Platelet Bacterial Interactions in the Pathogenesis of Infective Endocarditis — Part II: The Staphylococcus

Infective Endocarditis is a microbial infection characterised by the presence of septic vegetations on the surface of the endocardium (Moreillon and Que, 2004). Infection most commonly occurs on the heart valves that has been damaged by congenital defects such as previous disease or trauma (Durack, 1995). As a result these sites have the ability to gener‐ ate turbulent blood flow which in turn can cause damage to inner most lining of the blood vessels, the endothelium, which causes surface damage leading to exposure of underlying matrix protein (Ruggeri, 2009). Once exposed this highly thrombogenic surface leads to rapid platelet deposition and the formation of a fibrin network. Circulating bacteria from a transient bacteremia in turn binds to this sterile platelet fibrin nidus which allows a secondary accumulation of platelets that encase the bacteria leading to stable thrombus formation (Moreillon and Que, 2004).


Introduction
Infective Endocarditis is a microbial infection characterised by the presence of septic vegetations on the surface of the endocardium ( Moreillon and Que, 2004). Infection most commonly occurs on the heart valves that has been damaged by congenital defects such as previous disease or trauma (Durack, 1995). As a result these sites have the ability to generate turbulent blood flow which in turn can cause damage to inner most lining of the blood vessels, the endothelium, which causes surface damage leading to exposure of underlying matrix protein (Ruggeri, 2009). Once exposed this highly thrombogenic surface leads to rapid platelet deposition and the formation of a fibrin network. Circulating bacteria from a transient bacteremia in turn binds to this sterile platelet fibrin nidus which allows a secondary accumulation of platelets that encase the bacteria leading to stable thrombus formation (Moreillon and Que, 2004).
Despite improvements in medical and surgical therapy, invasive staphylococcal disease causing infective endocarditis is still associated with a severe prognosis and remains a significant therapeutic challenge. Once a disease primarily affecting younger patients presenting with rheumatic heart disease, modern times see a significant increase in newer 'at risk' categories including patients with long term indwelling central venous catheters, patients undergoing haemodialysis and invasive intravascular procedures such as arthroplasty, immunocompromized patients and intraveneous drug abusers (Thuny et al., 2012). Treatment of infective endocarditis usually requires a multidisciplinary approach involving specialists in infectious disease, cardiologists and cardiac surgeons. Current treatment regimes consist of aggressive prolonged antibiotic therapy, frequently combined with surgery (Prendergast andTornos, 2010, Wilson et al., 2007). Prolonged antibiotic use is often less than successful as 40% of patients relapse within 2 months of finishing clinically effective therapy. Furthermore, prolonged exposure to antibiotics leads to a greater risk of adding to the global problem of multiple antibiotic resistant strains of bacteria. Surgery is a costly and risky alternative, however necessary in up to 47% of patients (Castillo et al., 2000, Murdoch et al., 2009). In many cases surgery is not preferable due to risks associated with cardiac failure, further spread of infection leading to persistent sepsis due to surgical removal of an infected thrombus and/or life threatening embolisation (Jault et al., 1997, Heiro et al., 2000, Thuny et al., 2012, Remadi et al., 2007).

The Staphylococcus
Staphylococcus aureus is a gram positive pathogen that continues to cause a significant number of community-acquired and nosocomial infections. It is a normal commensal of the human body and usually lives in harmony with its host without causing symptoms. Its primary habitat is the anterior nares in 20% of the population and is transiently associated with the rest (Foster, 2009). The success of S. aureus as an opportunistic pathogen is due in part to its expression of a wide array of microbial surface components recognising adhesive matrix molecules (MSCRAMM's) (Patti et al., 1994). Using these MSCRAMM's S. aureus uses a multitude of mechanisms to attach either directly or indirectly to host cells including platelets (O' Brien et al., 2002, Kerrigan et al., 2008, Miajlovic et al., 2010, Fitzgerald et al., 2006, Pawar et al., 2004, George et al., 2006. It is for this reason that S. aureus is now the most common and most virulent etiologic pathogen in infective endocarditis.

