Learn more: PMC Disclaimer | PMC Copyright Notice
At the Bench: Neutrophil extracellular traps (NETs) highlight novel aspects of innate immune system involvement in autoimmune diseases
Basic Research Review for Clinicians: Immunogenic and vasculopathic roles of NETs in the pathogenesis of SLE and other related autoimmune diseases.
Abstract
The putative role of neutrophils in host defense against pathogens is a well-recognized aspect of neutrophil function. The discovery of neutrophil extracellular traps has expanded the known range of neutrophil defense mechanisms and catalyzed a discipline of research focused upon ways in which neutrophils can shape the immunologic landscape of certain autoimmune diseases, including systemic lupus erythematosus. Enhanced neutrophil extracellular trap formation and impaired neutrophil extracellular trap clearance may contribute to immunogenicity in systemic lupus erythematosus and other autoimmune diseases by promoting the externalization of modified autoantigens, inducing synthesis of type I IFNs, stimulating the inflammasome, and activating both the classic and alternative pathways of the complement system. Vasculopathy is a central feature of many autoimmune diseases, and neutrophil extracellular traps may contribute directly to endothelial cell dysfunction, atherosclerotic plaque burden, and thrombosis. The elucidation of the subcellular events of neutrophil extracellular trap formation may generate novel, therapeutic strategies that target the innate immune system in autoimmune and vascular diseases.
Introduction
The objective of this review is to highlight the role of neutrophils in the pathogenesis of autoimmune diseases, with a particular focus on the potential immunogenic and vasculopathic roles that NETs may play in the development and maintenance of autoimmunity. Although increasing evidence links NETs to a wide array of autoimmune disorders and other conditions, this review will focus primarily on the potential role of NETs in the pathogenesis of SLE as an example of a prototypical autoimmune disease. Definitions of key terminology used throughout the manuscript and the relevance to NET formation are provided in Table 1.
TABLE 1.
Selected index of key terminology
| Term | Definition | Relevance to NETs |
|---|---|---|
| Cellular events | ||
| Apoptosis | Form of programmed cell death characterized, in part, by cell shrinkage, chromatin condensation, and DNA fragmentation | Distinct form of cell death different from NETosis |
| Necrosis | Form of premature cell death caused by external factors with resultant detrimental effects | Distinct form of cell death different from NETosis |
| NETosis | Form of programmed cell death characterized by release of decondensed chromatin bound to cytosolic contents | Cell death related to extrusion of NETs |
| Vital NETosis | Form of NETosis in which the neutrophil remains intact and functional after extrusion of nuclear contents | Potentially different form of NETosis that does not result in immediate cell death |
| Mechanisms of NETosis | ||
| ROS | Chemically reactive molecules containing oxygen | Plays important role in cell signaling during NETosis |
| NADPH oxidase | Membrane-bound enzyme complex activated during neutrophil respiratory burst | A critical enzyme complex in NETosis |
| DPI | Inhibitor of flavoenzymes, particularly NADPH oxidase | Abrogates NETosis in in vitro experiments |
| Citrullination/deimination | Conversion of positively charged, protein-bound arginine/methylarginine residues to the uncharged, nonstandard amino acid citrulline | Critical step in NETosis to promote histone decondensation |
| PADs | A family of enzymes that catalyze citrullination | PAD4 inhibitors abrogate NETosis in animal and in vitro models |
| Pathways of NETosis | ||
| Autophagy | Physiologic process whereby cell degrades unnecessary or dysfunctional components | Pathway implicated in NET formation |
| AKT | Protein kinase plays role in multiple cellular processes, such as apoptosis and proliferation | May direct neutrophil toward NETosis rather than apoptosis |
| PI3K | Family of enzymes involved in cellular functions, such as cell growth, differentiation, and survival | Involved in subcellular signaling in NETosis |
| RAF-MEK-ERK pathway | Chain of proteins in the cell involved in cell surface receptor signaling | Activation of pathway is required for NET formation |
| Relevant animal models | ||
| MRL/lpr mice | Murine model of spontaneous lupus as a result of FAS mutation with glomerulonephritis, skin disease, autoantibody production, and vascular disease | Enhanced rate of spontaneous NETosis/inhibition of NETs improves phenotype |
| NZM2328 mice | Murine model of female-predominant, spontaneous lupus with glomerulonephritis, autoantibody production, vascular disease, and strong dependence on IFNs | Enhanced rate of spontaneous NETosis/inhibition of NETs improves phenotype |
| ApoE−/− | Murine model of spontaneous atherosclerosis | Inhibition of NETosis can improve atherosclerosis |
