key: cord-0280896-9h7wg6w2
authors: Stacey, Hannah D.; Golubeva, Diana; Posca, Alyssa; Ang, Jann C.; Novakowski, Kyle E.; Zahoor, Muhammad Atif; Kaushic, Charu; Cairns, Ewa; Bowdish, Dawn M. E.; Mullarkey, Caitlin E.; Miller, Matthew S.
title: IgA Potentiates NETosis in Response to Viral Infection
date: 2021-01-10
journal: bioRxiv
DOI: 10.1101/2021.01.04.424830
sha: c998de207b96758cdaad92af37551f2178fac40b
doc_id: 280896
cord_uid: 9h7wg6w2

IgA is the second most abundant antibody present in circulation and is enriched at mucosal surfaces. As such, IgA plays a key role in protection against a variety of mucosal pathogens, including viruses. In addition to neutralizing viruses directly, IgA can also stimulate Fc-dependent effector functions via engagement of Fc alpha receptors (FcαRI) expressed on the surface of certain immune effector cells. Neutrophils are the most abundant leukocyte, express FcαRI, and are often the first to respond to sites of injury and infection. Here, we describe a novel function for IgA:virus immune complexes (ICs) during viral infections. We show that IgA:virus ICs potentiate NETosis – the programmed cell death pathway through which neutrophils release neutrophil extracellular traps (NETs). Mechanistically, IgA:virus ICs potentiated a suicidal NETosis pathway via engagement of FcαRI on neutrophils through a toll-like receptor (TLR)-independent, NADPH oxidase complex-dependent pathway. NETs also were capable of trapping and inactivating viruses, consistent with an antiviral function.

The generation of NETs was first described by the Zychlinsky laboratory in 2004 as an 93 antibacterial effector mechanism 6 . NETs produced via a specialized form of programmed cell 94 dealth called "NETosis" and are composed primarily of decondensed chromatin studded with 95 antimicrobial proteins. Extensive work by many laboratories has since demonstrated that NETs 96

can have not only protective, but also pathogenic consequences in infections and many other 97 higher levels of NETosis than antibodies or virus alone, whereas IgG:IAV ICs did not induce 141

NETosis above background levels (Fig. 1A, B) . 142

In the context of IgG, bnAbs that bind to the stalk domain have been shown to potently elicit Fc-143 dependent effector functions, whereas antibodies that bind to the HA head domain and exhibit 144 hemagglutination inhibiting (HAI) activity do not. This is because HA stalk-binding bnAbs allow 145 for a two points of contact between target and effector cells 26, 27 . To determine whether broadly-146 neutralizing IgA:IAV ICs are primarily responsible for the induction of NETosis observed in the 147 context of IAV-specific polyclonal IgA, we used a panel of previously-described monoclonal 148 antibodies that bind to neutralizing epitopes on either the HA head or stalk domains 28-30 . The 149 antibody KB2 binds to the HA stalk domain of H1 viruses, while 29E3 is specific to the HA head 150 domain of Cal/09 31 . When human neutrophils were incubated with ICs containing a IAV:IgA 151 stalk-binding antibody (KB2), significant induction of NETosis was observed following 3-hour 152 stimulation (Fig. 1C ). In contrast, NETosis was not induced by IgG:IAV ICs, or by antibodies or 153 virus alone (Fig. 1C ). All ICs generated with a HA head-binding antibody (29E3) failed to induce 154 NETosis (Fig. 1D) . 155

In blood, IgA co-circulates with other antibodies, including IgG, which signals through distinct 156 FcRs (FcγRs) and can also induce NETosis 32 . Mixed ICs composed of IgG/IgA:HIV have also 157 been shown to act cooperatively to stimulate ADCC by monocytes 33 . We therefore tested whether 158 mixed ICs composed of IAV bound by IgA and IgG together would influence the magnitude of 159

