1 Introduction

Platelets are small (~2–4 μm in diameter) anucleate cellular fragments derived from bone marrow-resident megakaryocytes (~50–100 μm in diameter) and are classically known to be indispensable for hemostasis [1]. Platelet generation is a complex and highly regulated process [2], especially considering the multifunctional aspects of platelets. The importance of this tight regulation is demonstrated in vivo, where it is essential to prevent bleedings under conditions when platelet counts are low, but it is also imperative to prevent serious organ damage and vascular occlusion due to elevated platelet counts. It is becoming increasingly clear that platelets are involved in several other functional processes, apart from their well-known role in hemostasis, in both health and disease [1, 3–9]. These nonhemostatic functions, generally outlined as “platelets versus pathogens” and “platelet-target cell communication,” will be highlighted in this chapter and will shed light on the versatile role of platelets in an inflammatory setting.

2 Platelets Versus Pathogens

2.1 Pathogen Recognition by Platelets

Platelets have the ability to “sense danger,” such as pathogens or damage in case of sterile inflammation, as they express functional immune receptors called pattern recognition receptors. These include Ig- or complement receptors and Toll-like receptors (TLRs) [1]. Via these receptors, platelets are able to bind invading pathogens and microbes, including their derived materials. Pathogens first encounter TLRs on professional phagocytes, such as neutrophils, dendritic cells (DCs), or macrophages [1, 10, 11]. TLRs are germ line-encoded proteins, capable of binding various infectious molecular structures and potently stimulate innate immune mechanisms [1, 10, 11]. Many groups have reported the expression of TLRs 1–9 on human as well as on murine platelets, and some of these TLRs have been shown to be functional [1]. For instance, TLR4, which has been shown to mediate lipopolysaccharide (LPS, a gram-negative endotoxin)-induced thrombocytopenia and TNF-α production in vivo [12–18]. Engagement of TLR2 in human platelets by Pam3CSK4, a synthetic ligand mimicking bacterial lipopeptide, was shown to induce a thromboinflammatory response through activation of phosphoinositide 3-kinase [19]. Platelet TLR2 and TLR4 were also shown to be of importance in a study regarding periodontitis, which is associated with an increased risk for cardiovascular diseases [20]. The periodontopathogens involved (A. actinomycetemcomitans and P. gingivalis) were shown to induce expression of CD40L, known to mediate thrombotic and inflammatory processes, on human platelets via TLR2 and TLR4 [20]. Recently, new insights have emerged regarding platelet TLR3 and TRLR7, as human platelets expressing TLR3 were shown to respond to poly I:C, indicating an effect on innate immune responses when detecting viral dsRNA [21]. Platelet TLR7 was shown to mediate host survival and platelet counts during infection with encephalomyocarditis virus (EMCV) in mice, independently of thrombosis [22]. In contrast, platelet TLR9 seems to be important as sensor of internal danger signals, rather than external signals. It was shown that platelet TLR9 was functionally associated with oxidative stress, innate immunity, and thrombosis, as carboxy(alkylpyrrole) protein adducts, altered-self ligands produced during oxidative stress, through engagement of platelet TLR9, could enhance platelet activation, granule secretion, aggregation in vitro, and thrombosis in vivo [23]. Not many studies have been conducted investigating the role of TLRs on megakaryocytes. It was shown that endotoxemia could increase thrombopoietin (TPO) levels in vivo, accompanied by an increase in circulating young reticulated platelets with enhanced platelet-neutrophil aggregates [24]. In addition, the bone marrow treated with LPS demonstrated increased TPO levels, implicating a key role of infection and inflammation in thrombopoiesis [25]. Also, TLR4-knockout mice displayed decreased circulating platelet counts and reticulated platelets, suggesting TLR4 to be of importance in platelet production [12, 26]. A lot of evidence has accumulated over the years suggesting an important role for platelets as pathogen sensors within the blood, due to their expression of several receptors without any clear relation to hemostatic functions.

