IMMUNOPATHOLOGY

,


Introduction
NETs are networks composed of extracellular DNA containing histones and cytotoxic enzymes, including myeloperoxidase and elastase, which are released by activated neutrophils as part of their microbicidal arsenal (Brinkmann et al., 2004). However, NETs can also induce disseminated vascular coagulation, when released into the circulation, as well as tissue damage when released in the extravascular space (Knight et al., 2014;Carmona-Rivera et al., 2020;Czaikoski et al., 2016;Papayannopoulos 2018). In this context, NETs have been described as a key mediator of the pathogenesis of various inflammatory conditions, including COVID-19 (Papayannopoulos, 2018;Leppkes et al., 2020;Veras et al., 2020;Ackermann et al., 2021).
We and others demonstrated that patients with severe COVID-19 have an increased number of circulating-and lung-infiltrated neutrophils, which release a large amount of NETs (Veras et al., 2020;Ackermann et al., 2021;Radermecker et al., 2020), mediating lung epithelial damage and disseminated intravascular coagulation.
However, the molecular mechanisms involved in NET release during SARS-CoV-2induced response were not addressed.
Recent works showed that Gasdermin-D (GSDMD) is critical to the release of NETs (Sollberger et al., 2018;Chen et al., 2018). GSDMD cleavage by inflammatory caspases generates two fragments N and C-terminal. The N-terminal (GSDMD-NT) oligomerizes with plasma and nuclear membranes, forming membranes pores that mediate cell death by NETosis (Sollberger et al., 2018;Chen et al., 2018;Kambara et al., 2018;Broz et al., 2020). During sepsis, the genetic deletion or pharmacological inhibition of GSDMD with disulfiram prevented the formation of NETs, protecting mice from organ damage development and increasing survival . Disulfiram is a drug that inhibits aldehyde dehydrogenase (ALDH) and is used to treat alcoholism (Koppaka et al., 2012). It was demonstrated that disulfiram can act as a potent inhibitor of GSDMD , and an observational study based on clinical records from the national US Veterans Affairs healthcare system revealed a reduced risk of SARS-In the present study, we identified a key role of the GSDMD pathway on NET release during COVID-19. Notably, the inhibition of GSDMD with disulfiram, abrogated NET formation, reducing lung inflammation and tissue damage in a mouse model of . These findings indicate GSDMD as a novel potential target for improving the COVID-19 therapeutic strategy.

Expression of activated GSDMD is associated with NETosis during COVID-19
To investigate the molecular pathway involved in NETs production in COVID-19, we first performed a single-cell RNA sequencing (scRNA-seq) analysis in public data of bronchoalveolar lavage fluid (BALF) from COVID-19 patients and healthy controls (Liao et al., 2020). Confirming previous findings clustering analysis showed that BALFs of patients with severe COVID-19 contained higher proportions of neutrophils than healthy controls or patients with moderate COVID-19 ( Fig. 1 A). The analysis of gene expression in these neutrophils revealed that GSDMD is expressed and most part of these cells also expressed other inflammasome-related genes (Fig. 1 B). We identified neutrophils with three molecular profiles according to the expression of inflammasome genes (Pycard, Casp4, and Casp1) as indicated by linked points (Fig. 1 C). Then, we enrolled 63 hospitalized patients with moderate [n=18, age 63 (±16.58) years, and 55.5% women] and severe [n=45, age 58 (±16.85) years, and 48.8% women] COVID-19. Assisted mechanical ventilation was implemented in 100% of patients with severe COVID-19. The serum levels of CRP, LDH, and fibrin degradation products D-dimers, were found above the normal range, indicating ongoing inflammation, tissue lesions, thrombosis, and subsequent fibrinolysis. The identified causes of death were pneumonia, ARDS, or multi-organ failure. Existing comorbidities are also reported in Table 1. Confirming previous findings (Veras et al., 2020), COVID-19 patients showed higher levels of NETs in the airway fluid ( Fig. S1 A). Notably, the cleaved fraction of GSDMD (GSDMD-NT) was found on these NETs (Fig. S 1B). Moreover, the image analysis of lung autopsies from patients who died by COVID-19 showed the presence of NET structure associated with activated GSDMD-NT fraction (Fig. 1, D-F). As a control, NETs and GSDMD-NT were not found in the tissues obtained in lung autopsies from patients who died of acute myocardial infarction (Fig. 1 D). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 1 G). Then, we analyzed the expression of GSDMD in blood neutrophils from COVID-19 patients and healthy volunteers. Using confocal microscopy, we found elevated expression of GSDMD accumulated on the neutrophil plasmatic membrane and in typical NETs structures containing extracellular DNA (Fig. 1, H and I). The western blot analysis confirmed that neutrophils from COVID-19 patients expressed higher levels of GSDMD-NT when compared to controls (Fig. 1J). Additionally, we observed that patients with severe COVID-19 showed elevated serum levels of NETs and GSDMD than heath controls and moderate COVID-19 patients (Fig. 1, K and L). A positive correlation was found between serum levels of NETs and GSDMD (Fig. 1 M).
Thus, these results indicate that GSDMD pathway could be involved in the process of COVID-19-induced NETosis.

