogen activator, Streptokinase). Overdosing is an issue of serious concern and the therapeutic window of several of these drugs is relatively narrow. Dosing schedules may be adjusted on the functional platelet phenotype. Moreover, many novel 'blood thinning' drugs are being trialled or are under development. Whether these novel drugs will enter the market will depend on their clinical effectiveness and more importantly on the incidence and severity of side effects

(bleeding being the most important). Pharmaceutical and biotechnology industries will therefore have an interest to identify those patients as risk of bleeding. The platelet functional phenotype may be one of these.

vi) In donors. Blood services may more carefully manage donors with an EB platelet phenotype. In addition platelets from donors with an EB phenotype may be less effective in preventing bleeding in the recipient when compared with platelets from 'normal' donors. Donors with EB phenotypes may be barred from giving platelets.

Methods of the invention may be used to determine whether a subject has, or is at risk of having, an abnormal blood condition associated with an increased risk of thrombosis. A method of the invention is carried out on a platelet sample obtained from the subject to determine whether activation of platelets in the sample is increased compared to norma! platelet activation.

Platelets in the sample may be stimulated with a platelet agonist. If platelet activation is increased compared to normal platelet activation, the subject is diagnosed as having, or being at risk of having, an abnormal blood condition associated with an increased risk of thrombosis. Alternatively, platelets in the sample may be stimulated with a platelet agonist, and contacted with a platelet antagonist (either before, during, or after stimulation). If platelet activation is increased compared to normal platelet activation this indicates that the response of the platelets in the sample to the antagonist is less than normal, and the subject is diagnosed as having, or being at risk of having, an abnormal blood condition associated with an increased risk of thrombosis.

Subjects with increased activation of platelets or with a hyper-reactive functional platelet type are termed easy clotters (ECs). Such individuals may be normal healthy individuals, and the EC phenotype may be of major clinical significance in situations where increased risk of clot formation may cause severe disease, i.e. myocardial infarction (Ml), stroke or other thrombotic events in arterial circulation. EC identification is also expected to be of clinical use in certain categories of healthy individuals and in patients.

Identification of the EC phenotype by a method of the invention is expected to be of value for the following categories of individuals:

Patients with a risk of formation of blood clots or thrombi in arteries (atherothrombosis or AT) resulting in vessel occlusion (which may be partial or complete). AT leads to suboptimal (or a lack of) perfusion of target tissue (the tissue to which the supplying blood vessel is blocked by AT) with blood. This leads to insufficient (or loss of) oxygen supply causing loss of function or irreversible damage with death of tissue. The clinical relevance of AT is substantial with the following known examples:

i) Patients with AT in blood vessels of the heart (i.e. Myocardial

Infarction (Ml), unstable and stable Acute Coronary Syndrome(ACS)); ii) Patients with AT in blood vessels of the brain with cerebral stroke; iii) Patients with AT in peripheral blood vessels (peripheral artery disease or PAD), i.e. of the legs resulting in intermittent claudication and/or chronic leg ulcers; iv) Patients with AT in blood vessels supplying other vital organs

(i.e. the eye and blindness, the kidneys and kidney insufficiency).

If a healthy individual or a patient has an EC phenotype, special precautions can be taken or treatments prescribed. Some examples are:

i) Preventative measures. In healthy individuals with a family history of Ml or of another disease associated with AT (i.e. stroke, PAD) use of simple preventative medication can be applied from an early age (i.e. daily use of aspirin) to substantially reduce the risk of disease. Moreover, life-style advice can be used to further modify disease risk (i.e. cessation of smoking, reducing body weight, increased exercise, change of diet). Defining the EC phenotype of platelets by a method of the invention may be of significant value to better tailor preventative measures to those groups in the population most at risk of AT. This is important from a public health perspective since reducing the chance of AT in individuals considered to be at increased risk (i.e. because of a positive family history of AT related disease) is not without side effects. When applied to large cohorts of the population (i.e. in all individuals < 50 years of age and with a positive family history) even a small chance of side effects will cause pathology in a large number of individuals. The main side effects of prevention of AT with simple drugs (like aspirin) is major bleeding (i.e. in the brain or eye) or gastrointestinal complications, such as gastritis. Methods of the invention can be used to better target the preventative use of drugs like aspirin (or future alternatives) to those individuals most at risk (i.e. EC phenotype as determined by a method of the invention, positive family history for AT related diseases and age > 50 years). An EC phenotype as defined by a method of the invention may lead to the lowering of the age limit for the use of preventative measures (i.e. aspirin, life-style changes) in case of a positive family history of AT related diseases and an EC phenotype. Methods of the invention are expected to have widespread use since many individuals in the population are at risk of AT. AT-related diseases are the main cause of death in Western Society and the cost of prevention is substantial. With improved diagnosis of individuals at risk, overall health care system costs can be reduced; Acute measures. If a patient is diagnosed with an acute AT- related disorder (such as Ml or stroke) treatment needs to be instigated with urgency. Treatment in the acute phase consists of combinations of thrombolytic drugs, anti-platelet drugs and other drugs (i.e. lipid lowering drugs, beta-blockers, angiotensin converting enzyme inhibitors). Cocktails of anti-platelet drugs (i.e. aspirin, Clopridogel and GPIIb/llla inhibitors) are nowadays common practice. However, bleeding is one of the major side effects of combination anti-platelet drug therapy in up to 2% of patients. For patients with an EC phenotype the application of multiple anti-platelet drugs may be warranted but for patients with EB phenotypes use of multiple anti-platelet drugs may be contraindicated. Methods of the invention may aid the stratification of drug therapy in the acute phase of AT-related disorders reducing the risk of post-event complications

