Neutrophils are the first responders to infection and extravasate rapidly from the blood vessels into tissue. They are the most abundant leukocyte in blood, with about 10¹¹ neutrophils produced by the bone marrow each day, representing 40-60% of circulating immune cells in healthy adults (1). Neutrophils kill pathogens using oxidative burst, degranulation, phagocytosis and the release of neutrophil extracellular traps (NETs) (2, 3). Their role is most prominent in bacterial infection but they can also contribute to antiviral immunity.

Severe disease in COVID-19 is associated to increased neutrophil-to-lymphocyte ratio and high expression of neutrophil-related cytokines IL-8 and IL-6 in serum, and neutrophilia has been described as a predictor of poor outcome (4–14). Peripheral blood neutrophil counts in patients with COVID-19, although not as elevated as bacterial pneumonia, are higher in severe COVID-19 compared with mild cases and most other viral infections (4, 15). Neutrophils are associated with the development of thrombosis and pulmonary infiltrates found in post-mortem samples following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (16–18). In this Progress review, we focus on emerging data on the roles of neutrophils in the pathogenesis and response to SARS-CoV-2.

An altered neutrophil-to-lymphocyte ratio occurs in many conditions such as cancer, cardiovascular disease, sepsis and inflammatory disorders, including Systemic lupus erythematosus (SLE) and psoriasis (19). Patients with COVID-19 with severe disease had significantly higher absolute neutrophil counts (8) similar to the neutrophilia in both Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) (20). The limited antiviral response in COVID-19 may exacerbate neutrophil infiltration, resulting in exuberant inflammation (21).

A small gene ontology (GO) analysis of COVID-19 infected cells indicated that neutrophil activation and degranulation are the most activated cellular immune processes in COVID-19, but did not play a role in the antibody-mediated elimination of SARS-CoV-2 in a passive immunisation model (22). Neutrophils contribute to hypersensitivity pneumonitis in SARS-CoV-2 infection and altered neutrophil immunometabolism, with accumulation of succinate correlating with disease severity (21). A rat coronavirus (RCoV) model demonstrated that neutrophils produce cytokines and chemokines in response to alveolar epithelial cell infection with SARS-CoV-2, resulting in an inflammatory response which contributes to lung injury (23).

Neutrophil extracellular traps (NETs) are web-like chromatin structures released by neutrophils to degrade virulence factors and kill bacteria. Once unregulated in sepsis or severe COVID-19, they induce multiple organ damage, including arterial hypotension, hypoxemia, coagulopathy, renal, neurological, and hepatic dysfunction as consequence of a NETs-associated cytokine storm (24–26). Silva et al. found that gasdermin inhibition with disulfiram or genic deletion decreases NETs formation with reduced multiple organ dysfunction and mortality in a sepsis model (27). NETs concentration was markedly increased in the tracheal aspirate and plasma of patients hospitalised with COVID-19 as well as in SARS-CoV-2-infected lung airways and alveoli, with spontaneous NETs production from their neutrophils (13, 28–32). SARS-CoV-2 can directly induce healthy neutrophils to release NETs in vitro, which increase pulmonary epithelium cell death (28). NETs also appear to drive neuroinflammation in Ischemic Brain Damage (IBD) and IBD following COVID-19, by affecting the blood-brain barrier, promoting thrombosis, and by inducing neuronal damage through extruded NETs components, NETs-IL-1 loop and IL-17 cascades (33, 34), making them a promising target for therapy.

The first step in NETosis is cellular activation via pattern recognition receptors (PRR) such as Toll-like receptors 4 (TLR4), TLR7 and TLR8 in viral infections (24, 35, 36). Reactive oxygen species (ROS) are subsequently produced, resulting in the activation of protein arginase deiminase 4 (PAD4) which is responsible for chromatin decondensation (24, 37). Neutrophil elastase (NE), a granule protein, induces neutrophil nuclear membrane break down while granule protein gasdermin D facilitates pore formation in the cell membrane and mediates release of NETs into the extracellular space (Figure 1) (24, 31). NETs do play a role in viral clearance, but excessive NETs production exacerbates inflammation in acute respiratory distress syndrome (ARDS) and contributes to microvascular thrombosis (Figure 1) (38). These is potentially related to over-activation of the Stimulator of interferon genes (STING) pathway through cyclic GMP-AMP synthase (cGAS) in phagosomes, and by SARS-CoV-2 infection itself through Angiotensin-Converting Enzyme 2 (ACE2)-angiotensin II (39, 40). Pharmacological activation of the STING pathway may also regulate the effects of SARS-CoV-2 infection (41). NETs can also have different proteins cargo associated to their deoxyribonucleic acid (DNA), citrullinated histone 3 (cit-H3), NE, and myeloperoxidase (MPO) structure which can influence the type of immune response triggered (42). Severe COVID-19 patients were shown to have higher expression of the alarmin nuclear protein High mobility group box 1 (HMGB1), antiviral molecules like ISG-15 and LL-37, or functionally active tissue factor (TF) as protein cargo in NETs, produced mostly by normal density granulocytes (NDG) (43, 44). These cargo molecules induced thrombogenic activity and differential cytokines expression (43, 44).