| Other relevant biology | ||
| LDGs | A subset of proinflammatory neutrophils identified in SLE | Prone to spontaneous NETosis in absence of stimulation |
| TIIFNs | Group of proteins involved in innate immunity; overproduced in some patients with SLE | NET products can trigger production of TIIFNs |
| DNase1 | An endonuclease that cleaves DNA; can be dysfunctional in patients with SLE | Primary nuclease to degrade NETs |
| Inflammasome | A multiprotein structure within myeloid cells; a component of the innate immune system | NETs can activate the inflammasome to produce cytokines |
| Complement system | Part of the innate immune system that aids clearance of pathogens | NETs can activate the classic and alternative pathways |
| TF | Major initiator of physiologic coagulation and trigger of thrombosis | Expressed within NETs; contributes to hypercoagulability |
| Neutrophil proteins | ||
| LL-37 (CRAMP) | Antimicrobial peptide found in lysosomes of neutrophils | Common protein found within NETs; triggers TIIFNs |
| Elastase | A protease enzyme found within neutrophil granular proteins | Translocates to nucleus during NETosis |
| MPO | A peroxidase enzyme abundantly expressed in neutrophils with antimicrobial function | Translocates to nucleus during NETosis |
| MMPs | A group of enzymes capable of degrading extracellular matrix proteins | Mediates an interaction between NETs and vascular disease |
| Citrullinated peptides | A group of proteins, including vimentin, that includes major autoantigens in rheumatoid arthritis | Proteins are found citrullinated within NETs |
Neutrophils are the primary custodians of the innate immune system and play a critical role in host defense against microorganisms. Neutrophils respond to infectious insults through myriad defense mechanisms, including phagocytosis of microbes and toxic degranulation of cytoplasmic granular proteins that are microbicidal [1]. Impairments in neutrophil quantity or function render the host susceptible to a wide array of potentially life-threatening pathogens, highlighting the critical relevance of neutrophils in immune system homeostasis. Besides phagocytosis and degranulation, neutrophils stimulated by specific sterile and nonsterile stimuli can undergo a distinct form of cell death characterized by the extrusion of granular proteins bound to a meshwork of chromatin and other nuclear material [2]. These complexes of intermixed nuclear and cytoplasmic neutrophil contents released into the extracellular space have been termed NETs, and the process of NET formation as a novel form of cell death has been dubbed NETosis [3].
Despite the importance of neutrophils as agents of host defense, the current understanding of neutrophil function is superficial relative to the functional and phenotypical characterization of other immune cell populations. Part of these gaps in knowledge of neutrophil development and function reflect challenges inherent to studying these cells. Upon release from bone marrow, mature neutrophils are terminally differentiated cells that do not divide and cannot be genetically modified. In circulation, neutrophils are short-lived cells that do not readily survive freeze-thaw methods and die promptly in culture. As such, ideally, neutrophil functional studies must be carried out expeditiously on freshly isolated samples rather than samples collected and stored in existing bio-repositories. Murine neutrophil biology appears to differ critically from their human counterparts, somewhat limiting the use of mouse models of disease [4], and there are few established neutrophil cell lines [5]. Despite these limitations, the last decade has seen a revival of interest in neutrophil biology in the context of autoimmune disease.
SLE is an archetypal, idiopathic autoimmune syndrome characterized by profound clinical heterogeneity, the formation of autoantibodies to nucleic acids and associated nuclear proteins, multisystem organ involvement, unpredictable clinical course with periods of disease flare and remission, and potentially life-threatening end-organ complications. Pleiotropic abnormalities in both innate and adaptive immune responses characterize human and murine models of SLE.
A potential relationship between neutrophils and SLE was suggested >50 yr ago with the discovery of the lupus erythematosus cell phenomenon, defined as a neutrophil or macrophage that has phagocytized apoptotic nuclear material from another cell [6, 7]. Over the last decade, abnormalities in the phenotype and function of neutrophils, monocytes, macrophages, DCs, and other aspects of the innate immune system have been characterized in patients with SLE [8–11]. Accordingly, disease paradigms are shifting toward an increased awareness that profound abnormalities in both the innate and adaptive immune systems, triggered by genetic and environmental factors and a complex interplay between innate and adaptive immunity, likely shape disease susceptibility, pathogenesis, and outcomes in SLE.
