The possibility that plasminogen is released from platelet organelles or that platelets express specific receptors for plasminogen binding outlines a new chapter on the platelet functions in the physiological and pathological hemostasis. The increasingly accepted hypothesis is that platelets provide a mechanism for the localization of plasminogen at a site of fibrin formation, thereby enhancing the fibrinolytic cascade [37].

Firstly, Coppinger et al. showed that platelets are able to release plasminogen upon thrombin stimulation [38]. Proteomic analysis of platelet α-granules using mass spectrometry demonstrated that plasminogen is contained in α-granules, but it has not yet clarified whether the alpha-granular pool contains Lys- or Glu-plasminogen [35]. Notably, plasminogen was undetectable in megakaryocyte cultures unless the cell culture was exposed to the media containing plasma, probably due to an endocytic mechanism by these cells [39]. Secondly, it has been established that plasminogen interacts with human platelets, and the binding is enhanced fivefold by thrombin stimulation [40]. Multiple receptors for plasminogen have been described in various cell types, and a common feature of these receptors is that they interact with plasminogen via its lysine binding sites located in the kringle domains. This lysine-dependent binding mechanism was confirmed by the treatments with lysine analog ε-aminocaproic acid and carboxypeptidase B [40,41,42]. It has been established that plasminogen, associated with the platelet surface, assumes an open conformation that is more readily cleaved to plasmin by activators and is protected from inhibition by α2-antiplasmin [40,43,44,45].

Of the platelet surface proteins, the integrin α_IIbβ₃ seems to be the main plasminogen-binding site. Miles and Plow demonstrated that plasminogen binds to the fibrinogen receptor α_IIbβ₃ on platelets and that thrombin stimulation increases plasminogen binding [40]. Subsequently, Adelman et al. showed that plasminogen binding to platelet α_IIbβ₃ may be mediated via plasminogen associated with fibrinogen through lysine binding domains [42].

Whyte et al. provided new insights on the location and the role of plasminogen in in vitro model for thrombus formation and lysis under flow conditions. The authors found that adherent spread platelets accumulate plasminogen in an α_IIbβ₃-dependent manner, whereas PS-exposing platelets directly bind plasminogen through protruding caps or indirectly via platelet-associated fibrinogen. Notably, PS-exposing platelets did not express α_IIbβ₃, as demonstrated by the lack of PAC-1 (activated GP IIb/IIIa) staining in the platelet caps. Altogether, these data indicated that PS-exposing platelets foster the conversion of plasminogen into plasmin, thus modulating local fibrinolysis within the microenvironment of the thrombus [46].

Plg-R_kt is an integral transmembrane receptor synthesized with a C-terminal lysine, firstly identified on the macrophage surface, which promotes plasminogen activation, cell migration, and polarization. More recently, Whyte et al. demonstrated the presence of Plg-R_kt on the platelet surface. Genetic ablation of Plg-R_kt reduces plasminogen binding by roughly 50%, suggesting the central role of this receptor in the retention of plasminogen on the activated platelets [47].

It is conceivable that platelets are involved in the plasmin positive feedback loop; once released from platelets, plasmin can participate in mechanisms that further enhance platelet activities. However, it is still unclear whether plasmin exerts an activating or inhibiting action on platelets [48]. It is certain that experimental temperature, incubation time, plasmin concentration, and presence of plasma proteins modify plasmin effects on human platelets in vitro [37].

Niewiarowski et al. were the first to introduce the concept of plasmin as a platelet activator. The authors demonstrated that plasmin could induce reversible platelet aggregation and significant granule release. The short treatment of platelets with plasmin enhanced their sensitivity to ADP, a key agonist for platelet aggregation [49]. The study of biochemical mechanisms indicated that plasmin at high doses caused aggregation and secretion of washed human platelets associated with Ca²⁺ mobilization and phospholipase C and protein kinase C activation, suggesting the involvement of G_q protein [43]. In line with these studies, Ishii-Watabe et al. suggested that plasmin directly induced platelet aggregation via ADP secretion from platelets. In addition, the authors showed that platelet shape change occurred just after the stimulation with plasmin, while several minutes were required for granule release and aggregation [50]. Interestingly, platelet incubation with lower plasmin concentrations for 5 min at room temperature enhanced platelet aggregation, integrin α_IIbβ₃ expression, α-, δ-granules, and lysosomes release [51,52]. The data obtained from Quinton et al. indicated that plasmin induced platelet aggregation primarily through cleavage of Protease-Activated Receptors (PAR)-4 [51].