(recurrence of Ml) in EC patients or the risk of bleeding in EB patients; iii) As part of invasive procedures: Blockade of platelet activation with anti-platelet drugs is considered to be of benefit in the pre- intervention and peri-interventfona! treatment by angioplasty. I.e. there is growing recognition that such treatment is useful for NSTEMi (Non ST-Elevated Ml, i.e. patient with a heart attack without effects on the electrocardiogram that have been regarded as diagnostic, still with elevated creatine kinase, etc.). These patients are at high risk, and will be scheduled for angioplasty, and are put on anti-platelet drugs whilst they are waiting. However, as mentioned anti-platelet drugs are not without risks and stratification of use may be informed by the platelet phenotype as diagnosed by a method of the invention.

"Normal platelet activation" may be taken as a known value for normal platelets, or by performing the method on the platelet sample in parallel with a sample of normal platelets.

Methods of the invention may also be used in screening assays to identify compounds (for example from compound libraries) that are agonists or antagonists of platelet activation, or that enhance platelet activation. Such compounds may have therapeutic use for the treatment of abnormal blood conditions associated with an increased risk of bfeeding or thrombosis.

According to the invention there is provided a screening assay to screen for an antagonist of platelet activation, which comprises: stimulating platelets in a platelet sample; contacting the platelets with a candidate antagonist; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate antagonist if platelet activation is antagonised by the candidate antagonist. The platelets should be stimulated with a known platelet agonist (for example, any of the known agonists listed above). The candidate antagonist may be contacted with the platelets before, during, or after stimulation of the platelets, but preferably before or during stimulation.

According to the invention there is also provided a screening assay to screen for an agonist of platelet activation, which comprises: contacting platelets in a platelet sample with a candidate agonist; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate agonist if platelet activation is agonised by the candidate agonist.

There is further provided according to the invention a screening assay to screen for a compound that enhances platelet activation, which comprises: stimulating platelets in a platelet sample to cause activation of the platelets; contacting the platelets with a candidate enhancer of platelet activation; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate enhancer if platelet activation is enhanced by the candidate enhancer.

The candidate enhancer may be contacted with the platelets before, during, or after stimulation of the platelets, but preferably before or during stimulation.

Methods of the invention may also be used to screen for defective platelets (either hypo- or hyper-reactive platelets).

According to the invention there is provided a screening assay to screen for defective platelets, which comprises: stimulating platelets in a plurality of different platelet' samples; capturing activated platelets from the platelet samples in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying the platelet sample as containing defective platelets if platelet activation is reduced or enhanced compared to normal platelet activation.

Defective platelets identified by a screening assay of the invention may be further investigated to identify risk genes for cardiovascular diseases, in particular genes associated with an increased risk of bleeding or thrombosis.

Methods of the invention may readily be used in a high-throughput format to determine simultaneously the effect of several different concentrations of agonists or antagonists on platelet activation, or of several different agonists or antagonists on platelet activation, or to screen simultaneously several different candidate compounds to identify agonists, antagonists, or enhancers of platelet activation.

In preferred embodiments of the invention, methods of the invention are carried out using a microtitre plate in which a monoclonal antibody used to capture activated platelets is immobilised to each well of the microtitre plate. Such methods only require small sample volumes per well, and allow high throughput.

According to the invention there is also provided a kit for carrying out a method of the invention.

Thus, there is provided according to the invention a kit for assessing platelet activation, which comprises: a platelet agonist for stimulating platelets in a platelet sample to cause activation of the platelets; a capturing agent for capturing activated platelets from the platelet sample in preference to resting platelets; and a detecting agent for detecting captured activated platelets.

A kit of the invention may further comprise all reagents necessary to determine quantitatively the amount of captured activated platelets, for example labels or reagents required for controls or comparisons.

A kit of the invention may further comprise a platelet antagonist.

A kit of the invention may further comprise a solid phase to which the capturing agent is immobilised. Preferably, the solid phase comprises a plurality of sample wells or chambers, each sample well or chamber for receiving a portion of the platelet sample, and the capturing agent being immobilised to a wall of each sample well or chamber. In one embodiment, each sample well/chamber may comprise a different concentration of the same platelet agonist In another embodiment, each sample well/chamber may comprise a different concentration of the same platelet antagonist (optionally additionally with a platelet agonist: preferably the same concentration of platelet agonist in each sample well). In a further embodiment, each sample well/chamber may comprise a different agonist or a different antagonist. In a further embodiment, each sample well/chamber may comprise a different candidate agonist or enhancer of platelet activation, or a different candidate antagonist of platelet activation. The different candidate agonists, enhancers, or antagonists may be different compounds of a compound library which is to be screened for the presence of agonists, enhancers, or antagonists of platelet activation.