NETosis induction relative to IgA alone. When ICs were generated with a 1:1 ratio of IgG:IgA, 160 the magnitude of NETosis induction was similar to IgA alone (Fig. 1E ). In serum, IgG is 161 significantly more abundant than IgA (approx. 4:1 to 10:1). Thus, to recapitulate the physiological 162 stoichiometry of IgG:IgA, we purified each immunoglobulin from serum of matched donors, and 163 then recombined them at their natural physiological ratio. NETosis was observed when neutrophils were exposed to 0.05 and 0.2 mg/mL of purified virus 176 ( Fig. 2A) . We then purified IgA from convalescence serum of a SARS-CoV-2 infected individual, 177 and a SARS-CoV-2 naïve individual, incubated them with sub-stimulatory concentrations (0.0125 178 mg/mL) of spike pseudotyped lentiviruses to allow for IC formation, and then incubated these 179 mixtures with primary human neutrophils from healthy donors. As we observed in the context of 180 IAV, IgA:virus IC generated with IgA purified from SARS-CoV-2 convalescence serum was 181 capable of stimulating NETosis, whereas pseudovirus:IgA mixtures from naïve serum was not (Fig  182   2B ). These results confirm that IgA:virus ICs more potently stimulate NETosis when compared to 183 virus alone, and that ICs are required for this potentiation, since IgA from seronegative individuals 184 did not significantly induce NETosis when mixed with pseudotyped lentivirus. 185

We also incubated neutrophils with antibody:HIV ICs which contained HIV-specific IgA isolated 186 from the serum of HIV+ individuals. Following stimulation, a significant increase in NETosis was 187 Neutrophils are professional phagocytes, and ADCP is one of the many Fc-mediated effector 229

functions that contribute to their defense against pathogens 24 . To directly measure whether IgA 230

ICs induced phagocytosis, fluorescent, protein L-coated polystyrene beads were complexed with 231

IgA or IgG prior to incubation with neutrophils. After incubation with beads, cells were washed 232 extensively to remove any beads that had not been phagocytosed. Significantly greater bead uptake 233 was recorded for neutrophils that were exposed to the IgG-opsonized beads compared to those 234 coated with IgA, which actually inhibited phagocytosis relative to protein L-coated control beads 235 PMAa far more potent stimulant, significantly induced NETosis beginning at 90 min after 255 stimulation (Fig. 4A) . Conversely, to inhibit the production of ROS, a small molecule inhibitor of 256 the NOX complex, diphenyleneiodonium chloride (DPI), was pre-incubated with neutrophils prior 257 to stimulation with IgA:IAV ICs. DPI completely inhibited NETosis induced by IgA ICs (Fig. 4B) . 258

Together, these observations demonstrate that IgA:virus ICs stimulate suicidal NET release in a 259 NOX-dependent manner. 260 261

In the context of bacterial infections, NETs exert antimicrobial activity by trapping and killing 263 bacteria with antimicrobial effector proteins associated with NETs. We thus set out to determine 264 whether NETs were similarly capable of trapping and inactivating virus. Neutrophils were either 265 left unstimulated or were treated with PMA to induce suicidal NETosis (virus containing ICs were 266 not used to avoid the confounding issue of having viruses present during induction of NETosis). 267 IAV was then incubated in wells of stimulated or unstimulated neutrophils, and unbound virus was 268 washed away. Using immunofluorescence microscopy, we observed that IAV particles become 269 trapped in NETs induced following stimulation with PMA (Fig. 4C ). Using ImageJ software, we 270 quantified GFP pixel density and normalized this to the number of cells (and/or NETs) per field. 271

Consistent with the stark visual contrast observed in the images, significantly more virus was 272 associated with PMA-stimulated neutrophils that had undergone NETosis than unstimulated 273 neutrophils (Fig. 4D) . 274

To test whether IAV was inactivated after being trapped in NETs, we used an mNeon reporter 275 virus 42 . IAV-mNeon was incubated with unstimulated neutrophils, PMA-stimulated neutrophils 276 that had undergone NETosis, or PMA-stimulated neutrophils treated with DNase to digest NETs. 277

DNase digestion specifically allowed us to test whether being trapped in a NET was necessary for 278 inactivation, or whether factors released by neutrophils during NETosis were alone sufficient to 279 inactivate IAV ( Supplementary Fig. 1, Fig. 4E In the context of IgG, immobilized ICs have been reported to stimulate NETosis via FcγRIIA. 299

