Brothers in Arms: Platelets and Neutrophils in Ischemic Stroke

Abstract

Purpose of review:

In this review, we will describe how the combined ability of platelets and neutrophils to interact with each other drives ischemic stroke brain injury.

Recent findings:

Neutrophils are one of the first cells to respond during ischemic stroke. While animals stroke models have indicated targeting neutrophils improves outcomes, clinical trials have failed to yield successful strategies. Platelets play a critical role in recruiting neutrophils to sites of injury by acting as a bridge to the injured endothelium. After initial platelet adhesion, neutrophils can rapidly bind platelets through P-selectin and glycoprotein Ibα. In addition, recent data implicated platelet phosphatidylserine as a novel key regulator of platelet-neutrophil interactions in the setting of ischemic stroke. Inhibition of procoagulant platelets decreases circulating platelet-neutrophil aggregates and thereby reduces infarct size. Platelet binding alters neutrophil function, which contributes to the injury associated with ischemic stroke. This includes inducing the release of neutrophil extracellular traps (NETs), which are neurotoxic and pro-thrombotic, leading to impaired stroke outcomes.

Summary:

Platelet-neutrophil interactions significantly contribute to the pathophysiology of ischemic stroke brain injury. Better understanding the mechanisms behind their formation and the downstream consequences of their interactions will lead to improved therapies for stroke patients.

Introduction

Annually, 800,000 Americans suffer from stroke and for 1 in 6 stroke patients this will be fatal. Globally, 15,000,000 individuals have a stroke each year with 5,000,000 stroke related deaths occuring¹. The vast majority of strokes are ischemic and these are typically caused by a blood clot obstructing cerebral blood flow. Current acute ischemic stroke therapy focusses on the removal of the culprit thrombus via either pharmacological thrombolysis or mechanical thrombectomy. Due to a limited time window and other contra-indications, these therapies are only available for a small group of stroke patients². Moreover, even when thrombolytic therapy or thrombectomy is initiated, this is often not effective. Indeed, successful thrombolysis occurs only in approximately 30% of treated patients³ and thrombectomy does not guarantee improved stroke outcomes, even when the thrombus is successfully removed⁴.

Once cerebral blood flow stops and ischemia occurs, a dynamic process takes place in the brain initiating a detrimental cascade of inflammation, microvascular thrombosis and neurotoxicity. While successful restoration of cerebral blood flow is critical to improve stroke outcomes, preclinical evidence suggests this might not be sufficient to maximally salvage brain tissue. In an effort to improve stroke outcomes, several studies have investigated neuroprotection via either anti-inflammatory⁵ or anti-thrombotic strategies⁶. However, none of these approaches have successfully translated to the clinic⁷. Limitations associated with anti-inflammatory strategies include post-stroke immunosuppression and an increased risk for lethal infections in stroke patients. Additionally, the human brain is very sensitive to bleeding associated with classic anti-thrombotic drugs. Importantly, recent evidence suggests in particular the interaction between inflammation and thrombosis is a key contributor to ischemic stroke brain injury⁸. Interestingly, conventional anti-inflammatory and anti-thrombotic drugs don’t interfere with these interactions. This highlights the need for new management strategies to target thrombo-inflammation in ischemic stroke.

Platelet-neutrophil interactions are at the interface between inflammation and thrombosis. In several thrombo-inflammatory disorders platelet-neutrophil aggregates (PNAs) have emerged as markers for disease severity and potential therapeutic targets⁹. Similarly, in ischemic stroke PNAs are increased acutely^10–13, irrespective of use of anti-platelet drugs^14,15. However, circulating PNAs don’t correlate with stroke severity^11,14 or stroke etiology.^12,13 In this review, we will discuss recent findings on novel ways platelets and neutrophils interact during ischemic stroke and how these interactions exacerbate ischemic stroke brain injury.

Neutrophil Activation and Recruitment in Ischemic Stroke

Neutrophils are among the first cells in the blood to respond after ischemic stroke, with the number of circulating neutrophils rising within the first few hours of stroke onset¹⁶. The increase in neutrophils contributes to disruption of the blood brain barrier (BBB), cerebral edema, and brain injury. Importantly, the increase in circulating neutrophils is associated with stroke severity, infarct volume, and worse functional outcomes^17–19. Furthermore, the level of neutrophil accumulation in regions of brain ischemia correlates with stroke severity and worse stroke outcome in humans and mice^20,21. While these studies indicate neutrophils are critical regulators of ischemic stroke injury, modulators of their activation and recruitment to the injured stroke brain remain unclear.