A preferred example of a solid phase comprising a plurality of sample wells is a microtitre plate. Kits comprising a solid phase with a plurality of sample wells/chambers may be used for high-throughput screening assays in accordance with the invention. Embodiments of the invention are described in the example below with reference to the accompanying figures in which:

Figure 1 shows ADP-induced binding of platelets to 16N7C2 monoclonal antibody (moAb) (anti-GPllb/llla). (A-C) 5 μg/ml 16N7C2 was coated in a 96-well microtitre plate overnight at 4°C and subsequently incubated with platelets (PRP) in the presence of 0-20 μM ADP. Platelet binding was detected by addition of biotinylated-6B4 moAb (anti-GP!bα)/streptadivin- HRP. Sigmoidal regression fit was applied to the curves using ORIGIN software. Each curve represents data from one donor prepared in duplicate (Mean ± SE). (D) Binding of antibody 16N7C2 to resting platelets (grey area) and platelets stimulated with 20 μM ADP (black line) analyzed by flow cytometry. (E) An unrelated mouse monoclonal IgGI (MOPC-21) was included as an inert control;

Figure 2 shows TRAP (thrombin receptor agonist peptide)-induced binding of platelets to 16N7C2 moAb (anti-GPllb/llla). (A-C) 5 μg/ml 16N7C2 antibody was coated in a 96-weiI microtitre plate overnight at 4°C and subsequently incubated with platelets (PRP) in the presence of 0-40 μM TRAP. Platelet binding was detected by addition of biotinylated-6B4 moAb (anti-GP!bα)/streptadivin-HRP. Sigmoidal regression fit was applied to the curves using ORIGIN software. Each curve represents data from one donor prepared in duplicate (Mean ± SE);

Figure 3 shows U46619- and collagen-induced binding of platelets to

16N7C2 moAb (anti-GPllb/llla). 5 μg/ml 16N7C2 antibody was coated in a

96-well microtitre plate overnight at 4°C and subsequently incubated wjth platelets (PRP) in the presence of 0-30 μM U46619 (A) or 0-150 μg/ml collagen (B). Platelet binding was detected by addition of biotinylated-6B4 moAb (anti-GPlbα)/streptadivin-HRP. Sigmoidal regression fit was applied to the curves using ORIGIN software. Each curve represents data from one donor prepared in duplicate (Mean ± SE);

Figure 4 shows ADP-induced binding of platelets to SZ51 moAb (anti-P- sefectin). (A) 5 μg/ml SZ51 antibody was coated in a 96-well microtitre plate overnight at 4°C and subsequently incubated with platelets (PRP) in the presence of 0-20 μM ADP. Platelet binding was detected by addition of biotinylated-6B4 moAb (anti-GPIbα)/streptadivin. Sigmoidal regression fit was applied to the curves using ORIGIN software. Each curve represents data from one donor prepared in duplicate (Mean ± SE). (B) Binding of SZ51 antibody to resting platelets (grey area) and platelets stimulated with 20 μM ADP (black line) analyzed by flow cytometry. For control, cf. Figure 1D;

Figure 5 shows inhibition of binding of platelets to 16N7C2 moAb (anti- GPIlb/Illa).(A) 5 μg/ml 16N7C2 antibody was coated in a 96-well microtitre plate overnight at 4°C and subsequently incubated with platelets (PRP) in the presence of 0-20 μM ADP. PRP from the same donor was isolated before (closed circles) and 1 hour after aspirin intake (open triangles). Platelet binding was detected by addition of biotinylated-6B4 moAb (anti- GPIbα)/streptadivin. (B) PRP was incubated with 10 μM ADP with variable amount of antagonist SNAP 0-2 mM. Platelet binding was revealed as described above.

Example

Development of a High-Throughput Assay to Study Platelet Function

Since platelets play an essential role in AT-related diseases (i.e. cardiovascular disease (CVD)) they remain a main target for therapies. The "Bloodomics" project aims to identify risk genes involved in AT and coronary artery disease (CAD) to design better antithrombotics for the prevention and treatment of CVD. As part of this project, we have developed a sensitive high-throughput assay to study platelet function in which capturing monoclonal antibodies (moAbs) are used that are specific for activated platelets.