NETs
In 2004, Brinkmann et al. [2] reported the identification of neutrophil-generated extracellular fibers, which they termed NETs. These fibers are decorated with domains containing proteins from primary, secondary, and tertiary granules. In the original descriptions, NETs were induced in activated neutrophils upon stimulation with IL-8, PMA, or LPS and were visualized trapping gram-positive and -negative bacteria and fungi by electron microscopy [2]. Time-lapse video microscopy illustrates the sequenced, morphologic events of NETosis. Minutes after activation, neutrophils undergo a series of changes, including delobulation of the nucleus, chromatin decondensation, detachment of the nuclear membrane, disintegration of granular membranes, and intermixing of the nuclear and cytoplasmic contents [2, 12]. Loss of intracellular membranes before the integrity of the plasma membrane is compromised is a defining feature of NETosis. Ejection of cellular contents through the ruptured cell membrane results in extrusion of NETs that occupy a space 10- to 15-fold bigger than the volume of cells from which they originate [13] (Fig. 1).
LDGs isolated from a patient with SLE undergo spontaneous NET formation. NETs are visualized by confocal microscopy as wispy strands of DNA decorated with neutrophil granular proteins (blue, Hoechst stain for DNA; red, MPO). Image courtesy of Carmelo Carmona-Rivera, Systemic Autoimmunity Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, U.S. National Institutes of Health.
Some of the initial, functional studies of NETs were performed in animal models of infection, where NETs are regarded to play a role in host defense via the immobilization of microorganisms and facilitation of pathogen elimination [2, 14, 15]. Many different pathogens are known to induce NETs, including gram-positive and -negative bacteria, fungi, and parasites [16]. Containment of pathogens to sites of initial infections may be an important function of NETs, and a variety of pathogens may have adapted to evade NET-mediated elimination [15]. Neutrophils from neonates are less capable of forming NETs in response to various microbial stimuli compared with neutrophils from adult subjects, which may underscore some of the enhanced susceptibility to infections observed in this age group [17]. The putative role of extracellular traps in host defense is exemplified by the conservation of these structures in various animals [18, 19], insects [20], and even plants [21].
Beyond a role in combating infectious disease, NETs have been detected in association with malignancy [22, 23], atherosclerosis [24], and autoimmune diseases. The presence of NETs has been demonstrated in synovial fluid from patients with gout and rheumatoid arthritis [25, 26], in psoriatic skin lesions [27, 28], in diabetic wounds [29], and in affected glomeruli from patients with SLE and ANCA-associated vasculitis [30, 31], indirectly suggesting that NETs play a role in immune-mediated disease.
MECHANISMS OF NET FORMATION
NETosis is considered a unique form of cell death that is morphologically distinct from apoptosis, necrosis, and other forms of cell death [27]. The mechanisms regulating NET formation and the subcellular events of NETosis have not been fully elucidated and are likely dependent on the type of triggering stimuli and the context of stimulation. There are a few cellular events proposed to be critical in NET formation, namely production of ROS, migration of neutrophil elastase and later, MPO to the nucleus, and histone modification and decondensation.
As neutrophil granular proteins are potentially harmful to the host, activation of neutrophils and release of their cellular contents are strongly regulated by production of ROS. Generation of superoxide by the NADPH oxidase enzyme complex at the phagosomal membrane, a process known as respiratory burst, is considered a critical, early step in NET formation. Patients with chronic granulomatous disease display genetically determined defects in subunits of the NADPH oxidase enzyme complex, have decreased capacity to form ROS, and have impairments in NET formation following in vitro stimulation with specific stimuli [12, 32]. Pretreatment of neutrophils with DPI, a potent inhibitor of NADPH oxidase, can abrogate NET formation in human and murine neutrophils [12, 33, 34]. However, in 1 study, lupus prone MRL/lpr mice, genetically modified to be deficient in NADPH oxidase, surprisingly developed markedly exacerbated lupus with increased renal disease and elevated and altered autoantibody profiles [35]. Additionally, loss-of-function mutations in NADPH oxidase have been associated with SLE in humans [36, 37]. These conflicting observations may reflect the fact that NADPH oxidase, although involved in NET formation, is involved in other neutrophil-related functions, such as phagocytosis, and also has anti-inflammatory roles. Therefore, pharmacologic inhibition of superoxide production may have unintended consequences and seems unlikely to be a viable therapeutic strategy in SLE. More recently, pathways of NETosis that are NADPH oxidase independent and mediated by mitochondrial ROS have been described [38]. Use of medications that target mitochondrial ROS production may perhaps constitute a potentially more selective therapeutic approach for diseases associated with enhanced NET formation.