2.2 Pathogen Retainment by Platelets

Platelets are capable of harboring pathogens on their plasma membrane as well as internally [4, 27], as has been demonstrated for viruses [27, 28], bacteria [29–31], and parasites [5]. Platelets were also shown to be involved in acute and chronic hepatic disease due to hepatitis B virus, via upregulation of virus-specific CD8+ T cells and nonspecific inflammatory cells into the liver [32]. Interestingly, activated platelets were shown to surround or encapsulate Staphylococcus aureus, driving the pathogens into clusters resulting in reduced bacterial growth [33]. This occurred through secretion of the antimicrobial peptide β-defensin and signaling of neutrophil extracellular trap (NET) formation [33], which has now shown to be involved in many pathologies including thrombosis, transfusion-related acute lung injury, sickle cell disease, storage of red blood cells, and very recently diabetes [34–40]. Bacteria (methicillin-resistant Staphylococcus aureus and Bacillus cereus) were also trapped on the hepatic Kupffer cells, which were dependent upon interactions with platelet-adhesion receptor GP1b [6]. In that study, infected GP1bα-deficient mice suffered more endothelial cell and Kupffer cell damage and displayed more vascular leakage and rapid mortality [6]. Activation of platelets during sepsis can contribute to disseminated intravascular coagulation, which can lead to blood vessel occlusion, increased ischemia, and multiple organ failure, and it can also contribute to stimulation of pro- and anti-inflammatory cytokine production [41]. This platelet activation is evident from increased surface P-selectin expression [42, 43] and increased levels of triggering receptor expressed on myeloid cells-like transcript-1 [44] or PF-4 in mice [45]. During sepsis, neutrophils were also shown to be activated by platelet TLR4, causing the release of NETs, which subsequently trapped bacteria in blood vessels primarily of liver sinusoids and lung capillaries [7]. It was proposed that platelets act as circulating sentinels, sensing infectious agents and presenting them to neutrophils and/or the reticuloendothelial system [14–18]. Although platelet-dependent NET formation is an effective strategy in trapping bacteria, it may be detrimental as it may occur at the expense of injury to the host. When LPS-activated neutrophils come into contact with endothelium, there is little damage; however, if bound neutrophils encounter LPS-bearing platelets, neutrophil activation is enhanced resulting in NET formation together with reactive oxygen species release, which damages the underlying endothelium [7]. Furthermore, it was demonstrated that neutrophils are able to scan platelets for activation in the bloodstream via P-selectin ligand signaling, resulting in inflammation [8]. Also, platelet P-selectin, soluble or cellular, was found to trigger NET formation in mice through binding to neutrophil P-selectin glycoprotein ligand-1 (PSGL-1) [46]. Recently, in diabetes, in which neutrophils are more susceptible to NET formation, NETs were found to impair wound healing. Therefore, it was suggested that cleaving NETs or inhibiting NET formation may improve wound healing and reduce inflammation in diabetes [40].

2.3 Pathogen Elimination by Platelets

Platelets have been implicated to be involved in the clearance of bacterial infections. In infective endocarditis, for instance, thrombin-stimulated platelets were shown to facilitate clearance of streptococci [47]. Additionally, in mice infected with P. gingivalis, platelet TLR2 was implicated in the formation of platelet-neutrophil aggregates [19], and later it was demonstrated that phagocytosis of periodontopathogens mediated by neutrophils was dependent upon platelets, plasma factors, and TLR2 [48]. It was elegantly shown by McMorran and colleagues that activated platelets can kill the malarial parasite Plasmodium falciparum inside the red blood cell [5]. In a follow-up study, it was further elucidated that the mechanism of this platelet-mediated parasite killing was dependent upon platelet factor 4 (PF4 or CXCL4) and the erythrocyte Duffy-antigen receptor (Fy) [49]. This implies that in Duffy-negative individuals, thus lacking Fy, platelets would be incapable of eliminating this intraerythrocytic malarial parasite.

2.4 Pathogen Escape from Platelets

On the other hand, viruses and bacteria appear to have developed countermeasures to evade these immune responses elicited by platelets. This can be supported by the fact that acute viral or bacterial infections often lead to low platelet counts or thrombocytopenia. This has frequently been observed in immune thrombocytopenia (ITP), an autoimmune bleeding disease in which platelets are destroyed [50]. The pathogenesis of infection-related platelet targeting is incompletely understood, but several mechanisms have been described. These include molecular mimicry between viral/bacterial antigens and platelet antigens, resulting in cross-reactive autoantibody generation [51–55]. In addition, ITP patients infected with the gram-negative bacteria Helicobacter pylori demonstrated increased platelet counts following Helicobacter pylori-eradication therapy [56]. Similarly, the gram-negative bacterial endotoxin LPS also enhanced antiplatelet antibody-mediated platelet phagocytosis in vitro [15], as well as an increased platelet clearance in vivo, when antiplatelet antibodies and LPS were coinjected in mice [57]. Furthermore, C-reactive protein (CRP) was found to be a novel serum factor which potentiated antibody-mediated platelet destruction both in vitro as well as in vivo in mice [58]. CRP is an acute-phase protein, also present in healthy individuals, but known to vastly increase during acute bacterial or viral infections. CRP was found to be increased in children suffering from ITP and treatment with IVIg was correlated with increased platelet counts, with decreased levels of CRP, and with reduced clinical bleeding severity [58]. Interestingly, an elevated CRP value at diagnosis appeared to be predictive for slower platelet count recovery after 3 months [58]. From a functional perspective, the mechanism appeared to be independent of the platelet FcγRIIA but occurred via platelet oxidation triggered by antiplatelet antibodies and the phagocyte NADPH oxidase system, resulting in platelet-membrane phosphorylcholine exposure, to which CRP could bind and then subsequently enhance antibody-mediated platelet phagocytosis through interaction with FcγR [58].