GSDMD is required for NETs release by neutrophils from COVID-19 patients
Recent studies showed that disulfiram potently inhibits GSDMD . To investigate the potential role of GSDMD on NETs release during COVID-19, we added disulfiram on cell cultures of neutrophils from COVID-19 patients. We found that disulfiram inhibited the release of NETs in a concentration-dependent manner and the expression GSDMD-NT (Fig. 2, A-C). Importantly, the treatment with disulfiram also abrogated the activation of GSDMD and the release of NETs by SARS-CoV-2 infection in neutrophils (Fig. 2, D-F). Of note, disulfiram, at the used concentrations, did not inhibit the viral replication (Fig S 2). These results suggest that GSDMD is involved in the release of NETs triggered by SARS-CoV-2.

The cleavage of GSDMD depends on neutrophil infection by SARS-CoV-2
We previously demonstrated that SARS-CoV-2 is able to directly induce NET release by human neutrophils (Veras et al., 2020). Thus, considering that SARS-CoV-2 employs ACE2 and TMPRSS2 for host cell entry (Hoffmann et al., 2020), we investigated whether GSDMD activation in neutrophils is induced by SARS-CoV-2 via ACE2-TMPRSS2 axis during NETosis process. Neutrophils incubation with the inactivated SARS-CoV-2 show neither GSDMD activation nor NETs release.
Moreover, treatment of isolated neutrophils with a neutralizing anti-hACE2 antibody (αACE2) or camostat, an inhibitor of TMPRSS2, abrogated SARS-CoV-2-induced GSDMD activation and NETs release. Then, we treated SARS-CoV-2 infected neutrophils with tenofovir disoproxil fumarate (TDF), an antiretroviral that reduces SARS-CoV-2 replication through the inhibition of RNA polymerase (Clososki et al., is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022https://doi.org/10. .01.24.22269768 doi: medRxiv preprint 2020. Remarkably, TDF also reduced the GSDMD cleavage and release of NETs by neutrophils incubated with SARS-CoV-2 (Fig. 3, A-C). These results indicate that SARS-CoV-2 infection and replication activates GSDMD in neutrophils to release NETs. The lack of cleaved GSDMD and NET release in the culture with inactivated SARS-CoV-2 confirms that active viral infection might be necessary to trigger GSDMDdependent NETosis.
Inflammasome/GSDMD pathway is highly expressed in blood neutrophils from

COVID-19 patients
It is well established that inflammasome activation by canonical (caspase-1), or noncanonical pathway (caspase-4) cleaves and activates GSDMD (Chen et al., 2018;Shi et al., 2015). Considering these findings, we investigated whether these pathways are involved in GSDMD activation in COVID-19 patients. Confirming single-cell transcriptome data (Fig 1 C), neutrophils from COVID-19 patients also showed increased expression of active caspases-1 and caspase-4 ( Fig. 4 A). Also, we observed, that GSDMD activation and NETs production were abrogated in SARS-CoV-2-infected neutrophils from healthy individuals after treatment with inhibitors of caspase-1 (Ac-YVAD-CHO) or pan-caspases (Z-VAD-FMK) (Fig. 4, B-D). Neutrophils from COVID-19 patients also showed increased expression of RIG-I ( Fig. S 3), a viral sensor involved in the recognized RNA virus, which is implicated in inflammasome activation (Elion et al., 2018). These results indicate that the inflammasome is involved with the cleavage of GSDMD and NETs release triggered by SARS-CoV-2.
Epithelial and endothelial cell death elicited by SARS-CoV-2-induced NETs