Platelets play a critical role in thrombosis and haemostasis
Platelets are small anucleate cell fragments of the larger haematopoietic precursor cell, the megakaryocyte ( Thon and Italiano, 2010) and are crucial mediators of haemostasis. Platelets have no control over gene expression as they do not possess a nucleus however they have got limited capabilities in translational protein synthesis (Lindemann et al., 2001b). The primary role of platelets in haemostasis is to police the integrity of the endothelium to prevent blood loss (Nieswandt et al., 2009). Platelets circulate close to the endothelial cell surface at high shear as individual entities that ordinarily do not interact with any other cell types. A transition from this resting state to an activated state can be rapidly initiated if platelets are exposed to an appropriate stimulus. Disruption of the endothelial cell lining due to trauma or injury to the vascular endothelium platelets rapidly accumulate at the site of injury (Gawaz et al., 2005). Recruitment is a highly controlled event that is initiated by the adhesive interaction between the exposed extracellular matrix proteins in damaged endothelium and specific membrane receptors on the platelet (Tabuchi and Kuebler, 2008). Collagen (Santoro and Zutter, 1995), vonWillebrand factor (vWf) (Ruggeri, 1999), fibronectin (Savage et al., 1998, Kasirer-Friede et al., 2007 and thrombospondin (Jurk et al., 2003) constitute the exposed matrix proteins at the site of injury. Athough plasma proteins such as fibrinogen/fibrin and vitronectin are not synthesized by endothelial cells they can bind to exposed matrix proteins and increase adhesiveness at the damaged site (Ruggeri et al., 2006, Ruggeri andMendolicchio, 2007).
Platelets express a vast array of membrane receptors that play a critical role in recognition of matrix proteins. The initial interaction of platelets with the injured vessel wall occurs between GPIbα and immobilised vonWillebrand factor (Chesterman and Berndt, 1986). This interaction initiates the tethering of circulating platelets to the vessel wall. Platelets typically 'roll' over the vWf in the direction of flow driven by shear forces experienced by the vasculature (Ruggeri, 2009). A loss of interaction between GPIb and vWf on one side of the platelet leads to the formation of another GPIb-vWf interaction on the other side of the platelet which gives rise to a rolling phenomenon. This rolling mechanism is critical to slowing down the platelet long enough for a second interaction that anchors the platelet to the damaged site. This firm adhesion can be mediated by several membrane receptors, some of which will have become activated as a result of platelet rolling and others who are expressed on the platelet surface as a result platelet activation (Jackson et al., 2009). Once firmly adhered, the platelets rearrange cytoskeletal components which results in filopodia and llamelipodia extension leading to flattening or spreading of the platelet. Platelet spreading is critical following firm adhesion as it firstly allows the platelet withstand the shear forces experienced in the vasculature and secondly it increases the platelet surface area thus covering more of the damaged site.
Following attachment, platelets undergo a series of highly controlled intracellular signalling events that lead to the release reaction where platelets release the contents of its stored intracellular granules. Alpha granules contain proteins such as P-selectin which mediates adhesion of platelets to monocytes, neutrophils and lymphocytes, resulting in the formation of platelet leukocyte complexes (Diacovo et al., 1996a, Diacovo et al., 1996b, Larsen et al., 1989. These granules also contain many chemotactic agents which lead to the recruitment of various inflammatory cells; platelet derived growth factor (PDGF) and 12-hydroxyeicosatetraenoic acid (12-HETE) which recruit neutrophils (Herd andPage, 1994, Mannaioni et al., 1997); platelet factor 4 and platelet derived histamine releasing factor (PDHRF) which recruit eosinophils in airway disease (Brindley et al., 1983, Frigas andGleich, 1986); PDGF and transforming growth factor β (TGF-β) which recruit monocytes and macrophages and TGFβ which recruits fibroblasts (Deuel et al., 1982, Tzeng et al., 1985, Wahl et al., 1987. Platelet granules also contain several mediators of tissue damage such as oxygen free radicals and hydrolytic enzymes. Dense granules release cationic proteins that initiate vascular permeability and mediators that enhance aggregate formation such as adenosine diphosphate (ADP) and serotonin (5-HT) (Rendu and Brohard-Bohn, 2001). Bioactive amines are also secreted from platelets following activation including Thromboxane A 2 (TxA2) and platelet activating factor Once activation is complete the platelet forms a new surface for additional platelets to adhere, predominantly through GPIIb/IIIa crosslinking adjacent platelets through a fibrinogen bridge, resulting in aggregate formation. The final step sees and effective plug at the site of injury that is reinforced by the conversion of fibrinogen to fibrin through the coagulation cascade (Ruggeri et al., 2006).

