Compromised tolerance to self-antigens is an early step in the development of systemic lupus erythematosus (SLE) (1, 2). Among the diverse mechanisms that can mediate this process (1), NETosis is believed to play a major role. NETosis was first reported as a specialized form of cell death that occurs in neutrophils with neutrophil extracellular traps (NETs) released. This form of NETosis is now termed suicidal NETosis (3). Recently, it has been shown that NET formation can occur with preservation of neutrophilic functions, including phagocytosis and chemotaxis. This phenomenon is termed vital NETosis. Suicidal NETosis has been extensively evaluated and is considered as the classical NETosis, while research on vital NETosis is just unraveling (4). Stimuli and environment appear to be the most important determinants for neutrophils to undergo the suicidal NETosis or vital NETosis pathway (5). In this article, we evaluate the existing evidence on the role of NETosis in SLE ( Supplementary Figure S1 ). We will mainly focus on suicidal NETosis, as little is known on the role of vital NETosis in SLE.

During the process of NETosis, NETs are released. NETs are large, extracellular, and fibrillary structures, which are composed of cytosolic and granule proteins that are assembled on a scaffold of decondensed chromatin (6). NETs, which can neutralize and kill bacteria, fungi, viruses and parasites, are the important first-line in host immune defense (7). However, if dysregulated, NETosis can contribute to the breakdown of self-tolerance and consequently lead to autoimmunity. In this article, we review pathways of NETosis and the role of NETs as a contributor to the loss of normal immune tolerance in lupus. We present evidence that supports excessive NET formation and reduced NET degradation in lupus and discuss multiple mechanisms whereby NET may contribute to lupus pathogenesis. Furthermore, we discuss clinical implications of NETosis and potential target of therapy by modulating NETosis in lupus.

Stimuli, such as infections, drugs, ultraviolet light, and hormones, trigger neutrophil activation through innate immune receptors, which activates downstream intracellular mediators that include protein kinase C/Raf-MEK-ERK, and therefore lead to calcium influx and reactive oxygen species (ROS) production (8, 9). ROS, produced by nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) oxidase or mitochondria, activates myeloperoxidase (MPO), neutrophil elastase and protein-arginine deaminase type 4 (PAD4) to promote chromatin decondensation. Peptidylarginine deiminase 4 (PAD4)-dependent citrullination of histones induces decondensation of DNA resulting in a mixture of DNA and bactericidal proteins, including myeloperoxidase (MPO) and neutrophil elastase (NE), which are contained originally in intracytoplasmic granules. The increased cytosolic Ca⁺⁺ can also act as a cofactor for PAD4, which is a nuclear enzyme that promotes the citrullination of histone, to facilitates the interaction with DNA (10), and lead to NETs formation. Vital NETosis is also a complex of DNA, MPO and NE. However, unlike suicidal NETosis, vital NETosis is ROS independent (5). Notably, different stimuli utilize different pathways to induce NET formation, as shown in Figures 1 , 2 for pathways stimulated by phorbol 12-myristate 13-acetate (PMA) and calcium ionophore A23187, respectively (8). Patients with X-linked chronic granulomatous disease (CGD), who carry inactivating mutations of NADPH oxidase (NOX2), exhibit a specific pathway of NET formation, which is independent of NADPH oxidase (11) ( Figure 3 ).

The function for NETs was first discovered in the context of infection and later in autoimmunology. During infection, DNA in NETs presents a rapid bactericidal activity by sequestering surface bound cations, disrupting membrane integrity and lysing bacterial cells (6, 12). Via post-apoptotic NETs, neutrophils around the marginal zone (MZ) of the spleen, acting as B cell helpers, generate an innate layer of antimicrobial immunoglobulin defense (13). In autoimmune disease, such as SLE, DNA covariates with large amounts of neutrophil proteins including LL37 (a cathelicidin antimicrobial peptide) and high-mobility group box 1 (HMGB1), to activate plasmacytoid DCs (pDCs). The SLE NETs activated pDCs produce high levels of interferon (IFN)-α in a DNA- and TLR9 (Toll-like receptor 9)–dependent manner. IFN-α, in turn, activates neutrophils, resulting in more NETs formation through a positive feedback cycle (14). SLE patients were found to develop autoantibodies to both the self-DNA and antimicrobial peptides in NETs, indicating that NETs could also serve as autoantigens to trigger B cell activation (15).

Studies have shown that LL37 and HMGB1, which are released during NETs formation, are autoantigens and are playing important roles in immunity and inflammation (16). They act as autoantigens, in combination with DNA. LL37-DNA complexes derived from NETs can directly trigger polyclonal B cell activation via TLR9, which can lead to increased antibody (Ab) production (17). HMGB1–DNA complexes may also be recognized by autoreactive B cells through B cell receptor–TLR7/9 interaction, resulting in autoAb production (18). Preclinical mouse models demonstrated that in vitro monocyte derived DCs take up DNA particles from neutrophils undergoing NETosis. Transfer of these DNA-loaded monocyte-derived DCs led to production of antibodies against dsDNA, MPO, and proteinase-3 (PR3) in mice. Autoantibody production was most significant when mice were injected with DNA-loaded monocyte-derived DCs that were exposed to NET-ting neutrophils. Thus, the extruded DNA from NET can also be more immunogenic than the whole apoptotic material (19, 20).