It could be of clinical relevance to investigate the molecular aspects of the regulation of plasmin-mediated platelet activation. More recently, Pielsticker et al. indicated plasmin as a potent platelet activator, inducing P-selectin expression and fibrinogen binding. Moreover, the authors identified matricellular glycoprotein thrombospondin-1 (TSP-1) as a promoter or an inhibitor of plasmin-induced platelet activation, depending on its concentration, when bound to plasmin [53].

An important factor influencing the platelet response to plasmin in vitro is the platelet environment. Blockmans et al. reported the aggregatory effects of plasmin on gel-filtered platelets but no aggregation response in PRP, probably due to the presence of α2-antiplasmin in PRP [54]. In contrast to these data, Pielsticker et al. observed plasmin activating effects on platelets not only in the absence of plasma factors but also in diluted platelet-rich plasma and citrated whole blood, despite the presence of the plasmatic plasmin inhibitors α2-antiplasmin and α2-macroglobulin [53].

Related to the inhibitory effects of plasmin on platelets, Schafer and Adelman found that plasmin inhibited arachidonic acid mobilization from platelet membrane phospholipid pools [55]. Plasmin treatment of washed platelets resulted in progressive loss of GPIb, which contains receptor sites for von Willebrand (vWF), thereby reducing vWF-mediated platelet agglutination [48]. In the presence of plasma, the incubation of human platelets with streptokinase, able to produce plasmin, resulted in impairment of aggregation [56].

In conclusion, plasmin seems to exert contrasting effects on platelets, being able to both activate and inhibit platelet activity in vitro. This bivalent behavior could be caused by several variables, such as temperature and/or incubation time, that can modify and influence the platelet response to plasmin in vitro, but a consistent body of evidence predominantly indicates that plasmin acts as a platelet activator.

The potential role for platelets in the modulation of fibrinolysis became apparent when Vaughan et al. demonstrated that the platelet surface possessed for tPA a large number of specific, low-affinity binding sites that were amplified upon thrombin stimulation [57]. However, specific receptors for tPA on the platelet surface have not yet been identified. In addition, the presence of tPA has been detected by different techniques in isolated megakaryocytes, as well as in platelets [58,59].

tPA released from endothelium protects against the formation of platelet aggregates in vivo by modulation of platelet cyclic nucleotides [60].

The thrombolytic activity of tPA was impaired by platelets in vitro as a consequence of clot retraction and decreased binding of the plasminogen activator [33]. This may explain the resistance of platelet-rich clots to lysis in vivo.

Importantly, the “lisability” of platelet-rich areas depends on the complex interrelationships between the mechanical platelet contractile forces and the process of fibrinolysis. In fact, GP IIb/IIIa inhibitors facilitated the rate and the extent of fibrinolysis by improving recombinant tPA binding velocity and, subsequently, the lysis rate in platelet-rich areas, thus providing new insights into the synergistic potential of GP IIb/IIIa inhibitors and fibrinolytic agents in dissolving thrombotic vaso-occlusions [61].

A model of platelet lysate affinity chromatography showed that a number of platelet proteins that have previously been found to inhibit the fibrinolytic system were captured by tPA. Since there is evidence that platelets undergo lysis after trauma, the authors hypothesized that platelet lysis could shut down tPA-mediated fibrinolysis, with implications for therapeutic intervention, in order to balance fibrinolysis during resuscitation of trauma patients [62].