The growing role of platelets in infection and immunity
Platelets are poorly recognised for their role in infection and immunity even though just like professional phagocytes (neutrophils, macrophages and dendritic cells) platelets are derived from the same haematopoietic stem cell, undergo chemotaxis (

Mechanisms of interaction
Bacteria can interact with platelet in two ways, they can either support platelet adhesion or they can induce platelet aggregation. Platelet adhesion to immobilised bacteria is a measure of the strength of the interaction, whereas platelet aggregation induced by bacteria is a measure of the quality of the interaction. In contrast to typical platelet aggregation induced by physiological agonists such as adenosine diphosphate (ADP), collagen or thrombin, bacteria induce an all or nothing response. This means that the bacteria either induce a maximal aggregation or they don't induce platelet aggregation at all, there is no intermediate response. Another unique feature of bacteria induced platelet aggregation is a distinct pause in time before aggregation takes place. This is typically called the lag time. Increasing the concentration of bacteria shortens the lag time but never eliminates it. The average lag time to platelet aggregation following addition of Staphylococci is between 5-12 minutes. This is in contrast to the lag time observed upon the addition of typical platelet agonists ADP or thrombin which have a lag time less than 10 seconds.
There are 3 main interactions between bacteria and platelets. In the first interaction bacteria express proteins that can directly interact with a surface receptor on the platelet. In this case the bacterial protein express ligand mimetic domains that act as agonists on the platelet receptor thus triggering an intracellular signal that culminates in platelet activation. In the second interaction bacterial proteins bind a plasma protein that is a natural ligand for a platelet receptor. For example, bacteria can bind antibody which in turn bridges the bacteria to the antibody receptor (FcγRIIa) expressed on the platelet. Once engaged the receptor results in the generation of an intracellular signal leading to platelet activation. Finally bacteria may have the ability to secrete products or toxins that in turn activate platelets. Engagement of the product or toxin with a platelet receptor results in activation. These different mechanisms of interaction may help explain the lag time to platelet aggregation. For example, the lag time could be representative of the time taken to trigger a response or bind a plasma protein. A major challenge in studying platelet bacterial interactions is that most bacteria can interact with platelets using multiple mechanisms. This makes it incredibly difficult to identify either the platelet receptors or the bacterial proteins involved in triggering thrombus formation. Moreover not only are the interactions species specific but strain specific as well.