requires GSDMD
In several pathological conditions, the release of NETs is associated with tissue damage (Knight et al., 2014;Carmona-Rivera et al., 2020;Czaikoski et al., 2016;Papayannopoulos 2018;Veras et al., 2020). COVID-19 is characterized by extensive tissue damage, mainly in the lung (Veras et al., 2020;Radermecker et al., 2020, Zeng et al., 2021. Therefore, we investigate whether inhibition of GSDMD prevents NETinduced cell damage. To this end, neutrophils from the blood of healthy controls were treated with disulfiram incubated with SARS-CoV-2 for 1 h, the cells were washed twice, and then co-cultured with a human alveolar basal epithelial cell line (A549) or endothelial cell line (HUVEC) for 24 h. Cell viability was determined (viability dye + cells) by flow cytometry. We found that SARS-CoV-2-activated neutrophils reduced the viability of A549 and HUVEC cells compared with non-activated neutrophils. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; Importantly, disulfiram treatment reduced the cell death induced by SARS-CoV-2activated neutrophils (Fig. 5, A-D and Fig. S 4). These results indicate that the GSDMD inhibition can prevent tissue damage mediated by SARS-CoV-2-induced NETs.
Disulfiram prevents NETs release and organ dysfunction in a COVID-19 experimental model To investigate the importance of GSDMD-dependent NETosis to COVID-19 immunopathology, we infected K18 hACE2 transgenic mice with SARS-CoV-2 and treated them with disulfiram. According to Oladunni et al. 2020, mice submitted to SARS-CoV infection showed a dramatic reduction in the overall survival curves. In a preliminary experiment, we confirmed these results. Therefore, we perform euthanasia on day 5 after the virus inoculation to obtain tissue and blood samples for analysis. We found that disulfiram treatment reduced the circulating levels of NETs after SARS-CoV-2 infection compared to the group treated with vehicle ( Fig. 6 A). Additionally, the levels of inflammatory cytokines IL-6, IL-1β, and CXCL1/KC, but not TNF-α, in lung tissue were also mitigated by treatment with disulfiram ( Fig. 6, B-E). Histological analysis of the lung tissue from SARS-CoV-2 infected mice showed a septal thickening by intense neutrophil infiltration with alveolar-capillary barrier damage. At higher magnification, we also observed the parenchymal lung remodeling with architectural distortion and an intense inflammatory cells infiltration with damage of pneumocytes and endothelial cells of the alveolar septa. Importantly, the treatment of infected mice with disulfiram reduced these inflammatory events, avoiding alveolar septal thickening and preserving the tissue histoarchitecture ( Fig. 6 F). Furthermore, the confocal microscopy analysis of lung tissues showed that disulfiram treatment markedly reduced the GSDMD-NT expression after SARS-CoV-2 infection (Fig. 6, G and H). Although lung injuries are a hallmark of COVID-19, evidence has shown that other organs are also affected Shi et al., 2020). Thus, we analyze the protective effect of disulfiram in other tissues. We observed that the heart of animals infected with SARS-CoV-2 showed diffuse and sparse cardiac interstitial inflammatory infiltrate, with perivascular accentuation (Fig. S 5A). In the kidney tissue, the SARS-CoV-2 infection-induced ischemic tubulointerstitial nephritis, mimicking acute tubular necrosis, which was associated with cell glomerulitis (Fig. S 5B). In the liver of infected mice, central-portal necroinflammatory hepatitis and spillover with piecemeal necrosis were found (Fig. S   5C). The treatment with disulfiram promoted the preservation of tissue architecture, reduced the inflammatory infiltrate, and attenuated tissue damage. Collectively, these . CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 findings demonstrate that pharmacological inhibition of GSDMD with disulfiram prevents NETs release and organ dysfunction and can be used to improve the COVID-19 treatment.
Remarkably, we reported that SARS-CoV-2 can directly stimulate human neutrophils to release NETs (Veras et al., 2020). However, how SARS-CoV-2 leads to the release of NETs is still unclear. In the present study, we reveal that the virus that causes COVID-19 directly triggers NET release by a GSDMD pathway-dependent manner. We found that GSDMD is expressed in lung tissue of patients with severe COVID-19 in association with NETs structures. Furthermore, the release of NETs by neutrophils infected with SARS-CoV-2 or isolated from patients with severe COVID-19 was inhibited with GSDMD inhibitor, disulfiram. Moreover, in a mouse model of COVID-19, the treatment with disulfiram abrogated NETs release and reduced organ damage.