Many studies have focused on the role of the endothelium in regulating neutrophil recruitment and transmigration. Neutrophils express numerous endothelial adhesion receptors important in promoting neutrophil tethering and rolling on endothelial cells including P-selectin glycoprotein ligand-1 (PSGL-1), lymphocyte function associated antigen 1 (LFA-1) and macrophage-1 antigen (MAC-1)^22,23. Many of these receptors increase shortly after ischemia thus promoting neutrophil adhesion and infiltration^24–26. While many of these targets have been successful inhibited in animal models of ischemic stroke^27,28, the translation to human studies has been less fruitful. Clinical studies targeting intracellular adhesion molecule 1 (ICAM-1), a key endothelial cell receptor regulating neutrophil binding, MAC-1, and LFA-1 have all failed in various clinical trials with results ranging from no benefit compared to placebo controls to increased rates of infection^29–31. Given the importance of neutrophils in promoting neuroinflammation, resulting in worse outcomes in ischemic stroke injury, alternative mediators regulating neutrophil recruitment to the brain are important to consider.

Platelet-Dependent Neutrophil Recruitment: A Missing Link

Platelets play a critical role in the formation of obstructive thrombi causing stroke with robust clinical data demonstrating acute reductions in recurrent ischemic stroke with the combination of antiplatelet agents aspirin and clopidogrel. Long-term benefits have also been demonstrated with aspirin, aspirin-dipyridamole, and clopidogrel³². Additionally, platelets exert pathological effects in the ischemic stroke brain that go beyond classical thrombus formation, which is the focus of this review. Similar to neutrophils, platelets are capable of interacting with the endothelium upon activation. Upon exposure of subendothelial matrix or activation of the endothelium platelets will quickly adhere. Importantly, the efficiency of platelet adhesion increases in the presence of high shear rates, which is common in cerebral arteries affected by stroke^33,34. The principal regulator of platelet adhesion to injured or activated endothelium occurs through platelet glycoprotein Ibα binding to endothelial cell-bound von Willebrand factor (VWF), which is favored during high shear environments³⁵. In contrast, direct neutrophil interactions with the endothelium are less favored during high shear environments³⁶. Thus, the high shear environment of the cerebral arteries is primed for platelets to facilitate neutrophil adhesion to activated or injured endothelium. Once platelets are bound to the endothelium, platelets trap neutrophils using multiple receptors including platelet P-selectin binding to neutrophil PSGL-1³⁷ and platelet glycoprotein Ibα binding to neutrophil MAC-1^38,39 (Figure 1). Furthermore, neutrophil adhesion to activated platelets can proceed through flow-induced protrusions or FLIPRs that extend from adhered platelets and mediate neutrophil binding in in a P-selectin/PSGL-1-dependent manner⁴⁰. In addition to P-selectin, other platelets receptors including αIIbβ3⁴¹ and ICAM-2⁴² are also critical in forming platelet-neutrophil aggregates by binding MAC-1 and LFA-1, respectively. Recently, a novel interaction between platelet αIIbβ3 and neutrophil SLC44A2 was identified⁴³, which promotes venous thrombosis^44,45. Interestingly, this interaction specifically occurs under flow and is VWF dependent, implying it could also play a role in ischemic stroke^43,46. These findings highlight the need to further identify novel platelet-neutrophil binding partners that contribute to thrombosis.

Figure 1. Conventional and Novel Ligands Involved in Platelet-Neutrophil Aggregate Formation.

Platelets bind to neutrophils using classical receptors including α_IIbβ₃ and GPIbα which bind to MAC-1 as well as ICAM-2 which binds to LFA. Platelet alpha granule release results in P-selectin expression and the ability of the platelet to bind PSGL-1 on neutrophils. Besides these well-established pathways, the novel platelet ligand phosphatidylserine (PS) mediates platelet-neutrophil interactions through an unknown receptor to contribute to ischemic stroke outcomes. Neutrophil SLC44A2 has recently been shown to bind to platelet α_IIbβ₃ to mediate platelet-neutrophil aggregate formation.

Novel Regulators of Platelet-Neutrophil Aggregate Formation: Procoagulant Platelets

One particular subset of platelets that preferentially interacts with neutrophils are procoagulant platelets. Distinctive features of procoagulant platelets include high levels of externalized phosphatidylserine (PS), inactivation of integrin αIIbβ3 and their ability to generate microparticles⁴⁷. These features result in a unique phenotype of platelets that can drive coagulation on their surface, but are less able to aggregate with other platelets. In recent years, an increasing body of work has described a predisposition of procoagulant platelets to interact with neutrophils and mediate inflammation and subsequent tissue injury^21,40,48,49. Interestingly, ischemic stroke patients have increased levels of circulating procoagulant platelets^50,51 and platelets from ischemic stroke patients have an increased ability to become procoagulant when tested in vitro^52–57. Moreover, increased in vitro procoagulant platelet responses are associated with stroke severity and outcomes⁵⁷, and have been shown to predict stroke recurrence^53–56. Importantly, experimental stroke models have shown that targeting procoagulant platelets is protective^21,58, implying a causal role between procoagulant platelet formation and ischemic stroke outcomes.