Citrullination, also termed deimination, of histones is another important step in NET formation that occurs downstream of superoxide production. PADs are a family of 5 calcium-dependent enzymes that catalyze the process of citrullination of various target proteins [39]. PAD2 and -4 have been reported to be present in myeloid cells and can citrullinate histones [40, 41]. A PAD4-null strain of mice failed to produce NETs [42, 43]. As evidence of the potential therapeutic implications of targeting PADs, inhibition of these enzymes in 2 different lupus-prone mouse models (MRL/lpr and NZM2328) has been shown to reduce NET formation in vivo and to protect against lupus-related organ damage, including amelioration of rash, reduction of proteinuria and immune complex deposition in the kidneys, improvement of endothelial function, and alteration of circulating autoantibody profiles and complement levels [44, 45].
The Raf-MEK-ERK pathway, which has been implicated in NET formation upstream of NADPH oxidase, modulates expression of antiapoptotic proteins, suggesting that neutrophils might block apoptosis to allow NETosis [34]. Activation of protein kinase B, also known as Akt, is essential for NADPH oxidase-mediated NETosis, and Akt-specific inhibitors can direct neutrophils away from NET formation and toward apoptosis [46]. Therefore, Akt may serve as a molecular switch that regulates between an axis of NETosis and apoptosis. The autophagy pathway has also been implicated in NET formation. When neutrophils are stimulated by PMA, they develop large vacuoles that resemble autophagosomes [12]. Wortmannin, which inhibits PI3Ks and related enzymes, can block vacuolization and intracellular chromatin decondensation, thereby preventing NET formation and promoting neutrophil apoptosis [47]. Signal inhibitory receptor on leukocyte 1, a surface molecule exclusive to phagocytes, may represent a therapeutic target to inhibit NET formation by abrogating MEK-ERK pathway signaling and suppressing the formation of ROS [48]. The Src/Syk signaling pathway has been implicated in NET formation induced by β-glucan, 1 of the surface components of yeasts [49]. Whether Src/Syk signaling plays a role in NET formation in the context of autoimmune diseases is unknown. Mammalian target of rapamycin can regulate NET formation by post-transcriptional control of expression of hypoxia-inducible factor 1a, suggesting additional, potential points of molecular intervention in strategies to modify NET formation in disease [50].
Interestingly, NET formation may not always be synonymous with cell death [51]. A few groups have raised the idea that NETosis may occur in the absence of cell death [52–54]. Delivery of the NET without neutrophil membrane lysis may result in an enucleated neutrophil that retains some functions of host defense [55]. This process, termed “vital NETosis,” has been described in the context of neutrophil defense against gram-positive infections and warrants further study.
DEGRADATION OF NETs
Both enhanced NET formation and impairments in degradation of NETs may contribute to neutrophil-mediated immunogenicity. Defects in the clearance of apoptotic material have been demonstrated in human SLE [56]. In a parallel observation, clearance of NETs is also seemingly altered in patients with SLE. DNase1, a serum endonuclease, is essential to degrade chromatin within NETs [57]. Rare mutations in DNase1 have been reported in association with SLE [58], and mice that lack DNase1 develop a lupus-like phenotype [59]. In vitro studies identify impairments in NET degradation in serum from a subset of patients with SLE (30–40%), and these impairments can vary over time within individual patients in association with disease activity [57, 60]. Among patients with SLE, impaired NET degradation has been associated with lupus nephritis, hypocomplementemia, production of TIIFNs, and elevated autoantibody levels [60]. Potential mechanisms for impaired DNase1 activity in SLE include the presence of DNase1 inhibitors [61] and the ability of autoantibodies to NET components to stabilize NETs from degradation [57]. rDNase has been considered a potential therapy for SLE [62]; however, a phase I study in human patients demonstrated safety but did not demonstrate improvement in serum markers of disease activity [63]. More recently, contributions of macrophages to NET clearance have been described. Macrophages actively engulf NETs opsonized by C1q, an event that is facilitated by the preprocessing of NETs by DNase1 [64]. In addition to SLE, impairments in NET degradation have been reported in association with ANCA-associated vasculitis [65], antiphospholipid antibody syndrome [66], and drug-induced vasculitis [67]. In each of these conditions, defects in clearance of NETs may allow NETs to persist longer in vivo and thereby, facilitate NET-mediated immunogenic and pathogenic effects.