3 Platelet-Target Cell Communication

3.1 Platelet Release of Mediators

CD40L (CD40L/CD154) and CD40, besides their role in costimulation and perhaps thrombotic diseases [59], are also relevant in platelet immune reactions. When platelets are activated, most of their expressed CD40L is released, generating the soluble form (sCD40L), which is in fact the vast majority of all sCD40L in circulation [60]. Platelet CD40L can engage with endothelial cell-CD40 (in the membrane), which results in a cascade of inflammatory reactions leading to the release of several adhesion molecules including VCAM1, ICAM1, and CCL2 [61]. Platelet-secreted sCD40L, when interacting with CD40 positive vascular cells (including endothelial cells), can enhance the expression of adhesion molecules like P-selectin and E-selectin and stimulate the release of tissue factor and IL-6 [62, 63]. Therefore, a central role for platelet CD40L-CD40 interactions between endothelium/coagulation and inflammation is becoming apparent. Additionally, platelet-derived CD40L was shown to enhance CD8+ T cell responses and to stimulate T cell responses following infection with Listeria monocytogenes [64, 65], demonstrating a clear link between innate and adaptive immunity. Platelet CD40L was also shown to bind to dendritic cells (DCs) and thereby impair their differentiation, suppress the pro-inflammatory cytokines IL-12p70 and TNF by DCs, and increase IL-10 production by DCs [66]. Furthermore, activated platelets were shown to enhance lymphocyte adhesion to endothelial cells [67] and facilitate homing of lymphocytes in high endothelial venules [68] and migration toward inflammatory environments. Also, platelets can enable B cell differentiation and Ab class switching via their CD40L [69, 70]. Several other signaling pathways have been linked to platelet activation via the CD40L-CD40 axis, including NF-kB [71–76], illustrating that platelets have in fact several strategies for modulating adaptive immune responses through their CD40L and/or their derived sCD40L.

Platelets have many different cytokines and chemokines in their system, all differently impacting hemostasis and wound repair [77], and also pro-inflammatory and anti-inflammatory reactions, for instance, the immunosuppressant TGF-β [78]. Platelets appear to control the levels of TGF-β as is evident from patients suffering from immune thrombocytopenia (ITP), where low levels of TGF-β were observed during active disease, on one hand, but those levels normalized again upon treatment which increased platelet counts, on the other hand [79, 80]. Most of the platelet chemokines and cytokines are located within the different platelet granules. The α granules contain several immunomodulatory soluble factors, like chemokines, which included PF (CXCL4), RANTES (CCL5), β-thromboglobulin (β-TG, an isoform of CXCL7), and MIP-1α (CCL3) [81]. Platelet activation triggers release of these chemokines causing a diverse response of cellular interactions and responses. PF-4, for instance, renders monocytes resistant to apoptosis and stimulates their differentiation into macrophages [82]. Besides that, PF-4 is capable of enhancing neutrophil adhesion to unstimulated endothelial cells and of release of granule content [83]. In contrast, platelet-derived β-TGs, which are proteolytic products of inactive precursors, can either stimulate or inhibit neutrophil activity [84]. Also, platelet-derived MIP-1α can enhance histamine release from basophils [85] and is chemotactic for T cells [86].

