R. W. Farndale

The role of collagen in thrombosis and hemostasis

Collagens represent up to 40% of the total protein of the vessel wall, forming an insoluble scaffold which is essential for tissue integrity and which provides a surface for the attachment of other matrix constituents and for the adhesion of vascular cells. Thus collagens exercise important regulatory functions within the vasculature. At least 25 different types of collagen exist [1], a number of which, including major, widely distributed types I, III, IV, V and VI, occur in the vessel wall [2]. Three polypeptide chains (a-chains) form the fundamental structure of the collagen molecule, which is characterized by the presence of one or more triple-helical domains. Within these domains, the three a-chains, which are different gene products depending on collagen type, wind around one another in a characteristic lefthanded triple helix (Fig. 1). This conformation of collagen is of crucial importance, since the three-dimensional structure is needed for recognition of collagen by its ligands. The primary structure of the triple-helical regions of collagen chains is characterized by its repetitive Gly–X–Y sequence, with Gly– Pro–Hyp (GPO) as the most frequent, forming about 10% of the primary structure of collagen types I and III [3]. The presence of glycine as every third residue is essential for the close packing of the chains. The high content of the imino acids proline and hydroxyproline is responsible for hydrogen bonds which hold the three chains of the helix together, and which stabilize helix–helix interactions within the polymeric structure of collagen. Collagens can be divided into fibrous and non-fibrous types. Triple-helical monomers of the fibrous collagens, types I–III, V and XI, self-associate to form typical highly ordered tissue collagen fibers, cross-striated with a periodicity of 67 nm, as observed by electron microscopy (Fig. 1). In vivo, mixed fibers may be assembled from various types of collagens. In contrast, non-fibrous collagens form other higher order structures. For example, type IV collagen has a modular structure composed of triple-helical segments separated by globular domains. A two-dimensional network forms, stabilized via lateral interaction between the globular domains, and forming an important component of basement membranes. Collagens, located in the matrix underlying vascular endothelial cells, are not exposed to flowing blood. After injury, however, blood will flow directly over subendothelial structures including connective tissue that contains a high percentage of collagen. Thus collagen is ideally situated to initiate hemostasis. There is ample evidence that collagen is one of the major activators of the platelet response after injury. Collagen is the only matrix protein which supports both platelet adhesion and complete activation. When collagen becomes exposed to flowing blood, platelets rapidly adhere, spread, become activated and begin to form an aggregate. Figure 2 shows the intimate contact between aggregating platelets and collagen, and Fig. 3, the extent of platelet activation and aggregate formation on collagen, compared with the adhesion only that occurs on fibrinogen-coated surfaces. The early concept that collagen was also directly responsible for the activation of the intrinsic coagulation cascade [4] turned out to be incorrect [5]. However, collagen indirectly plays a crucial role in regulating thrombin formation because negatively charged phospholipids such as phosphatidylserine become exposed on the surface of platelets after interaction with collagen and form the catalytic surface for the assembly of active coagulation complexes and thrombin generation [6]. There is additional evidence (reviewed below) that collagen participates in other ways in the regulation of the coagulation cascade. Here we will review the role of collagen in the regulation of hemostasis, embracing both coagulation and platelet aggregation.

Mapping the platelet profile for functional genomic studies and demonstration of the effect size of the GP6 locus

Summary.  Background: Evidence suggests the wide variation in platelet response within the population is genetically controlled. Unraveling the complex relationship between sequence variation and platelet phenotype requires accurate and reproducible measurement of platelet response. Objective: To develop a methodology suitable for measuring signaling pathway‐specific platelet phenotype, to use this to measure platelet response in a large cohort, and to demonstrate the effect size of sequence variation in a relevant model gene. Methods: Three established platelet assays were evaluated: mobilization of [Ca2+]i, aggregometry and flow cytometry, each in response to adenosine 5′‐diphosphate (ADP) or the glycoprotein (GP) VI‐specific crosslinked collagen‐related peptide (CRP). Flow cytometric measurement of fibrinogen binding and P‐selectin expression in response to a single, intermediate dose of each agonist gave the best combination of reproducibility and inter‐individual variability and was used to measure the platelet response in 506 healthy volunteers. Pathway specificity was ensured by blocking the main subsidiary signaling pathways. Results: Individuals were identified who were hypo‐ or hyper‐responders for both pathways, or who had differential responses to the two agonists, or between outcomes. 89 individuals, retested three months later using the same methodology, showed high concordance between the two visits in all four assays (r2 = 0.872, 0.868, 0.766 and 0.549); all subjects retaining their phenotype at recall. The effect of sequence variation at the GP6 locus accounted for ∼35% of the variation in the CRP‐XL response. Conclusion: Genotyping‐phenotype association studies in a well‐characterized, large cohort provides a powerful strategy to measure the effect of sequence variation in genes regulating the platelet response.