Plasminogen activator uPA plays a central role outside the vasculature by interacting with its receptor uPAR on the cell surface, but several studies reported an active role also in fibrinolysis.

Platelets from normal subjects contain a small amount of uPA (up to 1.3 ng/10⁹), while insignificant levels were found in megakaryocytes [63]. It was shown that about 20% of pro-uPA intrinsic to blood was closely related to platelets and that platelets were able to rapidly incorporate exogenous pro-uPA [64].

pro-uPA seems to be involved in a mechanism for stimulation of fibrinolysis by platelets. In fact, it was observed that pro-uPA could mediate the clot lysis, and this mechanism was enhanced by platelets. Loza JP et al. reported that platelet-bound prekallikrein was able to enhance pro-uPA-induced clot lysis [65], but most probably, platelet-bound plasminogen is the essential player in this phenomenon. In fact, platelets from plasminogen-deficient mice were not able to activate pro-uPA, and the activation of pro-uPA was significantly enhanced by plasminogen on the platelet surface rather than plasminogen in the solution phase [44]. Interestingly, fragment E-2 from fibrin promoted plasminogen activation by pro-uPA, and fibrin specificity to pro-uPA didn’t require its conversion to uPA [66]. Furthermore, in the presence of plasma, activation of Glu-plasminogen bound to degraded fibrin was found to be more efficient with pro-uPA than with uPA [67]. In plasma, the pro-uPA fibrinolytic property was stable for several days and, unlike uPA, did not form stable inhibitor complexes [68]. Altogether, these findings suggest the use of pro-uPA as a thrombolytic agent against platelet-rich thrombi, which are more resistant to lysis.

Storage of large amounts of uPA in blood platelets was observed in a congenital bleeding disorder known as Quebec Platelet Disorder (QPD). Unlike normal platelets, QPD platelets contain approximately 400–600 ng/10⁹ of uPA without increased uPA in plasma or systemic fibrinolysis [69]. uPA from QPD platelets is predominantly detected as active two-chain uPA and in smaller amounts as single-chain and/or forms complexed to PAI-1. The high levels of active uPA were accompanied by abnormal intraplatelet generation of plasmin. During clot formation, uPA released by QPD platelets led to platelet-dependent increased fibrinolysis, thus provoking QPD bleeding. QPD platelets showed the loss of normal α-granule proteins, and this may possibly contribute to QPD bleeding. The genetic defect that causes QPD is characterized by a duplication mutation that selectively induces increased production of normal PLAU transcripts by megakaryocytes [70].

Another mechanism for uPA-mediated plasmin formation has been reported: glu-plasminogen bound to platelet surface was converted to plasmin with a high efficiency by uPA expressed on monocytes or endothelial cells-derived microparticles [45]. Importantly, uPA associated with platelet surface was able to upregulate uPA mRNA synthesis by endothelial cells [71]. Through this mechanism, platelets may control intravascular fibrin deposition.

The connection between platelet functions and plasminogen activator uPA has also been found in angiogenesis. It is known that platelets are an important circulating store of angiogenesis regulators, such as the angiogenesis inhibitor angiostatin. Jurasz et al. showed that angiostatin generation by platelets was inhibited by a selective inhibitor of uPA. The authors concluded that platelets constitutively generate angiostatin on their membranes, and this mechanism is mediated by uPA [72].

It has been established that uPA receptor (uPAR) is expressed on the platelet surface, but data on uPAR involvement in platelet pathophysiology are not available. However, a critical role of platelets uPAR for kinetics and endothelium adhesion associated with inflammation has been demonstrated [73]. Sloand et al. reported increased suPAR levels in the sera of patients affected by paroxysmal nocturnal hemoglobinuria (PNH). In PNH patients, GPI-negative granulocytes and platelets were the probable source of elevated plasma suPAR levels that were associated with thrombosis and inhibition of plasmin generation [74].