COVID-19 is characterised by a cytokine storm and the Pyrin domain containing 3 (NLRP3) inflammasome has been implicated. The inflammasomes are molecular mechanism involving multiprotein complexes which regulate the production of pro-inflammatory cytokines. NLRP3, a member of the nucleotide oligomerization domain (NOD)-like receptor (NLR) family, is present in neutrophils (17). After NLRP3 activation, pro-caspase 1 is cleaved to the active form caspase 1, leading to the cleavage of pro-inflammatory pro-IL-1β and pro-IL-18 into the active forms (Figure 2) (45). Single-stranded ribonucleic acid (ssRNA) viruses, such as SARS-CoV-2, induce Nuclear factor kappa B (NF-κB) activation and the further production of pro-IL-1β and pro-IL-18 (45, 46). Simultaneously, ROS and Adenosine 5’-triphosphate (ATP) produced by mitochondria trigger NLRP3 inflammasome assembly (46). Active NLRP3 inflammasome is present in peripheral blood mononuclear cells (PBMCs) and post-mortem tissues of COVID-19 patients, and high expression of its derived products such as Casp1p20 and IL-18 were seen to correlate with disease severity and poor clinical outcome (47). NLRP3 inflammasome activation has also been described in neutrophils of severe COVID-19 patients (48). Aymonnier et al. found that neutrophils from COVID-19 patients with respiratory failure demonstrated NLRP3 inflammasome molecule Apoptosis-associated speck-like protein containing a CARD (ASC) specks, and their early formation in NETosis. In patients with severe COVID-19 neutrophils with intact multilobulated nuclei, ASC specks formation and histone H3 citrullination was elevated (48). In a murine model they also showed transient presence of ASC specks at the microtubule organizing center, before nuclear rounding, early in NETosis (48). In addition, SARS-CoV-2 has been shown to directly activate the NLRP3 inflammasome through viroporin protein 3a, which most likely acts by the formation of K⁺ and Ca⁺ channels (49). Such direct activation of the inflammasome leads to the production of IL-1β and IL-18, perpetuating inflammation and resulting in further neutrophil activation (50). NLRP3 inflammasome activation in the blood of patients reveals an impaired immature neutrophil response in severe COVID-19. Inflammasome signature analysis in circulating myeloid cells allows COVID-19 patients to be stratified and predicts evolution of disease severity (51).

Heterogeneity within the neutrophil population during infection has been demonstrated in multiple diseases, and different subsets have defined roles in influencing the inflammatory response (38, 52). Neutrophil subsets varying in their density, maturity and expression of surface markers have been reported in COVID-19 (53, 54). Classically, in sepsis, immature neutrophils are released from the bone marrow and Carissimo et al. found increased immature neutrophils in whole blood that correlated with increased IL-6 and IP-10, and COVID-19 disease severity (55). The ratio of immature neutrophils to gamma delta (Vδ)2 T cells could predict severe COVID-19 (55). Additionally, a shift toward immature neutrophils as the driver of hyperinflammation is associated with severe COVID-19 disease (56).

Recently there has been a renewed interest in immunomodulatory neutrophil subsets, specifically in the field of cancer, SLE and sepsis, including low density granulocytes (LDGs) and myeloid derived suppressor cells (MDSCs) (38), but there is not a consensus on nomenclature and classification (57). MDSCs are a mixed population of mature and immature cells with differing immunomodulatory roles (58). There is a lack of clarity on the phenotypical and functional characteristics of MDSCs and their relationship to LDGs but their defining characteristic is suppression of the adaptive immune response (59). MDSC expansion is linked to G-CSF, a cytokine increased in the lungs of COVID-19 patients (60) and almost 90% of mononuclear cells in the severe disease cohort were MDSCs. Proportion of LDGs increases with disease severity in COVID-19 patients, as well as their production of NETs when compared to healthy controls (43, 61).