LDGs AND OTHER CELL SUBSETS THAT FORM EXTRACELLULAR TRAPS
Unlike other leukocytes, subsets of neutrophils with differing function are only recently starting to be identified. One of these recently characterized neutrophil subsets has been termed LDGs, based on their identification within the PBMC fraction in density gradient preparations [68]. LDGs are present in peripheral blood from patients with SLE in association with disease activity and are characterized in vitro as proinflammatory neutrophils that are toxic to endothelial cells and display enhanced ability to synthesize TIIFN and type II IFNs and other proinflammatory cytokines [69]. Compared with normal density neutrophils, LDGs have a strikingly enhanced capacity to form NETs in the absence of any triggering stimulus in ex vivo studies [31]. These cells have also been described in ANCA-associated vasculitis, a condition that shares many phenotypic features with SLE, where LDGs correlate with disease activity and may be a useful biomarker to predict response to therapy [70]. Although LDGs have a different gene-expression profile from autologous normal density neutrophils [31], distinct cell surface markers or epigenetic markings to readily identify and differentiate these cells remain elusive [71].
Neutrophils, eosinophils, mast cells, and basophils all contain cytoplasmic granular proteins, and each of these cell types has the capacity to form extracellular traps [27, 72, 73]. Expulsion of mitochondrial, rather than nuclear DNA, has been reported as a novel mechanism of extracellular trap formation in eosinophils and neutrophils [53, 73]. Associations between extracellular traps formed from eosinophils, mast cells, or basophils and autoimmune diseases have not been firmly established.
IMMUNOGENIC ROLE OF NETs
Source of autoantigens
Despite the fact that neutrophils have various types of granules containing hundreds of proteins, only a limited number of proteins (20–30) have been detected within NETs [26, 74]. Many of the proteins found within NETs represent the major autoantigenic targets in rheumatologic diseases: SLE (dsDNA, histones), ANCA-associated vasculitis (MPO, PR3), and rheumatoid arthritis (vimentin, enolase, histones). Several groups have proposed a “vicious loop” theory, whereby modified proteins found within NETs contribute as antigenic targets for autoantibody and immune complex formation that can, in turn, induce further NETosis. For example, MPO and PR3, which are the major antigenic targets of ANCA, have been demonstrated within NETs isolated from patients with ANCA-associated vasculitis, and ANCA antibodies can further induce NET formation in these patients [30]. Novel autoantibodies to NET-related proteins have been described in association with various autoimmune diseases. LL-37 and MMP-9 have been demonstrated within NETs, and autoantibodies to these proteins have been identified in patients with SLE and characterized as immunogenic and vasculopathic [75, 76]. Protein cargo within NETs may be influenced directly by the nature of the inciting stimuli. In rheumatoid arthritis, disease-associated autoantibodies induced a wider array of NET-protein content compared with NET induction with proinflammatory cytokines [26]. Ultimately, detailed analysis of NET contents across different diseases may reveal differences in protein abundance and post-translational modifications that could help to explain the differing autoantibody profiles associated with various autoimmune diseases.
Interaction between NET contents and APCs in the context of autoantibody production merits investigation. In coculture experiments, myeloid DCs more avidly uploaded components of netting neutrophils compared with contents from apoptotic or necrotic neutrophils [77]. Transfer of neutrophil granular proteins was observed by confocal microscopy and was inhibited by pretreatment with DNase. Immunization with NET-loaded DCs induced ANCA formation and vasculitis in murine models genetically predisposed to autoimmunity, whereas DCs loaded with apoptotic neutrophil debris did not induce pathology. A subset of neutrophils identified around the marginal zone of the spleen is prone to spontaneous NET formation and can activate B cells in a T cell-independent manner to induce Ig class-switching and antibody production [78]. Upon direct contact with T cells, NETs can reduce activation thresholds and increase T cell responses to specific antigens [79]. These findings support the concept that the complex of neutrophil cytoplasmic proteins and nucleic acids may play an important role in NET-mediated immunogenicity.
NETs and TIIFNs
Overproduction of TIIFNs or increased responses to these cytokines are a feature of SLE, and this phenomenon is considered critical for SLE pathogenesis [80]. Three groups have independently shown in SLE that NETs can activate pDCs to synthesize IFN-α in a TLR-dependent manner [31, 76, 81]. Immune complexes found in the sera of patients with SLE can stabilize DNA from extracellular degradation. These complexes, comprised of self-DNA bound to NET-related peptides, such as LL-37 or HMGB1, facilitate access of DNA to the intracellular compartments of pDCs with subsequent enhancement of IFN-α production. Autoantibodies to NET-derived self-DNA-peptide complexes can further promote transport of these complexes into pDCs and subsequently amplify production of IFN-α [76]. A similar process of NET-mediated TIIFN production has also been reported in psoriasis [28, 82] and Type 1 diabetes [83]. As an additional mechanism for NET-mediated TIIFN production, oxidized DNA present in NETs can activate the intracellular stimulator of IFN genes pathway [84]. These observations highlight a connection between neutrophils and the innate activation of pDCs and link 2 fundamental components of the pathogenesis of SLE and other autoimmune diseases.