Primary and secondary agonists can use P2X(1) receptors as a major pathway to increase intracellular Ca(2+) in the human platelet.

Summary.  In the platelet, it is well established that many G‐protein‐ and tyrosine kinase‐coupled receptors stimulate phospholipase‐C‐dependent Ca2+ mobilization; however, the extent to which secondary activation of adenosine 5′‐triphosphate (ATP)‐gated P2X1 receptors contributes to intracellular Ca2+ responses remains unclear. We now show that selective inhibition of P2X1 receptors substantially reduces the [Ca2+]i increase evoked by several important agonists in human platelets; for collagen, thromboxane A2, thrombin, and adenosine 5′‐diphoshate (ADP) the maximal effect was a reduction to 18%, 34%, 52%, and 69% of control, respectively. The direct contribution of P2X1 to the secondary Ca2+ response was far greater than that of either P2Y receptors activated by co‐released ADP, or via synergistic P2X1:P2Y interactions. The relative contribution of P2X1 to the peak Ca2+ increase varied with the strength of the initial stimulus, being greater at low compared to high levels of stimulation for both glycoprotein VI and PAR‐1, whereas P2X1 contributed equally at both low and high levels of stimulation of thromboxane A2 receptors. In contrast, only strong stimulation of P2Y receptors resulted in significant P2X1 receptor activation. ATP release was detected by soluble luciferin:luciferase in response to all agonists that stimulated secondary P2X1 receptor activation. However, P2X1 receptors were stimulated earlier and to a greater extent than predicted from the average ATP release, which can be accounted for by a predominantly autocrine mechanism of activation. Given the central role of [Ca2+]i increases in platelet activation, these studies indicate that ATP should be considered alongside ADP and thromboxane A2 as a significant secondary platelet agonist.

Nitric oxide specifically inhibits integrin-mediated platelet adhesion and spreading on collagen

Summary.  Background: Nitric oxide (NO) inhibits platelet adhesion to collagen, although the precise molecular mechanisms underlying this process are unclear. Objectives: Collagen‐mediated adhesion is a multifaceted event requiring multiple receptors and platelet‐derived soluble agonists. We investigated the influence of NO on these processes. Results: S‐nitrosoglutathione (GSNO) induced a concentration‐dependent inhibition of platelet adhesion to immobilized collagen. Maximal adhesion to collagen required platelet‐derived ADP and TxA2. GSNO‐mediated inhibition was lost in the presence of apyrase and indomethacin, suggesting that NO reduced the availability of, or signaling by, ADP and TxA2. Exogenous ADP, but not the TxA2 analogue U46619, reversed the inhibitory actions of GSNO on adhesion. Under adhesive conditions NO inhibited dense granule secretion but did not influence TxA2 generation. These data indicated that NO may block signaling by TxA2 required for dense granule secretion, thereby reducing the availability of ADP. Indeed, we found TxA2‐mediated activation of PKC was required to drive dense granule secretion, a pathway that was inhibited by NO. Because our data demonstrated that NO only inhibited the activation‐dependent component of adhesion, we investigated the effects of NO on individual collagen receptors. GSNO inhibited platelet adhesion and spreading on α2β1 specific peptide ligand GFOGER. In contrast, GSNO did not inhibit GPVI‐mediated adhesion to collagen, or adhesion to the GPVI specific ligand, collagen related peptide (CRP). Conclusions: NO targets activation‐dependent adhesion mediated by α2β1, possibly by reducing bioavailability of platelet‐derived ADP, but has no effect on activation‐independent adhesion mediated by GPVI. Thus, NO regulates platelet spreading and stable adhesion to collagen.