Procoagulant platelets are formed in response to a strong stimulus. In vitro this stimulus can be mimicked by dual agonist stimulation with thrombin and a GPVI agonist (i.e., convulxin or collagen related peptide). In stroke, the factors initiating procoagulant platelet formation are yet to be determined. Most likely, this is a combination of collagen exposed on damaged endothelium⁵⁹ and locally increased levels of circulating thrombin⁶⁰. Procoagulant platelet formation can then be potentiated by reactive oxygen species (ROS)⁶¹ released by ischemic brain tissue or by high shear⁶² occurring in the reperfused arteries, resulting in enhanced platelet PS exposure. Furthermore, we recently found that hyperglycemia, a common comorbidity for stroke patients, also potentiated procoagulant platelet formation and subsequent PNA formation in a mouse model of stroke⁶³.

Importantly, a causal role for procoagulant platelets in the exacerbation of ischemic stroke outcomes was established in murine ischemic stroke models. Indeed, mice with platelets lacking the ability to become procoagulant (platelet specific cyclophilin D KO mice) were protected from ischemic stroke brain injury²¹. Similarly, when procoagulant platelets were targeted with AnnexinA1⁵⁸ or lactadherin⁶⁴, ischemic stroke brain injury could be blocked. Mechanistically, these studies found that procoagulant platelets mediated the formation of detrimental platelet-neutrophil interactions in the reperfused ischemic brain. These interactions mediated trafficking of neutrophils to the brain, reduced local cerebral blood flow and thereby exacerbate stroke outcomes^21,58. One clinical study recently underscored these observations. When procoagulant platelets were isolated from ischemic stroke patients and incubated with healthy neutrophils, these platelets induced neutrophil extracellular traps (NETs), which subsequently initiated coagulation and induced endothelial cell death in vitro⁵¹. Combined, these studies highlight the detrimental effects of the interaction between procoagulant platelets and neutrophils in the ischemic brain.

The mechanisms through which procoagulant platelets interact with neutrophils have yet to be elucidated; however, targeting either platelet P-selectin⁴⁸ or PS on the platelet surface using recombinant AnnexinA1⁴⁸ was shown to abrogate interactions between procoagulant platelets and neutrophils.

Downstream Consequences of Platelet-Neutrophil Aggregate Formation

Platelet recruitment and activation of neutrophils alters the hemostatic balance in the brain to promote inflammation and create a procoagulant environment (Figure 2). Neutrophil activation results in the release granules containing cathepsin G, elastase and matrix metalloproteinases (MMP), all which disrupts the BBB^65–67. Besides inducing alterations in the BBB, cathepsin G and elastase also are important regulators of coagulation. Both enzymes inactivate natural anticoagulants such as tissue factor pathway inhibitor, thrombomodulin, and antithrombin, thus promoting a procoagulant environment^68–70. Cathepsin G also activates protease activated receptor-4 on endothelial cells⁷¹ and platelets⁷² to further amplify platelet activation and adhesion. In addition, platelet adhesion to neutrophils results in the generation and release of ROS from neutrophils⁷³. ROS has cytotoxic effects on neural and endothelial cells, leading to increased BBB permeability⁷⁴.

Figure 2. Platelet-Neutrophil Aggregate Formation Promotes Inflammation and a Procoagulant Environment After Ischemic Stroke.

During ischemic stroke, endothelial von Willebrand factor (VWF) traps platelets under high shear stress resulting in release of platelet alpha granule content, including P-selectin. Furthermore, this environment promotes activation of platelet integrins and PS expression. The combination of platelet alpha granule release, integrin activation, and PS expression fosters the formation of platelet-neutrophil aggregates. Platelet binding activates neutrophils to release cathepsin G (CatG), neutrophil elastase (NE), and matrix metalloproteases (MMPs), which alter coagulation and propagate additional platelet activation and disruption of the blood-brain barrier (BBB). In addition, platelets can induce intravascular neutrophil extracellular traps (NETs) resulting in further recruitment of platelets and neutrophils to the ischemic brain. Platelet binding to neutrophils also induces transmigration of neutrophils into the ischemic brain tissue where extravascular neurotoxic NET formation can occur.

