Neutrophils, NETs, and immunothrombosis

The blood coagulation system is activated by pathogens and contributes to host defense by reducing dissemination and facilitating destruction of pathogens.²  Recent studies have shown that immune cells, such as neutrophils and monocytes, actively participate in the formation of thrombi within blood vessels, a term called immunothrombosis.³  Exposure of the endothelium to pathogens and pathogen-associated molecular patterns triggers a proadhesive phenotype leading to leukocyte binding. Activated monocytes express tissue factor that initiates blood coagulation and intravascular fibrin formation. This fibrin network potentiates further recruitment of leukocytes and their activation via engagement of the integrin αMβ2 (Mac-1). Neutrophils facilitate trapping and killing of pathogens through an organized cell death pathway in which decondensed chromatin and antimicrobial proteins are expelled from the cell to form NETs.⁴ 

Immunothrombosis can also be pathologic. For instance, both arterial and venous thrombi have been shown to contain neutrophils and NETs.⁵  NETs can increase the size of thrombi by capturing platelets and microvesicles.⁶  However, it is not clear if leukocytes play a role in all forms of thrombosis or just particular types or sites of thrombosis. Mechanistic studies have demonstrated that neutrophils and NETs contribute to thrombosis in a murine inferior vena cava (IVC) stenosis model.^5,6  However, detailed analysis of the different receptor/ligand interactions involved in neutrophil recruitment, activation, and NET formation has not been described.

Neutrophils are recruited to sites of inflammation in a multistep process that involves rolling, activation, firm adhesion, and extravasation.⁷  Neutrophil rolling on the activated endothelium is initiated by engagement of selectins, including P-selectin, with P-selectin glycoprotein ligand 1 (PSGL-1). Selectin engagement further supports rolling and firm adhesion through conformational activation of integrin αLβ2 (LFA-1) and binding to intercellular adhesion molecule 1 (ICAM-1). Neutrophil recruitment to sites of inflammation is also mediated by chemokine signaling through an interaction between CXC chemokine ligand 1 (CXCL1) and its receptor, CXC chemokine receptor 2 (CXCR2). Finally, neutrophils extravasate to the site of inflammation along chemokine gradients.

Yago et al sought to elucidate the retinue of receptors and intracellular signaling pathways required for recruitment and activation of neutrophils in the IVC stenosis model of thrombosis. Similar to the recruitment of leukocytes to sites of inflammation, P-selectin–, CXCR2-, and β2-containing integrins were all required for recruitment of neutrophils to the activated endothelium (see figure). Importantly, the authors described a novel cooperative mechanism in the neutrophil in which PSGL-1 engagement, supported by CXCR2-mediated signaling, contributes to conformational activation of β2-containing integrins, such as αLβ2 (LFA-1), and firm adhesion of neutrophils. This is distinct from monocytes in which adhesion was dependent on P-selectin– and β2-containing integrins, but not CXCR2. Neutrophil P-selectin/PSGL-1–induced, integrin-dependent slow rolling required site-specific phosphorylation of the cytosolic adaptor protein Src homology domain–containing leukocyte phosphoprotein of 76 kDa (SLP-76). Interestingly, P-selectin and CXCL1 also cooperated to induce the formation of NETs, suggesting that regulatory pathways that mediate neutrophil adhesion also contribute to terminal effector functions. In contrast to extravasation of neutrophils to sites of inflammation, neutrophils do not appear to leave the vasculature during the early phase of immunothrombosis, likely due to the absence of a chemokine gradient. Critically, genetic deletion of components of this cooperative neutrophil-adhesion mechanism significantly impair thrombus formation in the IVC stenosis model.

What is the translational significance of the work? Venous thromboembolism (deep vein thrombosis and pulmonary embolism) is the third leading cause of cardiovascular death in the world. Preventing the recruitment of neutrophils or NET formation should reduce immunothrombosis without affecting hemostasis. There is a long history of targeting P-selectin as a novel antithrombotic strategy. P-selectin is expressed on the surface of activated endothelial cells and platelets. In 1992, it was shown that inhibition of P-selectin reduced leukocyte accumulation and fibrin deposition in an arteriovenous shunt model in baboons.⁸  This work was followed by a number of elegant studies by the Wakefield group showing that inhibition of P-selectin reduced venous thrombosis in animal models, including a baboon model.⁹  These studies indicate that targeting P-selectin reduces thrombosis and inflammation without increasing bleeding. The humanized monoclonal anti–P-selectin antibody crizanlizumab blocks PSGL-1 interaction with P-selectin and is currently being evaluated for the prevention of vaso-occlusive crises in patients with sickle cell disease.¹⁰  It would be very interesting to assess if crizanlizumab or other P-selectin inhibitors reduce venous thrombosis in humans.