Morrisey et al. described a population of LDGs correlating with disease severity and hypercoagulable state in COVID-19 patients (53). A population of CD45⁺CD66b⁺CD16^IntCD44^lowCD11b^Int LDGs was found in patients with severe disease, which displayed enhanced phagocytic capacity, spontaneous NETs formation and elevated cytokine production. Similarly, an immune-suppressive CD16^bright/CD62L^dim neutrophil subtype was increased in patients developing pulmonary embolism (PE) on the day of ICU admission (54). Using whole blood transcriptomics analysis, increased NLRP3 inflammasome, monocytes and LDGs were found in the lungs of COVID-19 patients, and neutrophil activation-associated signatures correlated to disease severity (62). In COVID-19, immature neutrophils are expanded and show increased programmed death ligand (PD-L) 1, which suppresses T cells, and reduced oxidative burst functions with no change in phagocytosis in severe COVID-19 (63). Chevrier et al. found higher LDGs were present in COVID-19 patients early in the course of the disease and decreased in convalescence using mass cytometry and serum proteomics, but CD16^low neutrophil population remained expanded over the disease course (64). COVID-19 induced-ARDS is associated with MDSC expansion, reduced lymphocyte function and arginine shortage, through increased arginase activity, therefore arginase supplementation may be therapeutic (65). Further study into the role of neutrophil subsets in COVID-19 is warranted, potentially as biomarkers of disease severity, or as new targets for therapeutic approaches.

Neutrophils express all known Toll-like receptors (TLRs) with the exception of TLR3 (72). TLR7, TLR8 and TLR9 are involved in the detection of ssRNA viruses such as SARS-COV-2 (73). Activation of these receptors leads to downstream activation of NF-κB and interferon regulatory factor (IRF7), and the subsequent production of pro-inflammatory cytokines and chemokines in neutrophils (Figure 2) (74). In conjunction with neutrophils, these pro-inflammatory cytokines and chemokines drive the characteristic hyperinflammation and pulmonary infiltration seen in severe COVID-19 (74). Neutrophils also produce type 1 interferons (IFN-α/IFNβ) through the activation of IRF proteins (75) and this broad, but dysregulated, pro-inflammatory and antiviral response puts selective pressure on these highly pathogenic respiratory viruses. The host response to SARS-CoV-2 has also been broadly defined as a significantly depleted type 1 IFN response, with a consistent upregulation of chemotactic signals (CCL8, CCL2, CXCL2, CXCL8 and CXCL9), most of which are key mediators of neutrophil recruitment. Liao et al. found that in the lungs of patients with severe COVID-19, macrophages exacerbate inflammation by producing chemokines that recruit neutrophils to the site of infection through chemokine receptors CC-chemokine receptor 1 (CCR1) and C-X-C chemokine receptor type 2 (CXCR2) (57). Using a SARS-CoV-2 animal model early induction of CXCL9 and CCL8 was found consistent with observations in primary human bronchial epithelial cells infected with SARS-CoV-2. At day 7, despite waning levels of virus, elevated CCR5, CCL2, CXCL9 and IL-6 were found in the animal model, suggesting neutrophil-mediated inflammation may persist after the virus has been cleared (76). This may correlate with the clinical findings of persistent symptoms and fatigue with post-viral infection complications in some patients.

The loss of IFN signalling is vital to understanding why SARS-CoV-2 elicits such a potent inflammatory and neutrophilic chemotactic response. For instance, bats appear to limit the inflammatory and neutrophilic chemotactic response when infected with coronaviruses endemic in the bat population (77). Banerjee et al. have proposed that bats possess repressors of NF-κB signalling, a potent inductor of pro-inflammatory and chemotactic responses, allowing these strains of the viruses to become endemic in the population. However, unlike bats, humans lack this repressor activity rendering us susceptible to this uncontrollable neutrophil-mediated inflammatory response following viral infection (77).

Coagulation cascade activation is a common finding in patients with COVID-19 and is associated with disease severity (78). Elevated levels of fibrin D-dimer degradation products, a marker of fibrin degradation indicating overactive coagulation, correlates with a worse clinical outcome (79). High plasma levels of plasminogen activator (tPA) and plasminogen activator inhibitor-1 (PAI-1) in hospitalised COVID-19 patients had strong correlations with neutrophil counts and activation, and extremely high levels of tPA increasing fibrinolysis (80). Plasmatic matrix metalloproteinase-9 (MMP-9) was likewise increased in COVID-19 patients which induced platelet and neutrophil activation, and NETs formation in vitro (81). Post-mortem studies have consistently shown that micro-thrombi are present throughout the pulmonary vasculature (82). Collectively, these data suggest that coagulation activation and vasculopathy within the lungs (pulmonary intravascular coagulopathy [PIC]) plays a role in modulating COVID-19 pathogenesis (78). The biological mechanisms through which SARS-CoV-2 infection causes PIC within the lung blood vessels remain poorly understood (83). However, recent autopsy studies have reported significant endothelial cell (EC) damage, apoptosis, loss of tight junctions and separation from the basement membrane (84). Local inflammation and dysregulated pro-inflammatory cytokine generation within the lungs are a major factor as well as local hypoxia and complement activation, which significantly enhance procoagulant pathways and downregulate anticoagulant pathways in vivo. Moreover, ECs express the ACE2 receptor through which SARS-CoV-2 gains entry into cells, and electron microscopy studies have reported viral inclusion bodies within ECs.