A wide array of stimuli can trigger NET release. NETs production can be stimulated by toll-like receptors (TLRs) agonists, inflammasome assembling activators, or by immune complexes via Fc-receptors (Papayannopoulos et al., 2018;Ackermann et al., 2021;Chen et al., 2018). Previous findings demonstrated that a range of bacteria and viruses, including immunodeficiency virus-1 (HIV-1), chikungunya, respiratory syncytial virus (RSV) (Saitoh et al., 2012;Funchal et al., 2015;Hiroki et al., 2020), are able to induce NETs through TLRs and ROS pathways. Although we do not exclude the participation of TLRs and ROS production, especially in the case of secondary infection associated with COVID-19, we are showing that SARS-CoV-2 induced NETs release via inflammasome/GSDMD activation. The activation of this pathway depends on the virus entry into the cell via ACE2 and TMPRSS2 and also effective viral replication. Previous findings demonstrated that SARS-CoV-2 activates caspase-1 in monocytes inducing the production of inflammatory cytokines (Rodrigues., 2020). According to these observations, we demonstrated that after entering and replicating into the neutrophils SARS-CoV-2 activates . CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint inflammasome/caspases1/4, which in turn cleaves and activates GSDMD to finally mediate NETs release.
A possible link connecting the activation of caspases by SARS-CoV-2 relies on mechanisms triggered by RNAs sensors. Previous reports indicate that the intracellular RNA sensor RIG-I is involved in the recognition of SARS-CoV-2 (Yamada et al., 2021).
The RIG-I activation assembles caspase-1-activating inflammasome complexes and could induce GSDMD cleavage (Elion et al., 2018;Rintahaka et al., 2021). In this context, we observed an increase of RIG-1 expression in neutrophils from COVID-19 patients. Then, the possibility of RIG-I being involved in GSDMD activation via caspases/NLRP3 during SARS-CoV-2 infection deserves future investigation.
However, it is important to mention that all findings described here pointed out GSDMD as the final signal for the NETosis induced by SARS-CoV-2. Therefore, we propose GSDMD targeting to ameliorate COVID-19 therapy.
Disulfiram is a drug approved for the treatment of alcohol dependence for its inhibitory effect on aldehyde dehydrogenase (ALDH) (Koppaka et al., 2012;Wright and Moore, 1990). However, a study showed that disulfiram at nanomolar concentration covalently binds and modifies human/mouse Cys191/Cys192 in GSDMD inhibiting its poreforming function . Furthermore, we recently demonstrated that inhibition of GSDMD by disulfiram prevents neutrophil death by NETosis . Considering this finding, we tested the effect of disulfiram on the COVID-19 experimental model. We use K18hACE2 mice with humanized ACE2. K18 hACE2 transgenic mice succumbed to SARS-CoV-2 infection by day 6, with virus detected in lung airway epithelium and brain (Oladunni et al., 2020). We reproduced these data in our conditions at the dose of 2x10 4 PFU and all experiments were performed with this viral load. Considering that we perform daily treatment starting 24h after the virus inoculation, and on day 5 mice were euthanized to collect samples. This window of therapy with disulfiram was sufficient to reduce NETs concentration, cytokine storm, and attenuate important tissues damages in several organs. Similarly, we observed that the inhibition of GSDMD by disulfiram in the sepsis model efficiently abrogates NETosis, systemic inflammation, and vital organs dysfunction, improving mice survival . Of note, the course of the disease in the mouse model is different when compared to humans, in which the symptoms are observed on 5-6 days after the infection and maintained for around 14 days (Zhou et a., 2020). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10. 1101/2022 Although the effect of GSDMD inhibition by disulfiram may be clearly associated with the reduction of NETs, we do not exclude its effect in other cells, such as macrophages, blocking the release of inflammatory cytokines, as observed in the lung tissue of infected mice treated with disulfiram.
Taken together, our findings demonstrate that GSDMD plays a critical role in the generation of NETs and organ damage induced by SARS-CoV-2 infection. Therefore, the pharmacological inhibition of GSDMD, as with disulfiram, represents a potential strategy to improve the treatment of COVID-19.