The cleaved form of uPAR (DII-DIII-uPAR) containing the SRSRY sequence corresponding to amino acids 88–92 at its N-terminus interacts with N-formyl peptide receptors (FPRs), thus mediating cell migration [75]. FPRs belong to a family of G protein-coupled receptors and are involved in the regulation of innate immunity and host defense. Three family members are identified: FPR1, FPR2/ALX, and FPR3, mainly expressed in several types of innate immune cells, including neutrophils and monocytes/macrophages; other cell types also express FPR members. Recently, our group assigned an important pathogenetic role to crosstalk between uPAR and FPRs in inflammation and cancer [75,76]. Despite numerous research focused on the role of FPRs in the modulation of immune responses, little data on their ability to regulate hemostasis and thrombosis are available. Czapiga M et al. demonstrated the expression of FPR1 on the platelet surface and its ability to induce chemotactic and migratory functions upon stimulation with E coli-derived fMLF [77]. Salamah et al. showed the expression of FPR2 and its effects in promoting platelet activation and thrombus formation under arterial flow conditions [78].

Further studies are necessary to establish the role of FPRs on platelets and their contribution to hemostasis and thrombosis, but it would be interesting to investigate whether crosstalk between uPAR and FPRs occurs on the platelet surface and whether it is involved in the regulation of FPRs-mediated effects on platelets.

The main endogenous inhibitor of plasminogen activators and the most important modulator of the fibrinolytic system is PAI-1. Several cell types are capable of synthesizing, storing, and releasing PAI-1, but platelets are considered a major reservoir of plasma PAI-1 [79]. However, there is an ongoing scientific debate on the effective contribution of platelet PAI-1 to fibrinolysis.

About 90% of the total PAI-1 in blood derived from platelet α-granules and platelet count were correlated with plasma PAI-1 concentrations [80]. Recent findings showed that more than 50% of platelet PAI-1 is in the active configuration, contrary to preliminary studies where the majority of PAI-1 stored in platelets has been considered to be inactive [81]. Because platelets retain mRNA from megakaryocytes and a capacity for synthesis of some proteins, platelets can de novo synthesize PAI-1, and the amount synthesized in vitro in 24 h is 35-fold higher than required to maintain normal plasma levels. Interestingly, Brogren et al. investigated the hypothesis that platelets might be the source of plasma PAI-1 and that the cellular source of PAI-1 can be determined by its tissue-specific glycosylation pattern [79]. PAI-1 isolated from macrophages, endothelial cells, and adipose tissue expressed heterogeneous glycosylation patterns, but no glycans were detected on PAI-1 isolated from plasma or platelets from healthy individuals, thus suggesting that platelets may be the main source of plasma PAI-1 [82].

The study of platelet PAI-1 provided multiple evidence that it plays a central role in fibrinolytic resistance to clots and thrombi. In vitro studies showed that platelet-mediated fibrinolytic resistance is reduced by neutralizing antibodies direct against PAI-1 [83]. In platelet-rich plasma clot lysis mediated by tPA, the release of PAI-1 from activated platelets resulted in a prolongation of the clot lysis time [84].

In order to elucidate the mechanisms responsible for the stability of platelet-rich thrombi, in vitro model thrombus (composed of thrombin-activated platelets, fibrin, plasminogen, and tPA) provided novel insights into the role of platelet PAI-1. In particular, PAI-1 released from activated platelets stabilizes fibrin deposited on the platelet surface and thus increases the lysis resistance of platelet-rich thrombi [83].

PAI-1 circulates in complex with vitronectin (VN), an interaction that stabilizes PAI-1 in the active form [85]. VN can be internalized into platelet α-granules from plasma, but its effects on the platelet pool of PAI-1 are not clear. Confocal microscopic analysis of platelet-rich plasma clots confirmed the co-localization of PAI-1 with VN and vimentin on the activated platelet surface, suggesting that platelet vimentin may regulate fibrinolysis by binding platelet-derived VN/PAI-1 complexes [86].