Platelet-rich plasma (PRP) was prepared from blood from healthy donors using citrate as anticoagulant, and platelets were stimulated with different agonists, Including ADP, TRAP (thrombin receptor agonist peptide), U46619 (thromboxane A2 analogue) or collagen. Levels of platelet activation were assessed using a quantitative sandwich ELISA where moAbs 16N7C2 or SZ51 raised against platelet receptors GPIIb/llla and P- selectin, respectively, were used to capture activated platelets. The biotinylated mouse anti-GPIbα moAb 6B4 was used to reveal the presence of platelets. GPIIb/llla and P-selectin dependent signals were significantly 1.5-3-times higher when platelets were stimulated as compared with resting platelets. Prior incubation of platelets with the nitric oxide donor SNAP inhibited the capture via GPIIb/llla or P-selectin induced by sub- maximal concentrations of agonists. In addition, platelets taken one hour after aspirin intake could no (onger be activated.

We believe that we herewith have developed a platelet function test that (i) is more readily applicable than the current aggregation test, (ii) allows multiple agonists and antagonists at different concentrations to be studied simultaneously and (iii) should not only allow easy detection of platelets with reduced reactivity but also, and importantly, with an increased reactivity, which so far has been especially difficult to determine.

Introduction

Platelets play a critical role in normal haemostasis and thrombosis. Under normal conditions, platelets do not interact with healthy blood vessels.

Upon vascular damage, the subendothelial proteins such as collagen become exposed, allowing platelet adhesion to the damaged vessel. In the high shear arterial system, platelets slow down by interacting reversibly to collagen-bound von Willebrand Factor (VWF) via the platelet receptor GPIb/V/IX, a process also known as "rolling". The mechanisms by which GPIbα generates signals remain controversial. Some studies suggest signal transduction occurs through interaction of proteins (14-3-3ζ, PI3- kinase, Src kinase, calmodulin) bound to the cytoplasmic tail of the GPIbMIX complex \ others through direct association of surface molecules (FcR γ-chain and FcRγlla) ^2"5, while others implicate indirect mechanisms such as ADP release ⁶ and thromboxane A2 (TXA₂) generation ^7>8. It is proposed that the interaction between VWF and GPIbα generates calcium spikes leading to a weak activation of GPIlb/IIIa. Intracellular calcium mobilization triggers extracellular influx which amplify GPIlb/IIIa activation, resulting in firm platelet adhesion ⁹. At lower shear rate, or once the platelets have slowed down on VWF-collagen, platelets firmly adhere to collagen via the main collagen receptors GPVl and integrin α₂βi, triggering a critical signalling that results in platelet activation. The GPVI receptor acts through tyrosine kinases (Fyn and Lyn) that phosphorylate ITAM domains on FcRγ-chain, which in turn activate Syk, triggering an activation cascade involving adaptor proteins LAT and SLP- 76. These proteins subsequently recruit various signalling molecules, including phospholipase Cγ (PLCγ) and phosphoinositol 3-kinase (P13- kinase)⁹. Activation of PLC leads to the activation of protein kinase C (PKC) responsible for platelet granular secretion containing soluble agonists such as ADP, ATP, serotonin and GAS6 that will lead to further activation of platelets and their recruitment. Another important agonist is TXA₂ that is formed upon activation of phosphoϋpase A2 (PLA₂) ¹⁰. Finally, thrombin is one of the most potent activators of platelets via its receptors: protease-activated receptors (PAR) and GPIbMlX^11"14. Secreted agonists mentioned above except ATP and GAS6 cause platelet activation through G-protein coupled receptors (GPCRs). The GPCR Gq family (PAR1 and PAR4 for thrombin, TPa for TXA₂ and P2Yi for ADP) activates PLCβ, The GPCR Gi family (PAR 1, PAR4, P2Yi₂ for ADP) inhibits adenyl cyclase and/or activates MAP-kinases and the GPCR G12/G13 family (PAR1 , PAR4, TPa) signals to small GTPases ¹⁰. The end point of platelet activation is aggregation in which integrin GPIIb/llla plays an important role by binding fibrinogen upon activation ¹⁵.

As part of the "Bloodomics project" that aims to identify putative genome sequence variations in genes or non-coding genome sequence involved in AT and CAD, we seek to develop a high-throughput assay testing platelet responses to the key endogenous platelet agonists described above. Currently available tests are time-consuming, vary in sensitivity and are not high-throughput. The assay described here allows simultaneous and reproducible screening of different signalling pathways of platelet activation such as tyrosine kinases and G-protein dependent pathways and is high-throughput. Although these are preliminary results as all the conditions have not been optimized for all four agonists, it will be of great interest to test this assay on a larger population of individuals in order to determine normal dose ranges for the agonists tested herein, and discriminate between individuals with reduced and increase risk of AT respectively.

Materials and Methods

Plasma from Healthy Individuals

Blood was drawn from healthy volunteers on 3.13 % sodium citrate (9:1 V/V) by free flow. Immediately thereafter, platelet rich plasma (PRP) was prepared by centrifugation at 1,000 rpm for 10 min. Platelet count was evaluated with cell-Dyn 1300 (Abbott-laboratories, IL) and adjusted to 200,000 cells/μl with autologous platelet poor plasma (PPP). Care was taken not to disturb the interface when aspirating the PRP. PPP was prepared by centrifugation at 13,400 rpm for 6 minutes. When indicated, blood was drawn from a healthy individual before and 1 hour after aspirin intake. PRP was prepared as indicated above. Monoclonal Antibodies (MoAbs)

MoAbs anti-GPlbα 6B4 and anti-GPIIb/llla 16N7C2, used in this study were produced and characterized as previously described ¹⁶'¹⁷. The moAb SZ51 binds CD62P on activated platelets and was a generous gift from Dr.

