The Role of Inflammation in Regulating Platelet Production and Function: Toll-like Receptors in Platelets and Megakaryocytes

Abstract

Platelets have been extensively studied as hemostatic regulators, stopping uncontrolled flow of blood from an injured vessel and allowing for repair. However, multiple studies have shown that platelets can interact with bacterial proteins, particularly seen during sepsis and inflammation. Immune cells recognize pathogens through Toll-like Receptors (TLRs). These same receptors allow platelets to recognize bacterial proteins and regulate platelet immunity and function. This review examines the TLRs expressed on platelets and megakaryocytes and how these receptors affect the function of these cells. Through TLRs, platelets go beyond hemostatic regulation and play a pivotal role in inflammation and infection.

Introduction

The immune system consists of two fundamental approaches to recognize and respond to harmful stimuli due to infection – innate and adaptive immunity. Innate immunity is the first line of defense against invading microorganisms and involves a series of reactions that prevent ongoing damage, isolate infective agents, and activate the repair process. Adaptive immunity involves the dynamic adaptation to new and unique epitopes on pathogens in the environment. [1] Innate immunity is primarily mediated by macrophages and neutrophils. These immune cells distinguish between pathogen and self by utilizing signals from Toll-like receptors (TLRs). Stimulation of TLRs results in NFκB and MAPK pathway activation, leading to the production of proinflammatory cytokines.

The Toll gene was first discovered as encoding for a receptor in Drosophila embryos. [2-3] Expression of TLRs is ubiquitous throughout species and has evolutionary conservation. TLR2 (with TLR1 or 6) and TLR5 primarily function at the plasma membrane, while TLRs 3, 7, 8, and 9 are reported to function intracellularly. TLR4 is found both in the plasma and intracellular spaces. Each TLR responds to a different set of ligands or pathogen associated molecular patterns (PAMPs). The most studied of the TLRs, TLR4, recognizes lipopolysaccharides (LPS) of gram-negative bacteria. The variety of TLR2 ligands is the greatest among all the TLRs and this is due to its heterodimerization with TLR1 and TLR6. TLR2 recognizes lipoproteins and lipopeptides, peptidoglycans and lipoteichoic acid of gram-positive bacteria, LPS from non-enterobacteria, lipoarabinomannan from mycobacteria, and zymosan from fungi, to name a few. The intracellular portion of these receptors consists of a Toll/Interleukin-1 Receptor (TIR) domain that binds to an adaptor molecule, MyD88. IRAK1/4 are recruited to the cell membrane, bind to MyD88, and are phosphorylated. TRAF6 then binds to and is activated by IRAK1. TRAF6 will go on to activate the NFκB and MAPK pathways. TLR signaling induces antigen presenting cell activation, pro-inflammatory cytokine production, and increased expression of co-stimulatory ligands. These events are important for induction of innate immune responses and improved acquired immunity. [4]

The interaction between infection and thrombosis has been largely studied in sepsis models, in which a systemic infection leads to the activation of the coagulation pathway, creating thrombi in the microcirculation of organs. These events lead to the consumption of platelets and coagulation proteins and multiple organ failure. Limited work has focused on the direct interaction of platelets with bacteria, leading not only to the formation of a platelet-rich clot, but also the activation of the innate immune system. In this setting, TLRs may be the link. Work done in 1977 showed that streptococcal derived lipoteichoic acid (LTA) could stimulate platelet granule release, independent of other well-characterized platelet receptors. [5] In addition, the synthetic lipopeptide, Pam₃CSK4, has been shown to induce physiological platelet activation. [6] This review will focus on work that has shown the role of TLRs in both platelet production and platelet activation.