Collagen surfaces to measure thrombus formation under flow: possibilities for standardization

Arterial thrombus formation is initiated by flow-dependent interactions of blood platelets and plasma components with the damaged vessel wall. In the last three decades, flow chamberbased methods have become increasingly popular for studying the process of thrombus formation in isolated whole blood, flowing at a predefined shear rate. Such devices containing immobilized collagens or other thrombogenic substances have yielded invaluable information on the mechanisms that regulate platelet adhesion and activation under flow, as well as on the role of the coagulation system in this setting. New developments are the use of small parallel-plate flow chambers, miniaturized variants of flow devices and small glass microcapillaries for measuring thrombus buildup in small volumes of blood within short intervals of time (< 10 min). The research is now at a stage that flow chamber assays can be developed for (pre)clinical screening of blood samples from patients. However, a still problematic factor is the precise control of the properties of the surface that mediates the formation of platelet thrombi. Earlier reports of the Biorheology Subcommittee of the SSC of the ISTH have provided overviews of the biorheological principles and the applications of flow-based assays [1,2]. These also describe the various methodologies and the various types of devices employed to measure thrombus formation. The present report focuses on the preparation and characterization of thrombogenic surfaces for flowbased assays. An inventory is made of the types of coating materials, of the procedures to coat flow chambers and of the factors influencing the activity of coated surfaces. Possibilities are discussed for optimized preparation of collagen-based coated surfaces. Finally, recommendations are given for standardization of the coating procedures for future use of flow devices in inter-laboratory comparisons and (pre)clinical trials. The full report is available on-line (see Data S1).

Gain- and loss-of-function mutants confirm the importance of apical residues to the primary interaction of human glycoprotein VI with collagen.

Summary.  Background: By site‐directed mutagenesis of recombinant receptor fragments, we have previously identified residue lysine59 of the platelet collagen receptor glycoprotein VI (GPVI) as being critical for its interaction with the synthetic ligand collagen‐related peptide (CRP) and the inhibitory phage antibody 10B12. Lysine59 is proposed to lie on the apical surface of the receptor near the linker joining the two immunoglobulin (Ig)‐like extracellular domains. Recently, others have postulated the involvement of a portion of the first domain distant from the interdomain hinge as being involved in an extended collagen‐binding site. Aim and Methods: To extend our knowledge of the primary collagen‐binding site of GPVI, a number of neighboring residues on the apical surface of recombinant soluble GPVI were mutated to alanine and binding of these mutants, as well as the lysine59 mutant, to fibrillar collagen was measured. Results: Binding of recombinant GPVI to collagen, like CRP, was dramatically reduced by the mutation of residue lysine59 to glutamate. Remarkably, the mutation of residues arginine60 in domain one and arginine166 in domain two, individually to alanine, which had no significant affect on CRP binding, reduced binding of recombinant GPVI to collagen. Mutation of the residue lysine41 to alanine dramatically increased binding to both CRP and collagen. This mutation abolished 10B12 binding, confirming its position in the epitope of our inhibitory phage antibody. Conclusions: Residues lysine59, arginine60, and arginine166, from both Ig‐like domains of GPVI, are critical for collagen binding by the receptor. This provides additional evidence for a basic patch on the apical surface of the receptor as the primary collagen‐binding site of GPVI.

A role for adhesion and degranulation-promoting adapter protein in collagen-induced platelet activation mediated via integrin α(2) β(1).

Summary.  Background: Collagen‐induced platelet activation is a key step in the development of arterial thrombosis via its interaction with the receptors glycoprotein (GP)VI and integrin α2β1. Adhesion and degranulation‐promoting adapter protein (ADAP) regulates αIIbβ3 in platelets and αLβ2 in T cells, and is phosphorylated in GPVI‐deficient platelets activated by collagen. Objectives: To determine whether ADAP plays a role in collagen‐induced platelet activation and in the regulation and function of α2β1. Methods: Using ADAP−/− mice and synthetic collagen peptides, we investigated the role of ADAP in platelet aggregation, adhesion, spreading, thromboxane synthesis, and tyrosine phosphorylation. Results and Conclusions: Platelet aggregation and phosphorylation of phospholipase Cγ2 induced by collagen were attenuated in ADAP−/− platelets. However, aggregation and signaling induced by collagen‐related peptide (CRP), a GPVI‐selective agonist, were largely unaffected. Platelet adhesion to CRP was also unaffected by ADAP deficiency. Adhesion to the α2β1‐selective ligand GFOGER and to a peptide (III‐04), which supports adhesion that is dependent on both GPVI and α2β1, was reduced in ADAP−/− platelets. An impedance‐based label‐free detection technique, which measures adhesion and spreading of platelets, indicated that, in the absence of ADAP, spreading on GFOGER was also reduced. This was confirmed with non‐fluorescent differential‐interference contrast microscopy, which revealed reduced filpodia formation in ADAP−/− platelets adherent to GFOGER. This indicates that ADAP plays a role in mediating platelet activation via the collagen‐binding integrin α2β1. In addition, we found that ADAP−/− mice, which are mildly thrombocytopenic, have enlarged spleens as compared with wild‐type animals. This may reflect increased removal of platelets from the circulation.