Indirect interaction (Released products)
Staphylococcus aureus was one of the first bacteria isolated from patients with acute endocarditis. Despite improvements in medical and surgical therapy, invasive staphylococcal disease causing infective endocarditis is still the most frequent etiologic microorganism found in patients (Rasmussen et al., 2011). Studies investigating the mechanism through which S. aureus contributes to endocarditis dates back as far as the early 1900's. By the mid 1900's significant attention had been placed on the involvement of S. aureus alpha (α)-toxin in contributing to IE. Alpha-toxin is produced by almost all strains of S. aureus. Typically α-toxin disrupts the cell membrane by binding to the lipid bilayer of platelets, erythrocytes and some leukocytes, forming an oligomeric structure that forms a water filled transmembrane pore. In 1964 Siegel and Cohen made two critical observations; first, addition of α-toxin led to the loss of single platelets as evidenced by turbidimetric aggregometry and second that addition of α-toxin to human platelets resulted in leakage of intracellular ions; NAD+, K+ and ATP but interestingly not protein, suggesting that α-toxin was not lysing the platelets (Siegel and Cohen, 1964). Further studies by Bernheimer and Schwartz confirmed these reports and demonstrated by electron microscopy that following exposure to α-toxin platelets swelled but did not show signs of lysis (Bernheimer, 1965). These early studies suggested that α-toxin may have the ability to generate a signal upon binding to the platelet. Focusing on this Arvand and colleagues demonstrated that α-toxin did indeed trigger a platelet signal upon binding and most importantly one that leads to secretion of intracellular contents including procoagulant mediators, platelet factor 4 and factor V. Secreted factor V in turn associates with the platelet membrane leading to assembly of the prothrombinase complex (Arvand et al., 1990). This explains the major pathway responsible for the procoagulatory effects of α-toxin. In contrast to these early findings Bayer et al. demonstrated that α-toxin did cause platelet lysis and this led to the release of platelet microbial proteins (PMP's) which was bactericidal to S. aureus.
Using an animal model of endocarditis the authors demonstrated that different strains of S. aureus differed in the expression of functional versus mutant forms of α-toxin. Under these conditions, the S. aureus strains producing either minimal or no α-toxin were less virulent in vivo than wild-type strains . Wild-type S. aureus strains or indeed an isogenic strain engineered to over-express α-toxin were associated with increased release of PMP from platelets. These results suggest that when S. aureus releases α-toxin, platelets release PMP's therefore leading to a protective role for the host by destroying S. aureus. . It remains to be seen whether TLR2 is mediating this inhibition of platelet signalling.

Indirect interaction (Cell wall proteins)
There are numerous cell wall proteins expressed on the surface of S. aureus that have been demonstrated to bind to platelets and trigger platelet activation. The majority of these cell wall proteins have been found to bind plasma proteins and bridge to a platelet receptor. Staphylococcal protein A is a widely expressed protein found on greater than 90% of S. aureus strains. In 1979, Hawiger et al., demonstrated that protein A is capable of binding to immunoglobulin G (IgG) which in turn bridges to the platelet antibody receptor, FcγRIIa. This interaction results in platelet signal generation, GPIIbIIIa dependent platelet aggregation and serotonin release from the platelet dense granules. Interestingly purified protein A failed to cause measureable platelet aggregation and release of serotonin was significantly reduced (Hawiger et al., 1979). Recent studies have demonstrated that protein A can bind to the A1 domain of the major plasma protein vonWillebrand factor with high affinity (low nM range) (O' Seaghdha et al., 2006) which serves as a receptor for GPIbα expressed on platelets (Andrews et al., 2003). More recent studies have investigated this interaction under fluid shear conditions and demonstrated that preincubating platelet rich plasma with a vonWillebrand Factor antibody or indeed blocking the platelet GPIbα receptor with an inhibitory monoclonal antibody partially inhibited the platelet-S. aureus interaction. Furthermore using a strain of S. aureus that is deficient in protein A expression reduced its interactions with platelets (Pawar et al., 2004).These results suggest that protein A plays a role in triggering platelet activation.