Ruan et al. ¹⁸. MoAb 6B4 was biotinylated using NHS LC Biotin

(sulfosuccinimidyl 6-(biotinamido)hexanoate), according to the manufacturer's instructions (Pierce, Rockford, IL). MOPC-21 was purchased from Sigma (St Louis, MO). Detection of unlabelled antibodies MOPC21, 16N7C2 and SZ51 in flow cytometry was by a goat anti-mouse

IgG-phycoerythrin (PE) (Jackson lmmunoresearch Laboratories).

Activation pathway-EUSA

96-weil microtitre plates (Greiner, Frickenhausen, Germany) were coated overnight at 4° C with 100 μl/well moAb 16N7C2, or SZ51 (5 μg/ml in PBS). Plates were blocked with 3 % milk powder (250 μl/well) for 2 h at RT. Agonists: ADP (Sigma), TRAP (Calbiochem), TXA₂ analogue U46619 (Caibiochem) and Horm collagen (Nycomed, Munich, Germany) were prepared in 25 μl by a serial dilution in PBS with starting concentrations of 20 μM, 40 μM, 30 μM and 150 μg/ml, respectively. The antagonist S- nitroso-N-acetyl penicillamine (SNAP; Calbiochem) at 2 mM was prepared in 25 μl by a serial, dilution in PBS with a constant concentration of ADP (10 μM). Immediately after the agonists/antagonists dilutions, 75 μl of PRP was added to each well and incubated at room temperature for 30 minutes. Every condition was tested in duplicate. Biotinylated moAb 6B4 was used as a primary antibody for platelet detection at 1 μg/ml in PBS- 0,3% milk powder for 1 hour at room temperature. MoAb 6B4 binding to captured platelets was revealed with streptavidin coupled with horse radish peroxidase (Roche, Mannheim, Germany) (1/5000 in PBS, 0.3% milk, 1 h at RT). The colour reactions were initiated by adding H₂O₂ and orthophenylenediamine (OPD; Sigma) and stopped with 4M H₂SO₄. Platelet binding was evaluated by measuring the absorbance at 490 nm.on a mϊcrotitre plate reader (EL340, Bio-Tek instruments). After each incubation, a washing step was performed three times with PBS.

FACS analysis Platelets (5 x 10⁶) were diluted in 200 μl PBS and incubated with 10 μM ADP, for 10 minutes at room temperature. At the end of the incubation, 1/5 of the reaction mixtures were first incubated for 15 minutes at RT with moAbs 16N7C2, SZ51 or MOPC-21. A secondary antibody PE-conjugated goat anti-mouse IgG was then added for 15 min before fixation with 0,2 % formyl saline (0,9 % NaCI, 0,2% formaldehyde). Samples were then read on a FACScan flow cytometer (EPICS^® XL-MCL, Coulter). A gate analysis on 10,000 events was performed based on forward and side scatter to exclude platelet aggregates and microparticles.

Results

The binding of resting and agonist-activated platelets to moAbs raised against GPI!b/llla or P-selectin was studied in an ELISA, as described in

Materials and Methods. The moAb16N7C2, was the first anti-GPIIb/llla that contains an echistatin-like RGD sequence in one of its complimentarity determining regions (CDRs) (GDR3 of the heavy chain variable domain) ^π and with a high affinity for its antigen (GPllb/llia). The moAb SZ51 ¹⁸ has been raised against P-selectin (CD62P) present in the alpha granules of resting platelets ¹⁹. We have selected to capture platelets with 16N7C2 and SZ51 because during platelet activation, there is an up-regulation of these two receptors¹⁹'²⁰ that could increase the number of antibody-bound platelets in our ELISA test. The moAb 6B4 targets the region amino acids 230-240 on the alpha subunit of the , heterodimeric GPIbα platelet receptor. It is a well described inhibitor of the GPIbα-VWF interaction¹⁶. MoAb 6B4 was chosen to reveal the binding of platelets as it does not differentiate between resting and activated states, and furthermore GPIbα is present in fairly high copy number on the platelet surface (25,000 copies) ²¹. Activation of platelets with various agonists led to an increase in platelet binding to the anti-GPilb/IIIa moAb as illustrated in Figures 1-3. Plotting the OD (490 nm) versus the concentration of ADP resulted in binding curves that could be best fitted by sigmoidal regression (Figure 1A-B)₁ although one curve could be best fitted through linear regression (Figure 1 C). Out of 7 donors tested, each could be categorized in one of these groups. Each donor had his/her platelets in an activated state when they were stimulated by 20 μM ADP, however the minimal amount required for activation to be able to differentiate between resting and activated platelets varied between the groups: 2.5 μM, 10 μM and 1.25 μM in group 1 (Figure 1A), group 2 (Figure 1 B) and group 3 (Figure 1 C)₁ respectively. To validate our results, we tested if we could obtain an increase in 16N7C2 binding to ADP-activated platelets in another assay, by flow cytometry. As expected, 16N7C2 bound to resting platelets but more interestingly, moAb 16N7C2 showed a 1.8 (based on mean fluorescence) increase in binding to platelets stimulated with ADP (Figure 1D-E), correlating with the results obtained in our ELiSA test (Figure 1A).