Platelets binding to neutrophils also promote the formation of NETs^75,76. NETs are extracellular DNA lattices generated upon neutrophil activation⁷⁷. While NETs primarily are thought to trap bacteria and pathogens to help fight infection, they are neurotoxic when they are formed in the brain parenchyma^78,79. NETs appear to promote thrombotic complications in animal models as they trap other blood cells including platelets and neutrophils as well as serve as a scaffolding to support thrombin generation⁸⁰. This combination leads to decreased blood flow and further exacerbates ischemia in the tissue. Mice deficient in PAD4, an enzyme critical for a histone modification required for chromatin decondensation, are unable to make NETs and are protected in a venous thrombosis model⁸¹. Dissolving NETs after their formation using DNase also reduces thrombosis, including during ischemic stroke⁸².

Platelet-neutrophil aggregates, themselves, also regulate blood flow and ischemia. Using intravital microscopy, Yuan and colleagues were able to demonstrate that rolling neutrophils extract large membrane fragments from procoagulant platelets in multiple organs after gut ischemia-reperfusion injury⁴⁸. In their model, procoagulant platelets bridged neutrophils to form macroaggregates and subsequently plug vessels in distal organs. Plugging of the cerebral microvasculature is a common observation in both clinical and experimental stroke and contributes to the no-reflow phenomenon after reperfusion, but also exacerbates outcomes in permanent stroke models^83–87. In line with this, a recent report observed platelets and neutrophils in high shear environments alter blood rheology to promote microvascular clotting, which could be critical in ischemic stroke injury⁸⁸.

Targeting Platelet-Neutrophil Aggregates Therapeutically in Stroke

Platelet-neutrophil aggregates are critical in the pathophysiology of ischemic stroke; therefore, therapeutic targeting may provide clinical benefit to stroke patients. Previous studies have directed therapies against PSGL-1 and P-selectin in the setting of myocardial infarction (MI). However, these studies demonstrated little benefit towards MI outcomes^89,90. Importantly, these therapies have not been examined in the setting of ischemic stroke, which has a distinctive pathophysiology compared to MI. Indeed, whereas targeting platelet αIIbβ3 was protective in MI⁹¹, this was ineffective and induced severe bleeding complications in stroke patients.⁹² While inhibition of platelet specific receptors that bind neutrophils has not been examined in clinical trials for stroke patients, several recent preclinical studies have suggested inhibition of procoagulant platelets plays a significant role in dampening ischemic stroke injury. In animal models deletion of cyclophilin D a key regulator of procoagulant platelet formation or administration of Annexin A1 in a cerebral ischemia-reperfusion model reduces platelet PS expression, platelet-neutrophil interactions and infarct size^21,58. Furthermore, in stroke patients the administration of cyclosporine, which inhibits procoagulant platelet formation, improved outcomes in patients where reperfusion was achieved after cerebral ischemia⁹³. Besides targeting therapies against neutrophil and platelets receptors, therapeutics directed against downstream consequences of platelets-neutrophil aggregates such as NET formation could be viable as well. NET formation is a known regulator of venous and arterial thrombosis, including ischemia/reperfusion injury and NETs have previously been demonstrated to be embedded in thrombi retrieved from ischemic stroke patients^94–96. Targeting NETs could be achieved by directing agents against the DNA lattice generated when NETs are released through DNase administration⁸². Furthermore, inhibition of molecular regulators of NET formation, including PAD4, could block DNA release altogether⁸¹. Targeting procoagulant platelet formation or NET release during ischemic stroke would be more beneficial compared to targeting platelet receptors as these interventions have less impact on basic hemostatic platelet or normal neutrophil function, meaning there would be minimal risk for bleeding or infections.

Conclusion

The unsuccessful translation of anti-neutrophil strategies from bench to bedside stimulated stroke research to further elucidate the precise mechanisms by which neutrophils mediate ischemic stroke brain damage. In recent years, the focus shifted from neutrophil transmigration and neutrophil-endothelial cell interactions, to platelet-neutrophil interactions and the associated cerebral microvascular complications. Novel therapeutic interventions targeting procoagulant platelets or the formation of NETs hold promise as safe ischemic stroke therapies and warrant further investigation.

Key points:

• Neutrophils are detrimental during acute ischemic stroke brain injury but critical to fight infections after stroke.

• Platelets-neutrophil interactions are critical in mediating injurious neutrophil function in the brain after stroke.

• Targeting novel mediators of platelet-neutrophil interactions such as SLC44A2 and phosphatidyl serine hold promise for stroke treatment.