Besides the immunogenic effects, NET can also have a direct cytotoxic effect on human epithelial and endothelial cells via the externalization of histones. Incubation of epithelial and endothelial cells with histone type-IIA (which includes all types of histones) prevented cell growth, and provoked cytotoxicity in a concentration-dependent manner. These data confirm the cytotoxic capability of histones in NETs on epithelial and endothelial cells (21). Extracellular histones originating from NETs have been shown to promote capillary necrosis and podocyte loss, which can lead to proteinuria and crescent formation in severe glomerulonephritis (22). Since endothelial cells have a limited capacity to internalize NETs, the persistent extracellular NETs can induce vascular leakage through the degradation of intercellular junction protein VE-cadherin. Taken together, NETs have specialized immune-protective functions, but can also elicit autoimmunity and direct cytotoxic effects on renal cells. The interplay between neutrophils and other cell types is presented in Figure 4 .

Although ample evidence suggests a pathogenic role of NETosis in lupus, a few animal studies suggest a no role or even a protective role of NETosis in lupus. The activation of the NADPH oxidase (Nox2) complex is required for NETs formation (52). Consequently, MRL/lpr mice that were rendered deficient in Nox2 had impaired NETs formation. Yet, Nox2-deficient MRL/lpr mice, which could not form NETs, had markedly exacerbated lupus disease. This suggests that NETosis may not contribute to SLE in vivo, and rather that Nox2 acts to inhibit disease pathogenesis in a way that is majorly NETs independent (53).

In addition, genetic deletion of PAD4, a distal mediator of NETs formation, also did not ameliorate loss of immune tolerance, immune activation or nephritis in MRL/lpr mice (54). In further support of this result, pharmacological inhibition of PAD did not improve disease in the Fas/lpr model of lupus, anti-glomerular basement membrane antibody-induced model of proliferative nephritis, and human-serum-transfer model of SLE (54). Compared to previous studies showing the effect of pan-PAD chemical inhibitors in ameliorating lupus in MRL/lpr mice (23) and of PAD4 and PAD2 deficiency in preventing lupus in FVB mice (24), PAD4 deficiency in MRL/lpr mice and pan-PAD chemical inhibitors in Fas/lpr did not improve lupus (54). The reasons for these discrepancies are not yet clear and may be related to the use of different mouse strains, the stage(s) of disease when the NET manipulations were performed, or the impact of partial inhibition versus complete inhibition of PAD activity. When it comes to the upstream and downstream mediators of NETs formation, both NOX2-deficient strain and PAD4-deficient strain exhibited impaired induction of NET formation, yet had elevated levels of antinuclear autoantibodies (ANAs) and exacerbated glomerulonephritis in the pristane-induced model of lupus. Corollary to this, treatment with Nox2-specific activators induced NETs formation, yet ameliorated pristane-induced lupus (55). Thus, while most studies thus far support the widely accepted notion that NETosis contributes to the pathogenesis of SLE, it is possible that NETs may have no role or even paradoxical roles in some model systems, in some patients, and in certain stages of disease.

In this section, we will review potential links between current treatments for lupus and NETosis, and discuss possible ways to therapeutically target NETosis. The latter include targeting of molecules such as NADPH oxidase, MPO, and PAD4 which are involved in NETosis pathway to prevent or reduce NET formation, accelerating the degradation of NET, deactivating molecules that are downstream of NET, and secondarily inhibiting NETosis ( Figure 5 ).

Hydroxychloroquine and corticosteroids, which are cornerstone of drug therapy in SLE, have been shown to decrease NET formation in vitro (20). Since autoAb and immune complexes can trigger NET formation, treatments that reduce autoAb could reduce NET formation (20). For example, treatment with a combination of rituximab and belimumab reduced autoAb levels and NET formation in patients with SLE (50).

Several approaches are being proposed to directly reduce the formation of NET and/or enhance their degradation (20). These approaches include targeting ROS with diphenyleneiodonium (56), targeting mitochondrial ROS with N-acetylcysteinine (57), inhibiting PAD enzymes by Cl-amidine (23), enhancing breakdown of NET with DNase1 (58), using DNASE1L3 to increase DNA digestion in apoptotic microparticles (27), and targeting histones with anti-citrullinated protein Ab (59). Some of these approaches are being tested in lupus in preclinical animal model and human studies. For example, lupus-prone MRL/lpr and NZM mice treated with PAD inhibitors, namely Cl-amidine and BB-Cl-amidine, had reduced NET formation and were protected from lupus (23). NADPH oxidase inhibitors can also suppress NET production. For example, bacterium Lactobacillus fermentum CECT5716 (LC40) protected kidneys in a mouse model of lupus by inhibiting NADPH oxidase activity (60). In a recent study, a Syk inhibitor fostamatinib attenuated NETs in neutrophils and reduced lupus characteristics (serum creatinine, proteinuria, and anti-dsDNA) in Fcgr2b-/- lupus mice (25).