in line with what was obtained with ADP, the binding curves of platelets activated with TRAP could be best fitted through sigmoidal fit, although out of four donors tested, one of them did not respond to TRAP (Figure 2C). Maximal activation of platelets with TRAP was achieved with 20 μM (Figure 2A), or 40 μM (Figure 2B).

All donors tested followed the same pattern of platelet activation by U46619 through GPllb/llla as the one presented here (Figure 3A)₁ while for one donor no difference could be observed between resting and U46619-activated platelets (data not shown). U46619 induced approximately a two-fold increase in binding through GPllb/llla (Figure 3A). Finally, an example of binding curve of platelets activated with collagen is presented in Figure 3B, however here the difference between resting and activated states was less obvious.

An increase in platelet binding to the anti-P-selectin moAb SZ51 was detected for platelets stimulated with ADP (Figure 4A) resembling the binding curve of ADP-stimulated platelets to 16N7C2 (Figure 1A). This increase was also observed in flow cytometry as presented in Figure 4B albeit with a ratio of binding superior to the one observed in ELISA (3.34 vs 2.2).

To determine the specificity of the test and to further explore its applicability, inhibition of platelet activation by aspirin and SNAP were carried out (Figure 5). The binding to moAb 16N7C2 of pfatelets from a donor that had taken aspirin 1 hour before the test was nearly abolished (Figure 5A). An increase in binding was observed upon activation with ADP although levels did not reach those observed for resting platelets prior aspirin uptake (Figure 5A). Binding could also be prevented by addition of SNAP along with the agonist ADP, in a dose-dependent manner (Figure 5B),

Discussion

Determination of platelet activation classically is done by using platelet aggregometry, originally described by Born²² , which has proven its usefulness during the years, as it rather closely mimics the physiological process where one (or more) stimulus induces a whole activation cascade that finally results in GPIIb/llla mediated platelet aggregation. However, this technique suffers from serious drawbacks: although the results are obtained "in real time", it takes around 5 minutes to have a full aggregation, which in view of the relatively short active ex vivo life time of platelets, precludes study of large numbers of (ant)agonists at different concentrations, unless a battery of aggregometers is used in parallel. The reproducibility of the test is not optimal, the interpretation is not always straightforward (which parameter: lag time, slope, maximal amplitude), and depending on the machine used 250-1000 μl of PRP is needed per test set. In addition, in contrast to the obvious detection of hypo-active platelets (e.g. Glanzmann thrombasthenia, Bernard Soulier syndrome, or due to platelet inhibitors) it has so far been particularly hard to determine 'hyper'-active platelets, to an extent that it is not even clear at present whether such platelets exist.

Other parameters, such as secretion of ATP (luminescence) or appearance of platelet activation markers (activated integrins, expression of granular proteins such as P-selectin or GMP-140 and lysosome integral membrane protein-CD63 ²³) detectable by amongst others FACS-analysis, have been put forward as alternatives for the aggregation tracings, which to some extent may still suffer from a lack of 'high-throughput' capacity.

We here designed a technology that would overcome most of the above mentioned problems by using an ELlSA setup where activated platelets are captured by coating antibodies that specifically distinguish between resting and activated platelet characteristics, and next quantified by an antibody that is invariant to changes in the platelet activation state.

We here provide proof-of-principle by using two antibodies: one against GPIlb/llla, of which both the numbers and the conformation changes upon platelet activation, and one against P-selectin, a transmembrane protein that in resting platelets is situated in the membrane of the α-granules and becomes surface exposed upon platelet secretion. In this respect, this could be surrogate endpoints for aggregation and secretion, respectively. We anticipate that capture with other antibodies such as PAC-1, IAC-1 , specific for activated integrins GPIlb/llla and GPIa/lla respectively, or any anti-P-selectin antibody might work equally well. The data presented here are intended to provide proof-of-principle and at present do not allow an estimate of the clinical value of the test, for which extended studies on platelets from normal controls and patients at risk are needed. However, the finding that with the four agonists applied, platelet activation can readily be detected, except in a couple of individual cases which need further scrutiny, together with the clear effect of aspirin intake and of increasing concentrations of SNAP, a NO-donor, indicate that the test is detecting relevant parameters.

The ultimate design of the test could consist in one ELISA plate per donor/patient, for which around 20 ml blood would be required, which then would be sufficient to study 8 concentrations of 4 agonists and of 1 antagonist, plus controls, in duplicate. The platelets are studied simultaneously such that variability due to different handling times is avoided, the data are produced in a digital form, making interpretation more straightforward, as the plates containing coated capturing antibody and the different concentrations of (ant)agonists can be prepared on beforehand and stored frozen, the remainder of the test can readily be automated, and finally, and potentially the more important advantage could well be that the possibility to detect hyper-active platelets, i.e. reacting at lower than on average required agonist concentrations, is built-in.