Collagen-mimetic peptides mediate flow-dependent thrombus formation by high- or low-affinity binding of integrin α2β1 and glycoprotein VI

Summary.  Background: Collagen acts as a potent surface for platelet adhesion and thrombus formation under conditions of blood flow. Studies using collagen‐derived triple‐helical peptides have identified the GXX’GER motif as an adhesive ligand for platelet integrin α2β1, and (GPO)n as a binding sequence for the signaling collagen receptor, glycoprotein VI (GPVI). Objective: The potency was investigated of triple‐helical peptides, consisting of GXX’GER sequences within (GPO)n or (GPP)n motifs, to support flow‐dependent thrombus formation. Results: At a high‐shear rate, immobilized peptides containing both the high‐affinity α2β1‐binding motif GFOGER and the (GPO)n motif supported platelet aggregation and procoagulant activity, even in the absence of von Willebrand factor (VWF). With peptides containing only one of these motifs, co‐immobilized VWF was needed for thrombus formation. The (GPO)n but not the (GPP)n sequence induced GPVI‐dependent platelet aggregation and procoagulant activity. Peptides with intermediate affinity (GLSGER, GMOGER) or low‐affinity (GASGER, GAOGER) α2β1‐binding motifs formed procoagulant thrombi only if both (GPO)n and VWF were present. At a low‐shear rate, immobilized peptides with high‐ or low‐affinity α2β1‐binding motifs mediated formation of thrombi with procoagulant platelets only in combination with (GPO)n. Conclusions: Triple‐helical peptides with specific receptor‐binding motifs mimic the properties of native collagen I in thrombus formation by binding to both platelet collagen receptors. At a high‐shear rate, either GPIb or high‐affinity (but not low‐affinity) GXX’GER mediates GPVI‐dependent formation of procoagulant thrombi. By extension, high‐affinity binding for α2β1 can control the overall platelet‐adhesive activity of native collagens.