TLRs and Megakaryocytes

Megakaryocytes primarily are found in the bone marrow where they constitute 1% of the total cell population [7]. Megakaryocytes are formed through the maturation process known as megakaryopoiesis, in which the cells express thrombotic markers, increase significantly in size and DNA content (16+ N). Upon stimulation by multiple factors, including thrombopoietin, megakaryocytes adhere to the endothelium of the bone marrow sinuses, increasing mRNA and protein levels, which are shuttled to the ever elongating pseudopodia that are forming. From these structures pro-platelets bud off and flow into circulation, where they undergo further maturation into platelets, a process called thrombopoiesis. Few studies have looked into the expression and function of TLRs in megakaryocytes and whether or not these receptors could have a role in platelet production. Initial work had shown that low grade endotoxemia increases thrombopoietin levels, which resulted in an increase in reticulated platelets and increased platelet-neutrophil aggregates. [8] Murine bone marrow treated with LPS also showed an increase in the levels of thrombopoietin and cytokines, factors important for thrombopoiesis. [9] Therefore, it is possible that inflammation and infection can modulate platelet production through TLRs. Work using the Meg-01 cell line, a human megakaryoblast cell line, has shown through RT-PCR and flow cytometry that megakaryocytes only express TLR1 and TLR6. [10] Both mRNA levels are increased in a dose dependent manner over time in the presence of interferon-γ. [10] Based on work done with macrophages derived from THP-1 monocytes, it is hypothesized that the levels of TLRs increase with cell differentiation since Meg-01 cells had a lower level of TLR1 and 6 expression compared to isolated platelets. [10] TLR4 surface expression was also verified by flow cytometry and immunohistochemistry on adherent Meg-01 cells; however, in this study, it was hypothesized that any TLR4 found on platelets was a remnant from megakaryocytes. [11] Further confirmation of TLR4 on megakaryocyte cell surface was shown through flow cytometry of murine megkaryocytes isolated from fetal livers. [12] As the cells matured, as indicated by the increase in CD41 levels, the TLR4 levels also increased. [12] Additionally, TLR2 and TLR4 were both detected by flow cytometry in human megakaryocytes isolated from patients, and both receptors were increased in megakaryocytes from myelodysplatic syndrome, a disorder of the hematopoietic stem cells causing multiple lineage cytopenias. [13]

As for the role of TLRs in megakaryocytes, in 2 different studies, TLR4-/- mice were shown to have a decreased circulating [12, 14] and reticulated [14] platelet count compared to WT, suggesting TLR4 may have a role in thrombopoiesis. After a single sublethal dosage of LPS, circulating platelets levels significantly increase compared to untreated mice; however, circulating platelets in TLR4-/- mice were still lower than WT. [14]

TLRS and Platelets

There is growing research examining the non-hemostatic role of platelets, specifically related to inflammation and infection. Platelets can have an early role in immune surveillance and the transfer of pathogen information to other innate immune cells. Platelets can also modify adaptive immune responses by localizing at sites of bacterial invasion, aggregating around bacteria, and promoting clearance. Coagulation abnormalities are common in severe sepsis and it is possible that the presence of TLRs on platelets could be the link between disseminated intravascular coagulation and sepsis. [15-16] While growing evidence has clearly established the relevance of platelets in inflammation, there is much circumstantial evidence that they are also relevant in the setting of infection. An early study showed that LPS, at high doses, actually inhibited platelet activation. [17] However, a much earlier study showed that E. coli endotoxin in dogs caused a sharp fall in circulating platelet numbers and changes in platelet morphology, suggesting platelet activation. [18] More recently, platelets were shown to bind and internalize pathogens and release microcidal proteins.[19-20] In anaphylaxis-like shock (ALS) induced by mannan, a component of fungal cell surfaces, severity increases when platelets have been depleted, suggestive that platelets play a role in the recognition and clearance of this pathogen. [16] These results are contrary to LPS- induced ALS, in which platelet depletion attenuated the disease state. [16]

Other bacteria have been shown to influence platelet function in a variety of settings. Interaction of Staphyloccus aureaus with circulating platelets has been implicated as a virulent mechanism in the induction of endocarditis. Previous investigations related to the activation of platelets in the setting of infective endocarditis have demonstrated both fibrinogen-dependent [21-22] and -independent pathways may trigger platelet aggregation, [23] as may engagement of glycoprotein Ib on the platelet surface. [24-25] Serotype polysaccharides from Streptococcus mutans have also been shown to induce platelet aggregation. [26] Salmonella typhimurium porin was shown to not induce platelet aggregation but enhance ADP- and thrombin-induced activation. [27] In the setting of infection, strokes are associated with increased platelet-leukocyte interactions. [28] Related to cardiovascular disease, Chlamydia pneumoniae has been shown to adhere to platelets, stimulate P-selectin expression, and trigger aggregation. [29-30] The pro-aggregatory effects of Porphyromonas gingivalis have also been established in mouse [31] and human platelets in vitro. [32-35] Thus, in the same way that TLRs are the sentinel receptors of the immune system, platelets may be a sentinel cell in the blood by modifying the acute response to infection and injury. As seen by these studies, not only are platelet thrombotic pathways activated, but platelet-white cell interactions are stimulated in the setting of infection.