We believe that we, with the present test have devised a technique that should provide a valuable addition to the currently available tests, and which with its high-throughput potential might be useful in both studies of large control and patient cohorts as required in functional genomic/proteomic approaches, as well as in screening tests for new antiplatelet agents. Reference List

(1) Berndt MC₁ Shen Y, Dopheide SM₁ Gardiner EE, Andrews RK. The vascular biology of the glycoprotein Ib-IX-V complex. Thromb Haemost. 2001 ;86: 178-188.

(2) Falati S, Edmead CE, Poole AW. Glycoprotein Ib-V-IX, a receptor for von Willebrand factor, couples physically and functionally to the Fc receptor gamma-chain, Fyn, and Lyn to activate human platelets. Blood. 1999;94:1648-1656.

(3) Wu Y, Suzuki-lnoue K, Satoh K et al. Role of Fc receptor gamma- chain in platelet glycoprotein Ib-mediated signaling. Blood. 2001 ;97.-3836-3845.

(4) Sullam PM, Hyun WC, Szollosf J et al. Physical proximity and functional interplay of the glycoprotein Ib-IX-V complex and the Fc receptor FcgammaRIIA on the platelet plasma membrane. J Biol

Chem. 1998;273:5331-5336.

(5) Torti M, Bertoni A, Canobbio I et al. Rap1B and Rap2B translocation to the cytoskeleton by von Wiliebrand factor involves Fcgammall receptor-mediated protein tyrosine phosphorylation. J Biol Chem. 1999;274: 13690-13697.

(6) Mazzucato M, Pradella P, Cozzi MR, De Marco L, Ruggeri ZM. Sequential cytoplasmic calcium signals in a 2-stage platelet activation process induced by the glycoprotein lbalpha mechanoreceptor. Blood. 2002; 100:2793-2800.

(7) Canobbio I, Bertoni A, Lova P et af. Platelet activation by von Willebrand factor requires coordinated signaling through thromboxane A2 and Fc gamma NA receptor. J Biol Chem. 2001 ;276:26022-26029. (8) Kroll MH, Harris TS, Moake JL, Handin Rl, Schafer Al. von Willebrand factor binding to platelet GpIb initiates signals for platelet activation. J Clin Invest. 1991 ;88:1568-1573.

(9) Jackson SP₁ Nesbitt WS₁ Kulkarni S. Signaling events underlying thrombus formation. J Thromb Haemost. 2003;1 :1602-1612.

(10) Deckmyn H, Vanhoorelbeke K, UIrichts H et al. Amplification loops and signal transduction pathways in platelets. In: Arnout J₁ De Gaetano G, Hoylaerts M, Peerlinck K, Van Geet C₁ Verhaeghe R₁ eds. Leuven university Press; 2003:75-91.

(11) Kahn ML, Nakanishi-Matsui M, Shapiro MJ, lshihara H, Coughlin SR. Protease-activated receptors 1 and 4 mediate activation of human platelets by thrombin. Journal of Clinical Investigation. 1999;103:879-887.

(12) Vu TK, Hung DT, Wheaton Vl, Coughlin SR. Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell. 1991 ;64:1057-1068.

(13) Okumura T, Janπieson GA. Platelet Glycocalicin - Single Receptor for Platelet-Aggregation Induced by Thrombin Or Ristocetin. Thrombosis Research. 1976;8:701-706.

(14) Okumura T, Hasitz M, Jamieson GA. Platelet Glycocalicin - Interaction with Thrombin and Role As Thrombin Receptor of Platelet Surface. Journal of Biological Chemistry. 1978:253:3435- 3443.

(15) Deckmyn H, UIrichts H₁ Van De WG, Vanhoorelbeke K. Platelet antigens and their function. Vox Sang. 2004;87 Suppl 2:105-111.

(16) Cauwenberghs N, Meiring M, Vauterin S et al. Antithrombotic effect of platelet glycoprotein Ib-blocking monoclonal antibody Fab fragments in nonhuman primates. Arterioscler Thromb Vase Biol. 2000;20: 1347-1353.

(17) Deckmyn H, Stanssens P, Hoet B et al. An echistatin-fike Arg-Gly- Asp (RGD)-containing sequence in the heavy chain CDR3 of a murine monoclonal antibody that inhibits human platelet glycoprotein I !b/I!la function. BrJ Haematol. 1994;87:562-571.

(18) Wu G, Xi X, Li P, Chu X, Ruan C. Preparation of a monoclonal antibody, SZ-51, that recognizes an alpha-granule membrane protein (GMP-140) on the surface of activated human platelets. Nouv Rev Fr Hematol. 1990;32:231-235.