Platelet glycoprotein VI as a mediator of metastasis

Platelets have previously been implicated in inflammatory processes, but there is a relatively short literature on their possible role in the progression of cancer. In this issue, Jain, Russell and Ware report that they injected metastatic cell lines into the tail vein of mice, and found that the development of metastases in the lung is impaired in platelet Glycoprotein VI knockout animals. Whilst the observed effect of ablation of the collagen receptor GPVI is partial, resulting in a halving of the number of metastases, this finding offers novel insight into the mechanisms underlying the recruitment of circulating tumor cells to susceptible tissues. How can the platelet exert a pro-metastatic effect? Several possibilities present themselves, for some of which there is experimental evidence. First, the activated platelet may release or generate products that enhance tumor cell survival in the circulation or recruitment to sites of metastasis. One thinks especially of stored platelet alpha granule components that are released upon platelet activation, such as PDGF that might be chemotactic for tumor cells [1] or increase their rate of proliferation. VEGF is similarly released [2] and is a key mediator of tumor angiogenesis. In the present study, however, there was no effect of GPVI knockout on primary tumor mass or its complement of microcapillaries, suggesting that VEGFmediated enhancement of tumor vascularity should be discounted. The activated platelet is also a rich source of newlysynthesized bioactive materials, including thromboxane A2 and thrombin generated on the activated platelet surface. Some of these products could activate the vessel wall leading to endothelial retraction and increased opportunity for invasion, or may cause vasoconstriction that contributes to the entrapment of circulating tumor cells. Thrombin and other coagulation pathway components have been suggested as mediators of metastasis [3]. Such an indirect, soluble mediator, model would require circulating platelets to become activated, for example by cytokines released by primary tumors, but such mechanisms need not require any direct interaction of tumor cells with the platelet. This is not the first such observation: a greater effect of complete platelet deficiency or PAR4 or fibrinogen knockout [3], or platelet GPIba deletion [4], has been reported. Both GPIb and GPVI are adhesion receptors that support platelet interaction with collagen, indirectly in the former case via VWF, and directly in the latter. GPVI is a rapid and effective signaling receptor, whilst the VWF/GPIb axis is considered unable to elicit full activation of platelets under most conditions. The greater effect of ablation of GPIb, a weak binder of immobilized VWF at low shear but essential for platelet adhesion at high shear, raises the question of whether the shear rate at the site of platelet activation may be an important determinant of themediation ofmetastasis, and suggests that in this system, the capture of platelets rather than their activation is the dominant process. The possibility that platelets or platelet-derived microparticles may adhere to tumor cells and hitch a ride around the circulation until a suitably exposed subendothelial collagen is encountered, in the lung, supports the idea that platelet adhesion receptors play a role in targeting tumor cells to their metastatic niche [3]. Binding to tumor cells might result in platelet activation through GPVI, with consequent enhancement of metastasis through any or all of the mechanisms outlined here. Several direct and indirect ligand–receptor systems might mediate platelet–tumor cell interactions, utilizing, for example, GPIb or GPVI, the two platelet adhesion receptors studied by Jain in this and their previous study. GPIb-bound VWF can interact with RGD-binding integrins, for example aVb3 that might be expressed on the tumor cell surface, whilst GPVI may bind laminin [5] and thence interact with apposed b1 integrins. Other direct interactions could occur, including that of the epithelial tumor-expressed podoplanin with platelet CLEC-2 [6], or tumor PSGL with P-selectin. The interaction of platelets with tumor cells has been found to perturb the ability of NK cells to target tumor cells [7], possibly by minimizing the scope for intercellular contact. The therapeutic potential of GPVI in metastasis is of obvious interest. Because the bleeding defect in mouse associated with GPVI knockout is minor, and that of Correspondence: Richard W. Farndale, Department of Biochemistry, University of Cambridge, Downing Site, Cambridge CB2 1QW, UK. Tel.: +44 1223 766111; fax: +44 1223 333345. E-mail: rwf10@cam.ac.uk Journal of Thrombosis and Haemostasis, 7: 1711–1712 DOI: 10.1111/j.1538-7836.2009.03566.x

The functions of the A1A2A3 domains in von Willebrand factor include multimerin 1 binding

Summary Multimerin 1 (MMRN1) is a massive, homopolymeric protein that is stored in platelets and endothelial cells for activation-induced release. In vitro, MMRN1 binds to the outer surfaces of activated platelets and endothelial cells, the extracellular matrix (including collagen) and von Willebrand factor (VWF) to support platelet adhesive functions. VWF associates with MMRN1 at high shear, not static conditions, suggesting that shear exposes cryptic sites within VWF that support MMRN1 binding. Modified ELISA and surface plasmon resonance were used to study the structural features of VWF that support MMRN1 binding, and determine the affinities for VWF-MMRN1 binding. High shear microfluidic platelet adhesion assays determined the functional consequences for VWF-MMRN1 binding. VWF binding to MMRN1 was enhanced by shear exposure and ristocetin, and required VWF A1A2A3 region, specifically the A1 and A3 domains. VWF A1A2A3 bound to MMRN1 with a physiologically relevant binding affinity (KD: 2.0 ± 0.4 nM), whereas the individual VWF A1 (KD: 39.3 ± 7.7 nM) and A3 domains (KD: 229 ± 114 nM) bound to MMRN1 with lower affinities. VWF A1A2A3 was also sufficient to support the adhesion of resting platelets to MMRN1 at high shear, by a mechanism dependent on VWF-GPIbα binding. Our study provides new information on the molecular basis of MMRN1 binding to VWF, and its role in supporting platelet adhesion at high shear. We propose that at sites of vessel injury, MMRN1 that is released following activation of platelets and endothelial cells, binds to VWF A1A2A3 region to support platelet adhesion at arterial shear rates.