Soluble ICs were primarily phagocytosed, but could be shifted to stimulate NETosis upon TLR7/8 300 activation, which resulted in furin-mediated cleavage and shedding of the FcγRIIA N-terminus -301

inhibiting further phagocytosis 32 . We observed that IgA ICs did not simulate phagocytosis, but 302 rather preferentially induced NETosis, even in the absence of TLR activation. While IgG ICs could 303 stimulate NETosis, the induction of NETosis was notably more pronounced upon stimulation of 304 neutrophils with IgA ICs. 305

In the context of IAV, bnAbs that bind to the conserved HA stalk domain have become a major 306 focus for the development of "universal" influenza virus vaccines and monoclonal antibody 307 prophylactics/therapeutics. Although bnAbs are relatively weak neutralizers of IAV, they confer suggests that IgA-stimulated NETosis is unlikely to occur in the airways where secretory IgA is 345 enriched, but instead would be expected to take place primarily in tissues and vasculature 52 . 346

The data presented here demonstrate that NETs can both trap and inactivate virus. This suggests 347 that they may have a protective antiviral function. We speculate that in individuals who lack virus-348 specific IgA, the high concentrations of virus needed to stimulate NETosis might exacerbate 349 inflammation and potentiate disease, as has been observed for those with COVID-19 14-16, 35 . 350

However, individuals with pre-existing immunitysuch as that conferred by vaccineslow levels 

Neutrophils were isolated from the peripheral blood of healthy male and female donors by 370 density gradient centrifugation as described previously 24 . Briefly, 3 mL of room temperature 371

Histopaque 1119 (Sigma-Aldrich) was added to a 15 mL falcon tube, followed by gentle addition 372 of 3 mL of Histopaque 1077 (Sigma-Aldrich). 6 mL of blood was layered on top and samples 373 were centrifuged at 930 x g for 30 minutes at room temperature (RT) with no deceleration in an 374 Allegra X-12R centrifuge (Beckman Coulter). The neutrophil layer was collected between the 375 Corporation). To purify monoclonal antibodies, clarified cell culture supernatants were applied 392 directly to Protein G-sepharose columns prior to washing and elution. 393

The variable light and heavy chain sequences of KB2 and 29E3 antibodies 30,53 were cloned into 395 cloned into pFuse vectors (pFUSE-hIgG1-Fc2 and pFUSE2ss-CLIg-hK, Invivogen). KB2 binds 396 to the stalk domain of H1 viruses, while 29E3 antibody is specific to the head domain of 397 A/California/04/09 (Cal/09). HEK293T cells co-transfected with pFUSE plasmids according to 398 manufacturer's recommendations and were subsequently purified from supernantants using 399

Protein G-sepharose columns, as described above. purified from the sera donors diagnosed with RA or healthy donors as described above. 474

Antibodies/viruses/ICs were then incubated with PMNs for 3 hours at 37°C before being fixed 475 with 3.7% paraformaldehyde (PFA) (Pierce Protein Biology) prior to staining and imaging. 476

Purified virus was plated on 15mm sterile coverslips in a 24-well plate at 2µg/mL and incubated 478 at 37°C for 18 hours. Wells were washed twice with PBS and 250 µg of either polyclonal IgA or 479

IgG was added for 30 minutes at 37°C. Wells were washed twice with PBS prior to the addition 480 of 4.0 x10 5 PMNs per well. PMNs were incubated for 3 hours at 37°C, fixed and stained as 481 described above. 482

15 mm glass coverslips were placed inside wells of a sterile 12-well plate and 4.0x10 5 PMNs 484

were added to each well in a total volume of 500 µL and allowed to settle for 1 hour. To block 485

FcαRI, 20 µg/mL of mouse anti-human CD89 antibody (AbD Serotec) was added to neutrophils 486 for 20 minutes at 4°C. PMNs were then stimulated with various conditions for 3 hours at 37°C 487 before being fixed with 3.7% paraformaldehyde (PFA) and stored at 4°C until staining. 488

Cells were fixed with 3.7% PFA (Pierce Protein Biology) at 4⁰C, washed in PBS three times and 490 then permeabilized using 0.5% Triton X-100 (Thermo Scientific) in PBS-T. 