(19) Stenberg PE, McEver RP, Shuman MA, Jacques YV, Bainton DF. A platelet alpha-granule membrane protein (GMP-140) is expressed on the plasma membrane after activation. J Cell Biol.

1985;101 :880-886.

(20) Chung AW, Jurasz P₁ Hollenberg MD, Radomski MW. Mechanisms of action of proteinase-activated receptor agonists on human platelets. Br J Pharmacol. 2002;135:1123-1132.

(21) Lopez JA. The platelet glycoprotein Ib-IX complex. Blood Coagul Fibrinolysis. 1994;5:97-119.

(22) Born GV Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature. 1962 Jun 9; 194:927-9.

(23) Metzelaar MJ, Korteweg J, Sixma JJ, Nieuwenhuis HK. Comparison of platelet membrane markers for the detection of platelet activation in vitro and during platelet storage and cardiopulmonary bypass surgery. J Lab Clin Med. 1993 Apr;121(4):579-87

Claims

1. A method for assessing platelet activation, which comprises: stimυiating platelets in a platelet sample to cause activation of the platelets; capturing activated platelets from the platelet sample in preference to resting platelets; and detecting the captured activated platelets.

2. A method according to claim 1, wherein activated platelets are captured using a monoclonal antibody that binds selectively to activated platelets.

3. A method according to claim 2, wherein the monoclonal antibody binds selectively to glycoprotein llb/llla or glycoprotein la/Ma of activated platelets, or to CD62P of activated platelets.

4. A method according to claim 2 or 3, wherein the monoclonal antibody is immobilised to a solid phase.

5. A method according to any preceding claim, wherein the captured activated platelets are detected using a monoclonal antibody that binds selectively to platelets, but does not differentiate between resting and activated platelets.

6. A method according to any preceding claim, wherein the amount of captured activated platelets is quantitatively determined.

7. A method according to any preceding claim which is a quantitative sandwich enzyme-linked immunosorbent assay (ELISA), radioimmunossay (RIA), immunoradiometric assay (IRMA), fluorescent immunoassay (FIA), chemiluminescent immunoassay (CLIA), or mirror resonance assay.

8. A method according to any preceding claim, wherein the platelet sample comprises a plurality of sample portions, and the platelets of each sample portion are stimulated with a different concentration of the same platelet agonist.

9. A method according to claim 8, which further comprises generating a dose-response curve of platelet activation to concentration of platelet agonist.

10. A method according to any of claims 1 to 7, which further comprises contacting the platelets with an antagonist of platelet activation.

11. A method according to claim 10, wherein the platelet sample comprises a plurality of sample portions, and the platelets of each sample portion are contacted with a different concentration of the same antagonist.

12. A method according to claim 11, which further comprises generating a dose-response curve of platelet activation to concentration of antagonist.

13. A method of determining whether a subject has, or is at risk of having, an abnormal blood condition associated with an increased risk of bleeding, which comprises carrying out a method according to any of claims 1 to 11 using a platelet sample obtained from the subject, and determining whether activation of platelets in the sample is reduced compared to normal platelet activation.

14. A method of determining whether a subject has, or is at risk of having, an abnormal blood condition associated with an increased risk of thrombosis, which comprises carrying out a method according to any of claims 1 to 11 using a platelet sample obtained from the subject, and determining whether activation of platelets in the sample is increased compared to normal platelet activation.

15. A screening assay to screen for an antagonist of platelet activation, which comprises: stimulating platelets in a platelet sample; contacting the platelets with a candidate antagonist; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate antagonist if platelet activation is antagonised by the candidate antagonist.

16. A screening assay to screen for an agonist of platelet activation, which comprises: contacting platelets in a platelet sample with a candidate agonist; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate agonist if platelet activation is agonised by the candidate agonist.

17. A screening assay to screen for a compound that enhances platelet activation, which comprises: stimulating platelets in a platelet sample to cause activation of the platelets; contacting the platelets with a candidate enhancer of platelet activation; capturing activated platelets from the platelet sample in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying and/or isolating the candidate enhancer if platelet activation is enhanced by the candidate enhancer.

18. A screening assay to screen for defective^' platelets, which comprises: stimulating platelets in a plurality of different platelet samples; capturing activated platelets from the platelet samples in preference to resting platelets; detecting the captured activated platelets to assess platelet activation; and identifying the platelet sample as containing defective platelets if platelet activation is reduced or enhanced compared to normal platelet activation.

19. A kit for assessing platelet activation, which comprises: a platelet agonist for stimulating platelets in a platelet sample to cause activation of the platelets; a capturing agent for capturing activated platelets from the platelet sample in preference to resting platelets; and a detecting agent for detecting captured activated platelets.

20. A kit according to claim 19, wherein the capturing agent is immobilised to a solid phase comprising a plurality of sample wells or chambers, each sample well or chamber for receiving a portion of the platelet sample, and the capturing agent being immobilised to a wall of each sample well or chamber.

21. A kit according to claim 20, wherein each sample well or chamber comprises a different concentration of the same platelet agonist, or a different concentration of the same platelet antagonist.