Other immunogenic mechanisms
Activation of the inflammasome machinery by NETs may represent another important amplifier of inflammatory pathways. Both NETs and LL-37 (a protein contained in NETs) can activate caspase-1, the central enzyme of the inflammasome, in human and murine macrophages [85]. Activation of the NLRP3 inflammasome by NETs results in the synthesis and release of active IL-1β and IL-18, and these effects are enhanced in macrophages derived from patients with SLE. This process appears to be mediated, at least in part, by LL-37 binding to P2X7R [85]. A feed-forward loop has been proposed, whereby NETs stimulate the inflammasome, triggering increased IL-18 and IL-1β production, which in turn, induces further NET formation.
NETs can activate the classic and alternative pathways of the complement system. Activation and consumption of components within the classic complement pathway are a hallmark of SLE. C1q can bind to NETs and activate complement with resultant deposition of C3b on NETs [61]. Thus, nondegraded NETs, particularly in the subset of patients with SLE who demonstrated impaired clearance of NETs, may play a role in complement consumption in SLE. Activation of the complement system via the alternative pathway rather than the classic pathway has been implicated in the pathogenesis of ANCA-associated vasculitis [86]. Serum from patients with ANCA-positive vasculitis can induce production of NETs that contain components that can activate the alternative complement pathway, including factor B and properdin [87].
NET formation in inflammatory diseases may not always have a deleterious effect. Gout is a painful crystal arthropathy caused by an acute inflammatory reaction and accumulation of neutrophils in response to MSU crystals deposited within synovial tissue. MSU crystals can induce NET formation and activate the inflammasome with resultant production of IL-1β [88, 89]. Interestingly, aggregation of NETs in gouty joints may serve an anti-inflammatory function by degrading cytokines and chemokines and by disruption of neutrophil recruitment and activation [89].
VASCULOPATHIC ROLE OF NETs
Toxicity to endothelial cells
Endothelial dysfunction and damage are a fairly shared finding across different systemic autoimmune diseases. Endothelial dysfunction can precede accelerated atherosclerosis in SLE and rheumatoid arthritis, and overt endothelial damage is a hallmark of systemic vasculitis [90]. Compared with normal-density neutrophils from healthy controls and patients with SLE, LDGs isolated from patients with SLE have a heightened capacity to induce endothelial cell cytotoxicity, and this toxic effect on the endothelium is abrogated by disrupting NETs with a nuclease [31]. Direct incubation of endothelial cell lines with NETs isolated from SLE LDGs promotes endothelial cell death. The mechanism by which enhanced NET formation by LDGs contributes to endothelial damage is, at least in part, mediated by MMPs [75]. MMP-9, externalized in NETs, activates MMP-2 in endothelial cells and induces a cell death program. Neutralization of MMP-9 prevents MMP-2 activation, decreases NET-mediated cytotoxicity, and restores endothelium-dependent vasorelaxation. The vasculopathic effect of NETs is supported by in vivo models. In murine lupus, use of chemical PAD inhibitors that abrogate NET formation improves endothelium-dependent vasorelaxation and endothelial progenitor cell phenotype and function [44, 45]. Whether biomarkers of NETosis correlate with measures of endothelial dysfunction in humans remains to be shown.
Atherosclerosis
Accelerated atherosclerosis, unexplained by traditional risk factors, is a feature of many rheumatologic conditions, and cardiovascular disease is a leading cause of mortality among patients with SLE and rheumatoid arthritis [91]. Several lines of evidence suggest a proatherogenic role of neutrophils and NETs. Inflammatory cells, including neutrophils and NETs, infiltrate human and murine atherosclerotic plaques. Citrullinated peptides and PAD4 are present in human atherosclerotic plaques [92]. Murine knockouts of CRAMP (the murine ortholog of LL-37) in the ApoE−/− mouse model of atherosclerosis protect these mice against atherosclerosis [93]. Likewise, antibody neutralization of HMGB1 or depletion of S100A8/S100A9, 2 neutrophil-related molecules, reduces atherosclerotic burden in ApoE−/−mice [94, 95]. Increased levels of IFN-α and autoantibodies to CRAMP/DNA complexes have been reported in murine models of atherosclerosis, and CRAMP/DNA complexes can stimulate IFN-α production in pDCs within murine atherosclerotic plaques [24], promoting proinflammatory responses in the vasculature. In the ApoE−/−mice, administration of chemical PAD inhibitors can reduce arterial-wall NET formation and IFN-α expression and significantly reduce arterial lesions, lipid-rich plaque, and prothrombotic phenotype in these mice in a TIIFN- and neutrophil-dependent manner [96]. These results support a role for NET formation in the pathogenesis of atherosclerosis through modulation of innate immune responses.