More recent studies demonstrated that multiple cell wall proteins expressed on S. aureus are capable of interacting with and triggering platelet aggregation (O' Brien et al., 2002). Among the cell wall proteins identified, clumping factor A (ClfA) and clumping factor B (ClfB) are possibly the most extensively studied. ClfA and ClfB has been shown to bind a number of plasma proteins including fibrinogen, IgG and complement, which in turn bridge the bacteria to specific platelet receptors and trigger activation. Binding each plasma protein individually is not sufficient to trigger platelet aggregation. To trigger full platelet activation ClfA or ClfB must bind specific IgG along with either fibrinogen or complement, IgG being the key molecule. There are at least two distinct sites on each ClfA or ClfB that allows IgG and fibrinogen bind at the same time. Once this occurs, fibrinogen binds to platelet GPIIbIIIa, IgG binds to platelet FcγRIIa and together induces receptor clustering leading to activation of signal transduction pathways culminating in platelet aggregation (Loughman et al., 2005, Miajlovic et al., 2007. Deletion of the fibrinogen binding domain on ClfA or ClfB led to the discovery of another much slower platelet aggregation (8-20 minutes versus 2-4 minutes). Loughman et al initially demonstrated that complement must assemble on the S. aureus surface and then cross link to complement receptors expressed on platelets. Similar to before, IgG binds to FcγRIIa, complement proteins binds to complement receptors on platelets and together induces receptor clustering leading to activation of signal transduction pathways culminating in platelet aggregation (Loughman et al., 2005). Much controversy surrounds the existence of complement receptors on platelets however the most convincing evidence of a complement receptor is that demonstrated by Nyugen et al., who demonstrated the expression of gC1qR/p33 following platelet activation. This suggests that an initial interaction leads to platelet activation which in turn triggers expression of gC1qR/p33 on the platelet surface. Expression of this receptor post activation most likely serves to anchor the bacteria to the platelet.
A critical part of S. aureus survival in the host is the wide array of cell wall proteins it expresses at various growth phases of its cell cycle. For example, ClfA is weakly expressed during the exponential phase and strongly expressed during the stationary phase. In contrast to this fibronectin binding protein A (FnbpA) is strongly expressed during the exponential phase of growth and weakly expressed during the stationary phase of growth. FnbpA also plays a key role in inducing platelet aggregation. The mechanism through which FnbpA induces platelet aggregation is more or less identical to the mechanism that ClfA uses to induce platelet aggregation. Fnbp contain a specific immunoglobulin binding domain (A domain) and a separate fibronectin binding domain (BCD). FnBPA possesses two different but related mechanisms of engaging and activating platelets (Fitzgerald et al., 2006). In the first mechanism, fibrinogen can bind to the A domain which crosslinks to GPIIb/IIIa, and specific immunoglobulin must crosslink to FcγRIIa to trigger platelet activation and aggregation (Fitzgerald et al., 2006). In the second mechanism fibronectin can bind to S. aureus via the FnBPA BCD domain (Meenan et al., 2007, Raibaud et al., 2005. The signal to trigger platelet activation/aggregation is complete when specific immunoglobulin binds the A domain of FnBPA and cross links to platelet FcγRIIa inducing receptor clustering.
As discussed in chapter 2 serine rich proteins expressed by viridans streptococci play a critical role in inducing platelet aggregation. S. aureus also expresses a highly glycosylated serine rich protein called SraP on its surface (Siboo et al., 2005). Strain of S. aureus deficient in expression of SraP has been shown to have reduced virulence in a rabbit model of endocarditis. Regardless of the fact that SraP shares significant similarities with a number of other serine rich glyco-proteins found in the streptococci that have been found to bind to platelet GPIbα (Kerrigan et al., 2007, Plummer et al., 2005, SraP does not appear to bind to this platelet receptor. While all of these studies are critical to our understanding of the molecular mechanisms involved in aggregate formation, one must be aware of the relevance of these findings to physiological conditions experienced in the vasculature. For example, almost all of the studies carried out to date have been carried out under non-physiological stirring or using static adhesion assays, neither of which are representative of the conditions experienced in the vasculature. Many reports in the literature in recent times have clearly demonstrated that the local fluid environment in the circulation critically affects the molecular pathways of cell-cell interactions (Varki, 1994). To address this several attempts have been made to create an environment more representative of conditions experienced in the circulation. Rheology is a useful technique that can be employed to shear cells at physiological rates. Using a cone and plate viscometer, Pawar et al. demonstrated that when S. aureus is mixed with whole blood isolated from a healthy individual thrombus formation could be observed. Additional studies demonstrated that the thrombus formation was dependent on multiple S. aureus cell wall proteins including protein A, ClfA, SdrC, SdrD, SdrE (Pawar et al., 2004). A potential limitation to using a cone and plate viscometer is that is measures thrombus formation in a soluble setting and it is well established that thrombus formation on a heart valve in IE occurs under stable conditions. To address this Kerrigan et al. developed a parallel flow chamber to assess the interaction between S. aureus and platelets in whole blood. To do this S. aureus was immobilised on a glass slide (to mimic the focal infection on a heart valve) and whole blood was perfused over the bacteria at both arterial and venous shear rates. This method demonstrated that platelets perfused over immobilised S. aureus under arterial shear led to a very strong adhesion, followed by rapid aggregate formation. Deletion of ClfA (but not protein A or FnbpA) from S. aureus abolished adhesion and subsequent aggregate formation. Using a plasma-free system, fibrinogen led to single platelet adhesion but not aggregate formation. Specific immunoglobulin failed to have any effect on either platelet adhesion or aggregation. However, addition of fibrinogen and specific immunoglobulin together to the plasma-free system led to platelet adhesion followed by aggregate formation thus highlighting the importance of fibrinogen and IgG in aggregate formation. Interestingly platelets did not adhere to or induce aggregate formation under low shear conditions using the parallel flow chamber (Kerrigan et al., 2008).