Analysis of single-cell RNA-seq data
We analyzed single-cell transcriptomic data from bronchoalveolar lavage fluid (BALF) from patients with varying severity of COVID-19 disease and their respective healthy controls from a previously published cohort (Liao et al., 2020). The dataset generated by authors is publicly available at https://covid19-balf.cells.ucsc.edu/ . Basically, the dataset was downloaded and the RDS file was imported into R environment version v4.0.5. UMAP plots were generated using Seurat v4 (Hao et al, 2021). The overlap between gene lists was performed using the UpSetR package (Lex et al, 2014).

Patients
The inclusion criteria were individuals hospitalized with moderate or severe forms of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; ≥ 30 times per minute or SatO2 ≤ 93% and all of them required mechanical ventilation.
Both forms were established on the first day of hospitalization Jin et al., 2020). Importantly, the samples were collected on the day of admission. Although the patients were classified as moderate at the time of hospital admission, they received nasal oxygen supplementation during hospitalization, but none required mechanical ventilation. Peripheral blood samples and airway fluid were collected from 63 and 11 confirmed individuals hospitalized with COVID-19, respectively. In the present study, the SARS-CoV-2-negative control group (blood n=20, and airway lavage n=8) was designed to include matched subjects of older age and chronic non-  (Veras et al., 2020;Duarte-neto et al., 2019). We used, as non-COVID-19 control, lung samples from the autopsy of two patients with acute myocardial infarction under familial consent.

Airway lavage
This procedure was performed as previously described (Veras et al., 2020). Briefly, the airway fluid from 11 patients with COVID-19 patients was obtained by aspiration of the . CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint orotracheal tube. This fluid was mixed 1:1 with 0.1 M dithiothreitol (Thermo Fisher Scientific; cat. R0862), incubated for 15 min stirring every 5 min at 37°C. In the control group (n=8), the airway lavage was obtained by injecting sterile isotonic saline solution through a nasal fossa. The injected solution was mixed with nasal and nasopharyngeal secretions before being evacuated from the other nostril when it was collected directly in a sterile tube. The samples were centrifuged 750 g at 4°C for 10 min. The supernatants were used for measurement of NETs and the cells were fixed for immunostaining in coverslips with Poly-L-lysine solution 0.1% (Sigma-Aldrich; cat P8920).

NETs quantification by MPO/DNA conjugates in picogreen assay
This procedure was performed as previously described (  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. at a 630x magnification. All images acquired were analyzed using Fiji by ImageJ.
Finally, 10 fields per sample were randomically analyzed at the x and y focal plane.

Cells, virus, and mock
To obtain the SARS-CoV-2 virus or the control (Mock) the SARS-CoV-2 Brazil/SPBR-02/2020 strain was isolated from a nasopharyngeal swab of the first confirmed Brazilian cases of COVID-19 and expanded on Vero-E6 (African green monkey kidney) cells. Vero-E6 cells were cultured in DMEM high glucose supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, Utah) and 100 U/mL penicillin, and 1% μg/mL streptomycin (P/S Corning; cat. 30-002-CI), 1% glutamine (Corning; cat. 15718008), 1% hepes (Corning; cat. 25-060-CI), and 1% fungizone (Gibco; cat. 15290-018) at 37°C in the 5% CO2 atmosphere. Experiments were performed after one passage in cell culture when Vero-E6 cells with DMEM plus 2% FBS in 150 cm 2 surface area flasks were incubated at 37°C in 5% CO2 atmosphere. All procedures related to virus culture were handled at a biosafety level 3 (BSL3) multi-user facility, according to WHO guidelines. Virus titers were determined as the tissue culture infectious dose at 50% (TCID 50/mL). Virus stocks were kept in −80 °C ultra-freezers. The virus strain was sequenced to confirm the virus identity and its complete genome is publicly deposited (GenBank accession No. MT710714). Non-infected control cultures (mock) were prepared using pure non-supplemented DMEM as inoculum. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

Mouse infection and treatment
The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint et al., 2020). K18-hACE2 mice were obtained from The Jackson Laboratory and were bred in the Centro de Criação de Animais Especiais (Ribeirão Preto Medical School/University of São Paulo). For the experimental infection, animals were transferred to the BSL3 facility. Eight-week-old male K18-hACE2 mice were infected with 2x10 4 PFU of SARS-CoV-2 (in 40 µL) by the intranasal route as previously described (Oladunni et al, 2020). Twenty-four hours after the virus inoculation and once daily until day 5 post-infection (dpi), animals were treated with Disulfiram (50 mg/kg, i.p.) (n = 7) or vehicle (n = 7). Naive mice remained uninfected and untreated.
On the 5 dpi, 6 h after the injection of Disulfiram or vehicle, animals were humanely euthanized for samples collection. All the experimental procedures were performed in accordance with the guide for the use of laboratory animals of the University of Sao Paulo and approved by the institutional ethics committee (066/2020).