An additional proatherogenic mechanism of neutrophils was described recently in SLE [97]. Oxidation of HDL is considered a risk factor for cardiovascular disease through impairment in the function of this lipoprotein and altering macrophage cholesterol efflux. Levels of oxidized HDL are substantially higher in patients with SLE and in lupus-prone mice compared with healthy controls. Neutrophils possess the machinery to oxidize HDL via NADPH oxidase, NOS, and MPO. These enzymes are externalized within NETs and strongly promote HDL oxidation. In vivo blockade of NET formation in lupus-prone mice by PAD inhibition significantly decreased HDL oxidation. As such, NET-mediated lipoprotein modification may be an additional mechanism to promote accelerated atherosclerosis in SLE.
Thrombosis
Many autoimmune diseases, characterized by dysregulated NET formation, including SLE and ANCA-associated vasculitis, have an associated, increased risk for thromboembolic events [98, 99]. The role of neutrophils as mediators of thrombosis is historically a controversial issue; however, accumulating evidence linking NETs with clot formation and propagation supports the involvement of neutrophils in thrombosis and suggests potential mechanisms for neutrophil-driven thrombogenicity [100]. NETs have been detected within thrombus isolated from patients with ANCA-associated vasculitis [101]. TF, a protein present in endothelial tissue and leukocytes, is the major initiator of coagulation and thrombin formation [102]. Whereas it has been controversial whether neutrophils synthesize functional TF or acquire TF produced by neighboring macrophages [103–105], the presence of TF has been demonstrated consistently within NETs in many different clinical settings. NETs, isolated from serum of patients with active small vessel vasculitis or patients with sepsis, contain TF [106, 107], as do NETs found within thrombi from coronary arteries during acute myocardial infarction [108]. TF within activated neutrophils is engulfed in autophagosomes and translocated to NETs, implicating autophagy-dependent, extracellular delivery of TF [107]. NETs containing TF may provide a scaffold for fibrin deposition and platelet entrapment and activation, and in a feed-forward loop, activated platelets may trigger further NET formation [109]. Additionally, serine proteases derived from NETs can activate the coagulation cascade by degrading TF pathway inhibitor [110] and promoting the activation of factor XII [111].
Animal models suggest the potential for neutrophil-targeted therapeutics to ameliorate thrombotic disease. Neutrophil depletion attenuates thrombus formation in an animal model of DVT [111]. In 2 murine models of lupus and a murine model of atherosclerosis, inhibition of PAD enzymes blocked NET formation and delayed time to carotid artery thrombosis induced by photochemical injury [44, 45, 96]. In addition, DNase can delay thrombus development in murine models of lupus and prevent extension of thrombus in a murine model of DVT [45, 111]. Therefore, the targeting of NET formation could have therapeutic efficacy to prevent the initiation and propagation of arterial or venous thrombosis [112].
POTENTIAL THERAPEUTICS
Medications that target key subcellular events in NET formation or that enhance NET clearance may provide much-needed novel therapeutic strategies in SLE and related autoimmune diseases. Potential therapeutic targets include inhibitors of NADPH oxidase, mitochondrial ROS production, actin cytoskeleton, and PAD enzymes. Whereas the safety of targeting NETosis as a therapeutic strategy will need to be evaluated rigorously, initial evidence from animal models and humans suggests that the targeting of PAD enzymes and other neutrophil-related targets may be effective to prevent NET formation without incurring substantial risks for treatment-related infectious sequelae [44, 45, 96, 113–116].
Additionally, medications with proven therapeutic efficacy in SLE should be investigated in the context of their potential effects on NET formation. Physiologically relevant concentrations of antimalarials, a mainstay of therapy for patients with SLE, can inhibit NET formation in vitro [97]. Cyclosporine A, a medication that is efficacious, albeit infrequently used to treat SLE, has also been shown to suppress NETosis via antagonism of the calcineurin pathway [117].
CONCLUDING REMARKS
Research focused on neutrophils, NETs, and how the innate immune system contributes to autoimmune and vascular diseases has generated novel insights into disease pathogenesis in SLE and other autoimmune conditions (Fig. 2). Given the crucial role that neutrophils play in host defense and the potentially devastating clinical consequences of neutrophil depletion, development of therapeutics that target neutrophils in autoimmune diseases may seem ill advised. However, an improved understanding in the molecular checkpoints that guide different types of neutrophil programmed cell death and elucidation of the critical mechanisms that govern pathways of NETosis may usher in a wave of novel therapeutics that target the innate immune system in autoimmune diseases. Identification of pathogenic subsets of neutrophils may facilitate the development of therapeutics that safely target specific components of the innate immune system, while persevering critical aspects of host defense (Table 2). Ongoing research conducted at the bench to fill in gaps in understanding about neutrophil development and function and well-designed clinical trials will be essential toward translating these findings effectively and safely into clinical practice.