Direct interaction (Cell wall proteins)
A growing concern about studies to date is the apparent lack of contrast with conditions experienced physiologically. In vivo, when S. aureus enters the blood stream it is in an environment where iron is sequestered in haem or haemoglobulin. The lack of iron available in vivo inactivates the Fur repressor in S. aureus that results in an up-regulation of a number of genes that typically wouldn't be expressed in when growing in normal laboratory bacterial growth media. A growing family of iron-regulated surface determinant proteins have been recently identified as expressed in S. aureus grown in iron limited conditions. Using surface plasmon resonance, Miajlovic et al. demonstrated that one family member, iron-regulated surface determinant B (IsdB), can bind directly (in the absence of plasma proteins) to the purified platelet fibrinogen receptor GPIIb/IIIa. As a result of this binding subsequent studies demonstrated the ability of the wildtype S. aureus strain to support platelet adhesion and induce platelet aggregation. A. S. aureus strain defective in expression of IsdB displayed a reduce ability to adhere to or induce platelet aggregation (Miajlovic et al., 2010).

Final thoughts and future directions
Infective endocarditis is notoriously difficult to treat as antibiotics are incapable of penetrating the growing thrombus to reach the encased microorganisms. As a result of this the in-hospital mortality rate can be as high as 36% (Botelho-Nevers et al., 2009). Even with treatment, 40% of patients with infective endocarditis relapse within 2 months of finishing clinically effective therapy (Netzer et al., 2002). Furthermore, approximately 25% of patients with infective endocarditis eventually require surgery, usually within 2 years after completing therapy (Olaison and Pettersson, 2003). These statistics reflect the poor delivery and penetration of antibiotic into the growing thrombus. The costs associated with hospitalization (of which the average stay in hospital is 30 days), surgery and prolonged antibiotic treatment is extremely high placing a severe burden on already over-stretched healthcare systems though out the world. The danger of S. aureus invasive disease is also compounded by the rapidly increasing global widespread occurrence of multiple antibiotic resistant strains (MRSA and VRSA) which is directly attributed to prolong use of antibiotics. The greater the duration of exposure of an antibiotic to bacteria, the greater the risk of development of resistance and this is irrespective of the severity of the need for antibiotics. If this is not addressed soon, acquired resistance may produce a virtually untreatable pathogen. Therefore it is of the utmost importance that we understand the molecular interactions that lead to the development of thrombus formation on the heart valves. This will serve two purposes, first it may lead to the development of novel therapies that will prevent the formation of a thrombus on the heart valve and secondly as a result will overcome the problem associated with getting clinically effective concentrations of antibiotic to the site of infection on the heart valve.
Potential drug targets identified from studies over the years suggest that blocking the interaction between IgG and platelet FcγRIIa may indeed prevent platelet receptor clustering and thus inhibit thrombus formation. Blockade of the platelet FcγRIIa receptor has distinct advantages over other anti-platelet agents as inhibitors of FcγRIIa do not affect the platelet response to other agonist and therefore does not compromise essential platelet functions.