3.2 Platelet Microparticle Shedding

Platelet microparticles (also referred to as microvesicles) are small extracellular vesicles produced by cell cytoplasmic blebbing and fission. Originally, they were described as “dust” released from activated platelets which supported thrombin generation, even without the presence of intact platelets [87]. Generally, the size of microparticles ranges from ~100 to 1,000 nm in diameter, although the majority are ~200 nm, and they are distinct from exosomes, which are ~50–100 nm in diameter and thus smaller in size and originating from multivesicular bodies via exocytosis [88]. The minimal experimental requirements for the definitions of extracellular vesicles and their functions are described by the International Society for Extracellular Vesicles [89]. Platelets appear to be particularly effective in the formation of microparticles, as compared to other cell types, as was demonstrated by the high abundance of platelet microparticles in circulation using cryotransmission electron microscopy and gold nanospheres conjugated to antibodies against the platelet CD41 [90]. Microparticle formation is associated with elevated intracellular calcium levels, cytoskeletal rearrangement, and membrane phosphatidylserine (PS) exposure [91], which supports coagulation considering its anionic properties, but platelet microparticles express modest levels of tissue factor (TF) and seem to be less procoagulant than monocyte-derived microparticles, which express PS as well as TF [92]. Examination of platelet activation under physiological flow conditions revealed elongated membrane strands (up to 250 μM) emerging from platelets, so-called flow-induced protrusions (FLIPRs) [93]. FLIPRs also expose PS, recruit monocytes and neutrophils, and appear to shed off PS+ microparticles [93]. Interestingly, PS-microparticles have also been described in body fluids [90, 94–96], demonstrating the complexity of platelet microparticle production. It has been challenging to properly assess microparticles in biological fluids, due to their small dimensions. Platelet microparticles have been described in various inflammatory conditions, in which platelets become activated [97, 98], and their clinically their levels were often associated with disease progression. For example, in blood and synovial fluid of patients suffering from rheumatoid arthritis (RA), platelet microparticles were found to be elevated [9, 99–102]. Using a murine model of RA, it was demonstrated that depletion of platelets attenuates inflammation [9, 103]. However, microparticles are observed in sterile, as well as in inflammatory diseases, making it unclear what triggers the platelets to produce microparticles. Several activation pathways may be driving the production of microparticles during inflammation, such as apoptosis, high shear forces, or platelet receptor signaling. The disease setting, at least partly, determines the route leading to microparticle formation, as in RA activation of the collagen receptor glycoprotein VI (GPVI) is a trigger, while in sepsis microparticles are produced via TLR-4 signaling through LPS [9, 104]. Both these signals, however, are accompanied by an increase in IL-1, indicating their role in enhancing inflammation. Additionally, signaling of immune complexes, consisting of bacterial components and well-conserved epitopes expressed by influenza viruses, through the platelet FcγRIIA [105, 106], was shown to lead to the production of microparticles.

From a functional perspective, platelet microparticles are thought to facilitate cell-cell communication. The platelet microparticle cargo is substantial and can consist of various cytokines and chemokines (e.g., IL-1, RANTES), potent lipid mediators (e.g., thromboxane A2), enzymes (e.g., inducible NO synthase), surface receptors (e.g., CD40L), autoantigens (e.g., citrullinated fibrinogen), nucleic acids (e.g., microRNA), transcription factors (e.g., PPARγ, RuvB-like2, STAT3, STAT5a), and interestingly even respiratory competent mitochondria, all of them potentially targeting and impacting a cell [96–98, 107–110]. As the microparticles can express PS and surface receptors, they interact with other cells through integrin and via the PS-binding proteins lactadherin [111] and developmental endothelial locus-1 (Del-1) [112]. These proteins appear to be involved in microparticle clearance and microparticle interaction with other cells, as Del-1 -/- and lactadherin -/- mice express elevated levels of plasma microparticles [111, 112]. Transcription factors transported within platelet microparticles can enable transcellular effects, such as PPARγ, which was demonstrated to be transported inside platelet microparticles and transferred to monocytes where it elicited transcellular effects [109]. Currently, however, more research is required to establish if specific internalization signals are required beyond the initial contact between microparticles and the cellular recipient. Microparticles appear to be important biomarkers in inflammatory disorders, but, further, delineation of their function, mechanisms of their generation, and technical improvements in their assessment are warranted, in order to better understand their role in health and disease.