Many studies have identified TLRs on platelets, but unlike megakaryocytes, there is more work being done to understand the physiological relevance of these immunological receptors. In addition, there is still some controversy. In one study, platelets were shown to not bind to LPS, nor express TLR4 or CD14. [36] It was determined that LPS primed platelets by binding to monocyte TLR4 and CD14 and triggered the release of platelet activating factor (PAF) and oxygen radicals, which increases agonist-induced aggregation and heterotypic aggregate formation with monocytes. [36] Other studies, however, have shown that TLRs are present on platelets, summarized in Figure 1. Through RT-PCR and immunoblotting, platelets were shown to express TLR1 and 6. [10] Further, tissue sections of coronary thrombi from patients with acute coronary syndrome show a colocalization of staining for CD41 (platelet marker) and TLR1 and TLR6. [10] Through flow cytometry, platelets were shown to express TLR2 and 4, but not TLR1. [11] Treatment with LPS (TLR4 ligand) and Pam₃CSK4 (TLR2 ligand) did not activate platelets nor augment agonist-induced platelet activation in this study. [11] Another study found that platelets do express TLR1, 2, 4, 6, 8, and 9 on the platelet surface. Intracellularly, there were higher levels of TLR2, 4, and 9. Upon activation, platelet TLR2 [37] and 9 [37-38] increased on the cell surface, while intracellularly, TLR2, 4, and 9 levels increased. [37] Taken in total, the data is highly suggestive of multiple TLRs in platelets.

Figure 1.

Summary of the TLRs shown to be expressed on platelets and the effects they have on platelet function.

The majority of studies looking into the functionality of TLRs in platelets have been focused on TLR4 and TLR2 as summarized in Figure 1. For TLR4, LPS stimulated platelets from mice deficient in TLR4 would not adhere to fibrinogen, suggestive that through TLR4, LPS can modulate platelet activation. [12] After a single injection of LPS, platelets did not accumulate in the lungs of TLR4 deficient mice compared to wildtype mice. [12] Similarly, LPS injected into a different TLR4 mouse model, showed that circulating platelet numbers were not affected [38-39], nor was there an increase in serum TNF-α levels compared to WT. [38] These studies suggest that through TLR4, platelets act as inflammatory sentinels, surrounding and isolating an infection, while modulating proinflammatory cytokine release. This idea is further supported by work showing platelet TLR4 induced platelet binding to adherent neutrophils. In this setting, however, LPS primed platelets and did not directly cause the formation of heterotypic aggregates. [40] In chicken thrombocytes, the avian equivalent to platelets, TLR4 stimulation with LPS increases IL6, COX-2, and PGE₂ levels through NFκB and MAPK pathways. [41] Additionally, sCD40L [42-43] and PAF4 levels increase after platelet stimulation with LPS in a TLR4 dependent manner, however, RANTES, angiogenin, and PDGF-AB all decrease. [43] Children with enterohemorrhagic E. coli (EHEC) have platelets bound with EHEC-LPS. In vitro studies further confirmed that EHEC-LPS was binding to platelets through TLR4 and CD62 and were being activated as indicated by the increase in activate α_IIbβ₃ and fibrinogen binding. [44] All of these studies continue to support the idea of platelets having a role in inflammation. But TLR4 can also regulate the hemostatic function of platelets. LPS treatment in vitro increases platelet CD63 [43], one marker of activation, but not CD62P [12, 43], another activation marker. Through TLR4, LPS reduced the time to occlusion in an in vivo mouse thrombosis model.[45] Interestingly, in this study, adhesion of the platelets to the endothelium was shown to be dependent on the presence of neutrohils. [45] Contradictory to this study, LPS alone was shown to not activate platelets but enhanced agonist-induced aggregation through TLR4, TLR2, and MyD88. [46] Aggregation through TLR4 and 2 involved the production of NO, increase in cellular cGMP, and activation of PKG. [46] Additionaly, this study also showed LPS, alone, was able to increase ATP secretion from dense granules [46] and P-selectin (CD62P) from α granules [46], contrary to what has been previously demonstrated. [12, 43] Finally, a TLR4 polymorphism was found to be cardioprotective because individuals with this polymorphism had reduced platelet thromboxane biosynthesis which limited platelet function. [47] Therefore, these studies suggest that TLR4 and platelet function can affect cardiovascular disease as well.