Sterilized glass coverslips were placed in a 24 well plate, and neutrophils at 4.0x10 5 cells/well 506 were allowed to settle for 1 hour prior to stimulation. Neutrophils were stimulated with PMA for 507 3 hours at 37 °C, 5 % CO2. Cal/09 at 10 5 PFU/mL was then allowed to settle on the pre-formed 508

NETs for 3 hours at 37°C, following this incubation cells were fixed with 3.7% PFA (Pierce 509 Protein Biology). To stain coverslips for immunofluorescent imaging coverslips were treated in 510 the same was as previously described. Primary antibodies used included primary rabbit anti- were fixed with PFA and incubated with 1 µg/mL Hoechst 33342, trihydrochloride, trihydrate 530 (Life Technologies). 5-fields per condition were taken on the EVOS FL microscope, and % 531 infectivity was determined as the number of infected cells / the total number of cells. 532

This protocol was performed as previously described 24 . Briefly, fluorescent carboxylate 534 microspheres 0.5µm (Polysciences) were coated with protein L and polyclonal IgA or IgG and 535 were incubated with neutrophils at a 500:1 ratio at 37°C for 15 minutes with gentle mixing. This 536 was followed by centrifugation at 930 × g for 10 minutes. Cells were washed twice with PBS 537 before being plated in a 96-well plate. Fluorescence was measured with the SpectraMax i3 plate 538 reader at 526nm (Molecular Devices). 539

Neutrophils were purified as described above and allowed to settle on glass coverslips for 1 hour 541

at 37 o C. While settling, neutrophils were incubated with 20 µm of Diphenyleneiodonium 542 chloride (DPI) (Sigma-Aldrich) a neutrophil NADPH oxidase inhibitor. Neutrophils were then 543 stimulated with IgA:IAV ICs or PMA (0.1 mg/mL, Sigma-Aldrich) as a positive control and DPI 544 was maintained in the media. Cells were then fixed with 3.7% paraformaldehyde (PFA) (Pierce 545

Protein Biology) and stored at 4°C until staining and imaging. 546

Graphs and statistical analyses were generated using Graphpad Prism v9 (Graphpad Software, 548

San Diego, CA). A P value of < 0.05 was considered to be significant across all experiments. Fluorescent polystyrene beads were coated with protein L, followed by either polyclonal IgG or 591

IgA. Human neutrophils were isolated and incubated with the beads at a 500 beads/cell ratio. 592

After washing, phagocytosis of beads was measured using a SpectraMax i3 plate reader 593 

the Weston Family Microbiome Initiative

Thoracic Society Grants-in-Aid (M.S.M.), and the Michael G. DeGroote Institute for Infectious

M.S.M. was also supporting, in part, by a CIHR New Investigator 624 REFERENCES 633

The human immunoglobulin A Fc receptor FcαRI: a 635 multifaceted regulator of mucosal immunity

Neutrophil recruitment and function in health and 637 inflammation

IgA Fc receptors

Neutrophil extracellular traps in immunity and disease

A role for neutrophils in viral respiratory disease

Neutrophil Extracellular Traps Effectively Control Acute Chikungunya 647 Virus Infection

Neutrophils recruited to sites of infection protect from virus challenge 649 by releasing neutrophil extracellular traps

Neutrophil extracellular traps mediate a host defense response to human 651 immunodeficiency virus-1

Extracellular vesicles from CLEC2-activated 653 platelets enhance dengue virus-induced lethality via CLEC5A/TLR2

Neutrophil extracellular traps cause airway obstruction during 656 respiratory syncytial virus disease

Host DNA released by NETosis promotes rhinovirus-induced type-2 658 allergic asthma exacerbation

Excessive neutrophils and neutrophil extracellular traps contribute to 660 acute lung injury of influenza pneumonitis

Neutrophil extracellular traps infiltrate the lung airway, interstitial, 662 and vascular compartments in severe COVID-19

Neutrophil extracellular traps contribute to immunothrombosis in 664 COVID-19 acute respiratory distress syndrome

Targeting potential drivers of COVID-19: Neutrophil extracellular 666 traps

Broadly neutralizing anti-668 influenza antibodies require Fc receptor engagement for in vivo protection

Broadly neutralizing hemagglutinin 671 stalk-specific antibodies require FcγR interactions for protection against influenza virus in 672 vivo

Alveolar macrophages are critical for broadly-reactive antibody-mediated 674 protection against influenza A virus in mice