3.3 Platelet RNA Transfer

Platelets are known to express and secrete many different molecules during platelet activation, and they do so via different signaling mechanisms [81, 113–115]. These molecules have different origins, such as those inherited from megakaryocytes, adsorbed from plasma or synthesized de novo. Despite being anucleate, platelets have been shown to express significant amounts of RNA, including mRNAs (e.g., (pre)mature RNA), structural and catalytic RNAs (e.g., ribosomal and tRNA), regulatory RNAs (e.g., microRNA), and noncoding RNA (e.g., antisense RNA) [107, 116–131]. Moreover, it was described that platelets also possess the molecular machinery for mRNA translation into proteins and that they are able to transfer RNA to recipient cells, in order to regulate cellular functions, such as platelet microRNA-223 transfer to human umbilical vein endothelial cells [107, 127–129, 131]. As mentioned before, intercellular transfer of platelet RNA to target cells can occur via platelet microparticles. However, the content of platelet RNA transcript does not fully match to the platelet proteome content [132]. These molecular tools have opened up a new area of investigation of platelet mRNA and its impact on platelet function in both health and disease [133].

3.4 Platelet MHC Class I Signaling

Major histocompatibility complex (MHC) class I molecules are present on both the platelet plasma membrane and intracellularly [134]. The MHC class I molecules on the platelet plasma membrane are mainly adsorbed from plasma and basically consist of denatured H chains. The platelet-membrane MHC class I molecules appear to be somewhat instable, as they can passively dissociate from the platelet during storage or can be eluted from the membrane due to chloroquine diphosphate or acid treatment, without affecting the platelet membrane integrity [135–141]. Interestingly, denatured MHC class I can elicit faulty interactions with CD8+ T cells, anergizing Cytotoxic T-lymphocytes (CTLs), following transfusions. For instance, allogeneic platelet MHC class I molecules are incapable of stimulating CTL-mediated cytotoxicity on their own [141] but can facilitate the so-called transfusion effect, an immunosuppressive-like reaction to transfused blood cells. CBA mice transfused with allogeneic BALB/c platelets accepted donor-specific skin grafts, in contrast to nontransfused recipients [142]. This implies that allogeneic platelets may inhibit T cell-mediated cytotoxicity reactions, like skin graft rejections. On the other hand, intracellular platelet MHC class I molecules are associated with α granules and generally consist of intact integral membrane proteins associated with β2-microglobulin [143]. Furthermore, it was also demonstrated that platelets contain the entire proteasome system, including TAP molecules, but the endoplasmic reticulum is absent. In syngeneic settings, platelet activation can lead to expression of nascent MHC class I molecules, which are capable of presenting antigens to CD8+ T cells. Activated platelets were shown to present malarial peptides to malaria-specific T cells, resulting in enhanced immunity against the parasite [144]. Therefore, the type of platelet MHC class I (platelet plasma membrane-bound or intracellularly) will determine the effect on T cells (suppression or activation).

A summary of the key nonhemostatic functions of platelets is summarized in Fig. 10.1.

Fig. 10.1
figure 1

The key roles of platelets in modulating inflammatory processes. 1 Platelets can uptake infectious agents, and, via the expression of TLR, they can activate neutrophils to, for example, secrete NETs. 2 Platelet CD40L expression allows them to interact with different cells of the immune system and either activate (arrows) and/or suppress (T bar) them. 3 Intact platelet MHC class I molecules are located intracellularly but upon activation are expressed and can activate antigen (e.g., malaria)-specific CD8+ T cells. On the other hand, the MHC class I molecules on the surface of resting platelet are denatured and lead to CD8+ T cell inhibition. 4 Platelets release Platelet Microparticles (PMPs) under a variety of stress conditions and these PMPs can carry multiple cargos to other cells and sites of inflammation. 5 Platelets contain many different pro-inflammatory and anti-inflammatory cytokines and chemokines and upon activation can release them to the extracellular space. 6 Immune interactions with platelets can lead to severe thrombocytopenic states such as in the case of sepsis, where infections can bind to platelets and cause their sequestration and/or destruction or immune thrombocytopenia (ITP) where the combination of antibody and infectious particles or CRP leads to increased platelet destruction. 7 Platelets contain several different species of RNA and these can be exported via PMPs of mRNAs can be translated into nascent protein synthesis. The culmination of these events makes platelets a formidable immunomodulatory host (Reproduced from reference [3])

Full size image

Conclusions

Platelets have traditionally been viewed as primary regulators of hemostasis and thrombin generation. However, it has become increasingly clear that platelets have multiple functions in inflammation and immunity. Platelets can not only enforce sophisticated protection mechanisms against invading pathogens but are also capable of impacting immune functions in a large variety of recipient cells. They do so by utilizing numerous mechanisms, via diverse surface molecules, through secretion of several pro- and inflammatory mediators and shedding of platelet microparticles carrying a heterogeneous cargo. These relatively novel aspects have shed new light on platelet functions beyond hemostasis.