The functionality of the TLR2 has also been studied in platelets, but not as extensively as TLR4. In the setting of infection, TLR2 in concert with PAR-1 and endothelial-derived CX3CL1 was shown to react to Rickettsia africae to increase sCD40L levels from platelets, which could not be reduced with doxycycline, a widely used treatment for such infections. [48] Using TLR2-/- mice, the formation of platelet-neutrophil heterotypic aggregates was unaffected by the presence of P. gingivalis, a bacteria known to be recognized by TLR2. [49] Additionally, through TLR2 and PI3K/Akt, platelets formed platelet-neutrophil heterotypic aggregates that were dependent on the presence of P-selectin on the platelet cell surface. [49] These studies further confirm the role of platelets and platelet TLRs in infection.

Unlike TLR4, studies have shown decisively that stimulation of platelet TLR2 can directly activate platelets. Activation of this receptor with Pam₃CSK4, a synthetic TLR2 ligand, results in platelet aggregation and adhesion [49-50] that is dependent on PI3K/Akt. [49] Akt signaling results in increased P-selectin levels, ROS production, and α_IIbβ₃ activation. [49] Interestingly, although the signaling cascades activated by TLR2 might be the same as thrombin, the activation of Akt, ERK1/2, and p38 occur at different times and with different levels of phosphorylation. [50] Not only does the signaling differ between agonists, the contents of the α-granule (specifically, FXIIIa, thrombospondin, fibrinogen β, gelsolin, PBP, and PF4) released by each agonist differ. [50] This work suggested that, depending on the type of agonist, platelet function could vary. Platelets stimulated with thrombin were able to form stable clots in vitro. However, stimulation with Pam₃CSK4 resulted in greater formation of heterotypic aggregates compared to thrombin. [50]

Conclusions

In summary, megakaryocytes and platelets express various TLRs, however, the function of these receptors in megakaryocytes is not completely understood. The available data suggests that TLRs are the link between thrombopoiesis and infection, as seen with modulation of platelet production during endotoxemia. TLRs on platelets have been more extensively studied. Two TLRs, TLR2 and TLR4, both have been shown to augment platelet activation and alter its function from hemostatic regulator to immune sentinel. As further work is done in this field, we will further understand not only the basic function of both megakaryocytes and platelets, but also the involvement of these cells and TLRs in various disease pathologies.

Abbreviations

TLR

toll-like receptor

NFκB

nuclear factor kappa B

MAPK

mitogen activated protein kinase

PAMP

pathogen associated molecular patterns

LPS

lipopolysaccharide

TIR

toll/interleukin-1 receptor

MyD88

myeloid differentiation primary response gene 88

IRAK1/4

interleukin-1 receptor associated kinase 1/4

TRAF6

tumor necrosis factor associated factor 6

LTA

lipoteichoic acid

WT

wildtype

ALS

anaphylaxis-like shock

PAF

platelet activating factor

TNF-α

tumor necrosis factor α

IL6

interleukin 6

COX-2

cyclooxygenase-2

PGE₂

prostaglandin E₂

sCD40L

soluble CD40 ligand

RANTES

regulated upon activation, normal T cell expressed and secreted

PDGF-AB

platelet-derived growth factor-AB

EHEC

enterohemorrhagic E. coli

NO

nitric oxide

cGMP

cyclic guanosine monophosphate

PKG

protein kinase G

ATP

adenosine triphosphate

PAR-1

protease activated receptor-1

PI3K

phosphoinositide 3 kinase

ROS

reactive oxygen species

ERK1/2

extracellular signal related kinase 1/2

FXIIIa

factor XIII activated

PBP

platelet binding protein

PF4

platelet factor-4