Drugs
To assess the pathways involved in the release of NETs, neutrophils were treated with is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint more precise result. The data are expressed as relative PFU (%) to that for the untreated virus-infected control cells, which was defined as 100%.

Cytopathic effect of SARS-CoV-2 infected neutrophils on A549 and HUVEC cell lines by flow cytometry
Neutrophils were isolated from healthy controls (1x10 6 ) incubated or not with disulfiram (10 μM) for 1 h. Next, were incubated or not with SARS-CoV-2 (MOI = 1.0) for 1 h. Then, the infected neutrophils were washed twice and co-cultured with Human alveolar basal epithelial (A549) or Human Umbilical Vein Endothelial Cells (Huvec) cell lines ( 2× 10 5 ) previously stained with CellTrace™ Violet Cell Proliferation Kit (Thermo Fisher #C34557) following manufacturer protocols, for 24 h at 37°C. Subsequently, the cells were harvested and suspended in buffer containing 2% FBS in PBS for further evaluation of cell viability by flow cytometry through staining with Fixable and viability dye eFluor™ 780 (eBiosciente™) following manufacturer protocol (Veras et al., 2020).
The acquisition of the cells was performed in a flow cytometer (FACS Verse) and the analyses were made using the FlowJo software (FlowCytometry Analysis Software v10).

Cytokine assays (ELISA)
The quantification of cytokines in mouse lung was conducted by a commercial ELISA kit (R&D Systems) according to the manufacturer's instructions. The individual sample's optical density was measured at 450 nm using a spectrophotometer (Spectra Max-250, Molecular Devices, Sunnyvale, CA, USA).

Histological examination
Mice were euthanized 5 days post-SARS-CoV-2 infection. The lung tissue was harvested and fixed in 4% buffered formalin and posteriorly embedded in paraffin blocks. Sections (3 μm) were then stained with hematoxylin and eosin for histological examination. Images were acquired by DMI 6000B microscope (Leica Microsystems) at a 400x magnification. Histological evaluation was performed by a pathologist.

Statistics
Statistical significance was determined by either two-tailed paired or unpaired Student t-test or one-way ANOVA followed by Tukey's post hoc test. Absolute numbers and percentages were compared with Fisher's exact test. Spearman's rank-order correlation (r) was calculated to describe correlations. P<0.05 was considered statistically significant. Statistical analyses and graph plots were performed with GraphPad Prism 8.4.2 software.

Supplementary material
In supplementary figure 1, we showed that GSDMD expression is associated with is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022.  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022   is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022  Human neutrophils were isolated from healthy control (n=7). Cells were treated with a . CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 neutralizing anti-hACE2 antibody (αACE2, 0.5 µg/ml), an inhibitor of the serine protease TMPRSS2 (camostat, 10 µM), or an antiretroviral that reduces SARS-CoV-2 replication through the inhibition of RNA polymerase -tenofovir disoproxil fumarate (TDF; 10 µM). After 1 h the cells were incubated with SARS-CoV-2, or virus control The data are expressed as mean ± SEM (*or # p<0.05, one-way ANOVA followed by Tukey's test in B and C). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ; https://doi.org/10.1101/2022.01.24.22269768 doi: medRxiv preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022.  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 Supplementary figure 2. Evaluation of the antiviral effect of disulfiram against SARS-CoV-2. Vero E6 cells were treated with disulfiram (using a 4-fold serial dilution) and then incubated 1 hour at 37ºC with approximately 90 PFU (Plaque Forming Units) of SARS-CoV-2. Four days after infection, viral release to the media was measured by the plaque-forming unit (PFU) assay.
Supplementary figure 3. RIG-I is highly expressed in neutrophils from COVID-19 patients.
Neutrophils were isolated from healthy controls (n=7) and COVID-19 patients (n=8). The neutrophil lysates were harvested for immunoblot analysis of RIG-1. The α-actin was used as a loading control.
. CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 Supplementary figure 4. Gating strategy for flow cytometry analysis. Blood isolated neutrophils (10 6 cells) from healthy donors, pretreated, or not, with disulfiram (30 µM) were incubated, or not, with SARS-CoV-2. After 1 h, these cells were washed twice and co-cultured with lung epithelial cells (A549, 2 × 10 5 cells) or endothelial cells (HUVEC, 2 × 10 5 cells) previously stained with viability dye for 24 h at 37°C. Gating strategy for flow cytometry analysis of A549 or HUVEC viability.
. CC-BY-ND 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 25, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022