Studies have shown that SLE NETs decorated with downstream molecules, including tissue factor and interleukin-17A (IL-17A), promoted thrombin generation and the fibrotic potential of cultured skin fibroblasts (61). In mouse model of lupus, mice lacking IL-17 were protected from the development of glomerulonephritis and had improved survival (62). This suggests that inhibition of IL-17A or tissue factor expressed on NET could be another potential therapeutic target.

In human SLE neutrophils, NETosis could be inhibited by adding a therapeutic anti-citrullinated protein Ab; this Ab also prevented NET-mediated organ damage in animal models of inflammation (59). JAK inhibitor tofacitinib, a drug approved to treat inflammatory diseases, modulated NET formation and ameliorated lupus in MRL/lpr mice (44), and is under clinical investigation in SLE patients (NCT02535689). Using a peptide inhibitor of complement C1, PA-dPEG24, to reduce NET formation by human neutrophils (63) offers another potential NETosis-based therapeutic approach.

In a recent study, 6-gingerol, the most abundant bioactive compound of ginger root, attenuated NET release in response to lupus- and APS-relevant stimuli, such as RNP ICs and aPL (APS IgG), in human neutrophils in vitro. Administration of 6-gingerol to mice reduces NET release in various models of lupus and APS, while also improving disease-relevant endpoints, such as autoantibody (anti-dsDNA and anti-β2GPI) formation and large-vein thrombosis (64).

A recent report identified inositol-requiring enzyme 1 α (IRE1α) as a critical mediator of lupus-derived immune complex–mediated NETosis in vitro (65). Importantly, pharmacological inhibition of IRE1α using KIRA6 reduced mitochondrial ROS formation, NET release (in both human neutrophils and a mouse model of lupus), and autoantibody formation in multiple lupus mouse models. Thus, inhibition of the IRE1α pathway could be an effective strategy for neutralizing NETosis in lupus.

Quantifying NET formation through the course of lupus disease may serve as a biomarker for disease activity, as suggested by studies showing increased circulating NET in patients with active SLE and reduced NET in SLE patients in remission (39, 50). However, currently there is no gold standard to measure NET formation (4, 66). The current strategies to detect NETs including microscopy, ELISA, immunoblotting, flow cytometry, and image-stream-based methods suffer from drawbacks such as being subjective, error-prone, non-specific and not being high throughput. This can be alleviated by new imaging based high throughput methods. For example, a novel imaging flow cytometry approach for the measurement of NETs both in vitro and in whole blood samples can provide an unbiased, accurate and rapid quantification of NET formation (67). Another study employed membrane-permeable and impermeable DNA dyes in situ to stain NET-forming cells, and used automated algorithm-driven single cell analysis of change in nuclear morphology, nuclear area and intensities to precisely detect NET-forming cells (68). Such high throughput approaches may provide a good platform to evaluate NETs as a surrogate marker of disease activity in the clinic and for the discovery of potential inhibitors of NET formation.

Taken together, observations in humans and animals with SLE support the idea that excessive NETosis and reduced NET degradation play a role in autoAb production, inflammation, and tissue damage in lupus ( Table 1 ). Impaired NETosis is believed to mark an early step in lupus pathogenesis. The extruded nuclear Ag released by NET serve as autoAg, and the failure to dismantle NET plays a role in the breakdown of normal immune tolerance. The persistent exposure of nuclear particles then activates immune cells, which facilitate immune response against self-Ag. Excessive NET can also induce Endo-MT in cultured endothelial cells, which contributes to activated myofibroblasts and extracellular matrix production. Thus, NETosis may play different pathogenic roles at different stages of lupus. However, a few animal studies suggest that NETosis may have no role at all or even a protective role in the development of lupus in certain model systems. These potentially dual, pathogenic versus protective roles, of NETs will need to be precisely delineated at different stages of lupus disease before NETosis-targeting treatments can be used in the clinic. Nonetheless, preclinical discoveries on mechanisms of NET formation, together with clinical findings in lupus, will pave the way for further investigations into targeting NET formation therapeutically and as a biomarker.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

MW and RS developed the outline and objective of the article, MW wrote the first draft of the article. TI and DN assisted with literature review, preparation of figures, and writing. YL added part of the content. MW and RS revised the article critically for its intellectual content. All authors approved the final version to be published, and take responsibility for the integrity of the content covered in this article.

This work was supported in part by the Science and Technology Program for Basic Research in Shenzhen (No. JCYJ20190809095811254 and JCYJ20200109140412476), Hospital Clinical Research Project (20213357002), NIH R01 AR050797, R01 AI080778, and R01 AR056465 grants.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.