Neutrophils (PMNs) release NETs with potential immunogenic roles. Protein-DNA complexes within NETs (e.g., LL-37/DNA) activate pDCs to synthesize TIIFNs. NETs interact with B and T lymphocytes, linking the innate and adaptive immune systems and promoting their activation. NETs can activate platelets and coagulation factors and promote thrombosis. NET contents are potentially toxic to endothelial cells (EC), promoting vasculopathy and apoptosis through MMPs and histones. Several amplification loops of NETosis exist, whereby products found within NETs (autoantigens, complement activating factors, proinflammatory cytokines) or antibodies (Abs) and immune complexes (ICs) to/containing NET-related proteins stimulate neutrophils to undergo NET formation. In a feed-forward loop, NET proteins stimulate the inflammasome (NLRP3) in macrophages (Mϕ) with resultant release of proinflammatory cytokines (IL-1 and IL-18) that promote additional NET formation in granulocytes.
TABLE 2.
Summary of questions for future clinical-translational research
| • What is the extent of human diseases in which LDGs are present? Is there a way to readily distinguish and selectively target LDGs? Does quantity of LDGs serve as a biomarker to predict clinical outcomes in SLE and ANCA-associated vasculitis? |
| • Do extracellular traps originating from immune cell populations other than neutrophils play a role in human disease? |
| • Does protein content within NETs differ across autoimmune diseases in association with distinct autoantibody profiles? Are novel autoantibodies targeting NET contents differentially expressed in autoimmune conditions, and what is the potential functional significance of such antibodies? |
| • Are amplification of TIIFN responses by NETs unique to SLE or similarly found in other related autoimmune diseases? |
| • In addition to potential deleterious effects, do NETs play a beneficial or anti-inflammatory role in certain autoimmune diseases? Does enhanced NET formation confer a protective effect against pathogens? |
| • Can a standardized biomarker for in vivo NET formation be developed and used to predict cardiovascular events in autoimmune and vasculopathic diseases and target preventative approaches to minimize future cardiovascular events? |
| • To what extent do markers of NETosis explain cardiovascular risk factors in autoimmune diseases independent of traditional risk factors? |
| • Would inhibition of NET formation be a useful therapeutic strategy to prevent initiation and propagation of arterial or venous thrombosis? If so, would inhibition of NETs have broad efficacy in all types or specific types of thromboembolic disease? |
| • Would inhibition of specific critical subcellular events of NETosis be efficacious to treat SLE or other related autoimmune diseases, and what would be the potential infectious and other risks of such approaches? What is the role of current effective therapies in autoimmunity with regard to modulation of NETosis? |
AUTHORSHIP
P.C.G. and M.J.K. provided the study conception and design; acquisition, analysis, and interpretation of data; and drafting and review of the manuscript.
ACKNOWLEDGMENTS
This research was supported by the Intramural Research Program at the National Institute of Arthritis and Musculoskeletal and Skin Diseases, U.S. National Institutes of Health.
Glossary
| AKT | protein kinase B |
| ANCA | antineutrophil cytoplasmic antibody |
| ApoE−/− | apolipoprotein E-deficient |
| CRAMP | cathelicidin-related antimicrobial peptide |
| DC | dendritic cell |
| DPI | diphenyleneiodonium |
| DVT | deep vein thrombosis |
| HMGB1 | high-mobility group box 1 |
| LDG | low-density granulocyte |
| LL-37 | cathelicidin |
| MMP | matrix metalloproteinase |
| MPO | myeloperoxidase |
| MSU | monosodium urate |
| NET | neutrophil extracellular trap |
| NETosis | process of neutrophil extracellular trap formation |
| NLRP3 | nucleotide-binding oligomerization domain-like receptor family, pyrin domain containing 3 |
| PAD | peptidylarginine deiminase |
| pDC | plasmacytoid dendritic cell |
| PR3 | proteinase 3 |
| ROS | reactive oxygen species |
| SLE | systemic lupus erythematosus |
| TF | tissue factor |
| TIIFN | type I IFN |
Footnotes
SEE CORRESPONDING ARTICLE ON PAGE 265
DISCLOSURES
The authors declare no conflicts of interest.