Immunogenicity of chimeric haemagglutinin-based, universal 676 influenza virus vaccine candidates: interim results of a randomised, placebo-controlled, 677 phase 1 clinical trial

Antibody-dependent cellular cytotoxicity in HIV infection

Broadly-Neutralizing Anti-Influenza Virus Antibodies: Enhancement of 681 Neutralizing Potency in Polyclonal Mixtures and IgA Backbones

Neutrophils Ameliorate Lung Injury and the Development of Severe 684 Disease during Influenza Infection

Broadly neutralizing hemagglutinin stalk-specific antibodies induce 686 potent phagocytosis of immune complexes by neutrophils in an Fc-dependent manner

Novel cell death program leads to neutrophil extracellular traps

Epitope specificity plays a critical role in regulating antibody-dependent cell-691 mediated cytotoxicity against influenza A virus

Optimal activation of Fc-mediated effector functions by influenza virus 694 hemagglutinin antibodies requires two points of contact

A Pan-H1 Anti-Hemagglutinin Monoclonal Antibody with Potent Broad-697 Spectrum Efficacy In Vivo

Characterization of a Broadly Neutralizing Monoclonal Antibody That 699 Targets the Fusion Domain of Group 2 Influenza A Virus Hemagglutinin

In Vivo Bioluminescent Imaging of Influenza A Virus Infection and 702 Characterization of Novel Cross-Protective Monoclonal Antibodies

Influenza Viruses Expressing Chimeric Hemagglutinins: Globular Head and 705 Stalk Domains Derived from Different Subtypes

TLR7 / 8 activation in neutrophils impairs 707 immune complex phagocytosis through shedding of FcgRIIA

Cellular Phagocytosis of HIV-1-Infected Cells Is Efficiently Triggered by IgA Targeting 710 HIV-1 Envelope Subunit gp41

SARS-CoV-2 triggered neutrophil extracellular traps (NETs) mediate 712 COVID-19 pathology

Neutrophil extracellular traps in COVID-19

IgA Complexes in 715 Plasma and Synovial Fluid of Patients with Rheumatoid Arthritis Induce Neutrophil 716 Extracellular Traps via FcαRI

The multifactorial role of neutrophils in 718 rheumatoid arthritis

Insights into IgA-mediated immune 723 responses from the crystal structures of human FcαRI and its complex with IgA1-Fc. 724

Species-Specific Recognition of Single-Stranded RNA via Toll-like 726 Receptor 7 and 8. Science (80-. )

NETosis: how vital is it?

Rationally designed 729 influenza virus vaccines that are antigenically stable during growth in eggs

NETs: a new strategy for using old weapons

High Level of Neutrophil Extracellular Traps Correlates With Poor 734 Prognosis of Severe Influenza A Infection

The role of nasal IgA in children 738 vaccinated with live attenuated influenza vaccine

Comparisons of the Humoral and Cellular Immune Responses Induced 740 by Live Attenuated Influenza Vaccine and Inactivated Influenza Vaccine in Adults

Comparative Immunogenicity of the

Trivalent IIV and LAIV against Influenza A Viruses in Children. Vaccines

Serum levels of immunoglobulins (IgG, IgA, IgM) in a 748 general adult population and their relationship with alcohol consumption, smoking and 749 common metabolic abnormalities

IgA Targeting Human 751 Immunodeficiency Virus-1 Envelope gp41 Triggers Antibody-Dependent Cellular 752

Cytotoxicity Cross-Clade and Cooperates with gp41-Specific IgG to Increase Cell Lysis. 753 Front

The era of the immunoglobulin A Fc 755 receptor FcαRI; its function and potential as target in disease

Protection of mice against lethal challenge with 2009 H1N1 758 influenza A virus by 1918-like and classical swine H1N1 based vaccines

WHO | Manual for the laboratory diagnosis and virological surveillance of influenza

NF-B Inhibition Facilitates the 763 Establishment of Cell Lines That Chronically Produce Human T-Lymphotropic Virus 764 Type 1 Viral Particles

Protocol and reagents for pseudotyping lentiviral particles with 766 SARS-CoV-2 spike protein for neutralization assays

This work was funded by grants from the Canadian Institutes of Health Research (CIHR) 621