Animal model studies established that platelet PAI-1 was primarily involved in arterial thrombi, which contain high platelet content [37]. In particular, Zhu et al. noted that, in a murine model of carotid artery injury, the reperfusion after tPA treatment occurred in PAI-deficient mice but not in wild-type animals, thus demonstrating the main role of PAI-1 in the thrombolytic resistance of platelet-rich thrombi [87].

In line with in vitro studies and data from animal models, the immunohistochemical analysis of human thrombi suggested that active recruitment of platelets contributes to the high PAI-1 concentration in thrombi [88].

Diverse pathological conditions are associated with unregulated PAI-1 levels, but only plasma PAI-1 levels in most cases were examined.

Platelets were capable of immediate degranulation when exposed to thrombin leading to the release of active PAI-1, which is complexed with tPA and inhibits tPA-induced fibrinolysis [89]. Following degranulation, platelets synthesize more PAI-1, but such synthesis occurs over a long period of time, rendering these platelets dysfunctional. Since thrombin levels are elevated in trauma and intensive care unit (ICU) patients, these observations deserve more attention.

Platelet PAI-1 could play a central role in diverse pathological conditions associated with hypoxia, such as chronic obstructive pulmonary disease (COPD). In fact, exposure of human platelets to hypoxia enhanced PAI-1 expression. In addition, circulating platelets isolated from COPD patients had higher PAI-1 levels compared to controls, and probably, the prothrombotic phenotype associated with hypoxia may be attributable to the synthesis of PAI-1 by platelets [90].

It has been proposed that miRNA plays a central role in the regulation of fibrinolysis [91]. Platelets of patients with type 2 mellitus diabetes (DM2) release significantly more PAI-1 at the same level of platelet aggregation. This may contribute to enhanced thrombosis in diabetes [92]. Moreover, recent studies showed that mir-30c negatively regulated PAI-1 mRNA and protein expression in megakaryocytes; hyperglicemia-induced repression of mir-30c enhanced PAI-1 expression and thrombus formation in DM2 [93].

In congenital bleeding disorders, low levels of intraplatelet PAI-1 were found, but it is difficult to establish the relationship between deficiency of platelet PAI-1 and bleeding [37].

A significant increase in platelet PAI-1 levels was observed in essential thrombocythemia (ET) patients with thrombotic complications compared to ET patients without thrombotic complications and control group [94].

Taken together, the evidence strongly suggests that platelet PAI-1 plays a central role in arterial thrombosis and in hemostasis, as it is a determinant factor for fibrinolytic resistance.

PAI-2 is classically considered as an inhibitor of fibrinolysis via inhibition of plasmin generation by uPA and, to a lesser extent, tPA. Novel functions were attributed to PAI-2, which seems to be involved in hemostasis and to be associated with platelets. In particular, PAI-2 deficiency in mice reduced bleeding times via dysregulated platelet activation. The immunohistochemistry analysis of blood clots demonstrated that PAI-2 was associated with platelet aggregates; however, PAI-2 was not detected in human platelets [95].

α2-antiplasmin is the principal inhibitor of plasmin and can be detected in platelet α-granules. α2-antiplasmin antigen contained in platelets corresponded to 0.05% of the total α2-antiplasmin present in the blood on a volume basis, and 87.5% of this material can be released when platelets are stimulated by thrombin. Immunohistochemical analysis of human thrombi showed that α2-antiplasmin was present at high concentrations, but platelets released small amounts of α2-antiplasmin [37,88,96]. However, thrombus stabilization against fibrinolysis involves this mechanism: platelet FXIII-A through exposure to the activated platelet membrane exerts antifibrinolytic function by cross-linking α2-antiplasmin to fibrin [97].

Takei M et al. examined the effects of α2-antiplasmin on platelet aggregation using animal models and found that lack of α2-antiplasmin increased platelet micro-aggregation when platelets were stimulated with ADP. Microaggregates produced in the early phase of platelet activation have the potential to aggravate thrombus formation, leading to vascular occlusions by the formation of a platelet-rich thrombus [98].