Neutrophils and platelets are key modulators of thrombosis. Significant NETosis is found in patients with severe COVID-19 and is important in thrombus aetiology (85). NETs can bind to platelets, triggering platelet activation, and through their citrullinated histone H3 (citH3) they can also interact with procoagulant von Willebrand factor (VWF) (85). In addition to their effects on primary hemostasis, NETs also enhance local thrombin generation. In particular, NETs initiate coagulation activation through the alternative contact pathway and trigger thrombin generation by enhancing the intrinsic tissue-factor dependent pathway. NETs have also been described to over-activate the STING pathway through the cGAS sensor in phagosomes (40). The over-activation of the STING-pathway increases hyper-coagulability via interferon-β and tissue factor, released by monocytes-macrophages, and can be inhibited upstream the STING-pathway by aspirin, intravenous immunoglobulins and Vitamin-D (40). NETs histones can activate platelets by stimulating platelet TLR4 and TLR2; neutrophils can bind to these active platelets through surface glycoprotein Ib to induce NETosis and, consequently, result in thrombosis (85). Platelet activation is associated with disease severity in COVID-19 (86). Finally, NETosis has potent pro-inflammatory effects on ECs, which serve to attenuate the normal ability of ECs to regulate procoagulant pathways (87, 88). NETs and thrombosis have been implicated in several disorders including cancer, SLE, rheumatoid arthritis (RA), atherosclerosis and ischemic stroke. NETs have been shown to invade microthrombi in septic patients and contribute to organ damage, hence it is likely that neutrophils are a mediator of organ dysfunction in COVID-19 (31).

The severity of COVID-19 differs between age-groups, and children, especially neonates, exhibit milder disease with only a small proportion require intensive care with acute respiratory illness. There are many theories about this discrepancy, which is also seen with other similar viral illnesses, and the decreased expression of ACE2 and NETs formation may be contributory (89). However, a multisystem inflammatory disease in children (MIS-C) or paediatric multisystem inflammatory syndrome temporally associated with COVID-19 (PIMS-TS) has emerged in children, occurring weeks after the primary infection with SARS-Cov-2, that can lead to serious and life-threatening illness in previously healthy children (90). There is no internationally accepted single definition of MIS-C/PIMS-TS, but most case definitions require multi-organ dysfunction, systemic inflammation evidence of recent a SARS-CoV-2 infection, and the exclusion of other causes. The clinical presentation and laboratory findings in MIS-C are similar to Kawasaki’s disease and toxic shock syndrome, and considered to be a spectrum of disease (90).

Similar to adults with COVID-19, neutrophilia and lymphocytopenia are common in MIS-C. Neutrophils play a key functional role in Kawasaki disease with recent descriptions of NETosis and neutrophil activation in the form of CD11b and CD64 production (91). Neutrophil counts predict responsiveness of patients with Kawasaki disease to intravenous immunoglobulin therapy, also used in MIS-C (92, 93). Neutrophils activation marker Fc γ receptor I (FcγRI; CD64) was described to be highly expressed on neutrophils of treatment-naive MIS-C patients in acute phase compared with healthy controls (94). These patients also showed increase levels of the neutrophils chemoattractant cytokine IL-8 (94). Ramaswamy et al. talk of a potential myeloid dysfunction in MIS-C patients based on the high expression of alarmin-related S100A genes in neutrophils and monocytes, and the significant reduction in key antigen-presentation molecules such as HLA class II and CD86 (95). Additional research is required to fully understand the role of neutrophils in MIS-C and to determine whether treatments used in Kawasaki disease such as intravenous immunoglobulin therapy could also be used with MIS-C patients.

The clinical syndrome of severe COVID-19 has several unique features, including, unusually for a viral infection, an increased neutrophil-lymphocyte ratio. Neutrophils play a role in viral clearance in terms of NETs and the production of IFN. However, neutrophils can have detrimental effects by aiding the pathogenesis of SARS-CoV-2 and exacerbating complications of COVID-19 such as ARDS, thrombosis and MIS-C. Understanding the role of neutrophils in the pathogenesis of severe COVID-19 may lead to identification of key therapeutic targets and/or biomarkers for early identification of patients who may benefit from immunomodulatory agents to control hyperinflammation and reduce mortality rates.

EM and AMM have performed literature research, designed the review layout, wrote, and revised the review. RW, AUM, NC, JO’D, CNC, TH, and NS have performed literature research, wrote, and revised the review. JI-C designed the review layout, and revised the review. ML has performed literature research, designed the review layout, wrote, and revised the review. EJM has performed literature research, designed the review layout, wrote, and revised the review. All authors agree to be accountable for the content of the work. All authors contributed to the article and approved the submitted version.

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.