Cohort information

The definition of clinical spectrum of SARS-CoV-2 infection followed the COVID-19 treatment guidelines of National Institutes of Health. The mild patients with COVID-19 were defined as those individuals who had any of the various signs and symptoms of COVID-19 (e.g., fever, cough, sore throat, malaise, headache, muscle pain, nausea, vomiting, diarrhea, loss of taste and smell) but did not have shortness of breath, dyspnea, or abnormal chest imaging. The moderate/severe patients with COVID-19 were defined as those individuals who experienced a fever (≧38℃) or respiratory symptoms and subsequently developed pneumonia requiring oxygen therapy or other complications within 14 days (inclusive), leading to hospitalization (including emergency room admission). The vaccinated COVID-19 patients received their vaccinations before being infected with SARS-CoV-2.

Recombinant protein, monoclonal antibody and F(ab')2 fragment generation

RBD recombinant protein (YP_009724390) was expressed by S2 cells and its purification was performed as previously described [21]. The ACE2-cross-reactive anti-RBD (mAb 127) and the anti-RBD mAb (LGSV201), which does not bind to ACE2, were generated from RBD-immunized mice as previously described (Fig. S1) [20]. Recombinant human ACE2 (rACE2) was purchased from Genetex (Irvine, CA). ZMAB-mouse IgG2b (cmIgG) was used as an isotype-matched antibody control (AB Biosciences, Concord, MA). The cmIgG and mAb 127 F(ab')2 fragments were prepared using a Pierce™ F(ab’)2 Preparation Kit (Thermo Fisher Scientific, Waltham, MA).

Indirect ELISAs

Diluted sera (1:100 for anti-nucleocapsid or anti-RBD antibodies, 1:50 for anti-ACE2 autoantibodies) or diluted monoclonal antibodies were added to the indicated protein-coated plate (2 µg/mL). The binding antibodies were detected by either goat anti-human IgG-horseradish peroxidase (HRP)-conjugated antibodies (Thermo Fisher Scientific) or goat anti-mouse IgG HRP antibodies (Leadgene Biomedical, Taiwan). For color visualization, the tetramethylbenzidine (TMB) reagent (Clinical Science Products, Mansfield, MA) was added, and the reaction was stopped by 2 N H2SO4. The absorbance at OD 450 nm was read by a VersaMax microplate reader (Molecular Devices, Sunnyvale, CA).

Serum preadsorption assay

Diluted sera (1:50 in PBS) were added to bovine serum albumin (BSA) (10 μg/mL, 100 μL)- or RBD (10 μg/mL, 100 μL)-coated wells. After incubation at 37 °C for 1 h, the diluted sera were transferred to wells that were coated with rACE2 (2 μg/mL, 100 μL). Bound anti-ACE2 IgG was detected using HRP-conjugated goat anti-human IgG antibodies (1:4,000) (Thermo Fisher Scientific). Positive anti-ACE2 autoAb is determined by the OD of sera binding to ACE2-coated ELISA plates when exceeding 0.39, as determined by the cutoff value (> 2.1-fold of OD values of anti-ACE2 IgG in unvaccinated healthy donors). Positive CR Ab was determined by the OD of sera binding to ACE2-coated ELISA plates when (OD of BSA preadsorption)-(OD of RBD preadsorption) > 0.1.

Isolation of human leukocytes, peripheral blood mononuclear cells, neutrophils, and platelets

Whole blood was collected in EDTA vacutainers and centrifuged to obtain buffy coats. The cells were subjected to RBC lysis and washed with PBS to isolate human leukocytes. Peripheral blood mononuclear cells (PBMCs) were isolated from the buffy coats by density gradient centrifugation using Histopaque-1077 (Sigma‒Aldrich, St. Louis, MO). The purity of the PBMCs was confirmed by flow cytometry (CytoFLEX S, Beckman Coulter, Indianapolis, IN) using an anti-human CD14 antibody (sc-1182, Santa Cruz Biotechnology, Santa Cruz, CA). Neutrophils were isolated from the buffy coats by density gradient centrifugation using Polymorphprep™ (ProteoGenix, Schiltigheim, France). The purity and viability of the neutrophils were confirmed by flow cytometry (CytoFLEX S) using an anti-human CD11b antibody (F-2648, Sigma‒Aldrich) and 7-AAD (BD Pharmingen, La Jolla, CA). Platelets were isolated from human whole blood that was collected in ACD vacutainers by centrifugation at 200 × g for 20 min. Platelet-rich plasma was obtained and then further centrifuged at 800 × g with 100 nM prostaglandin E1 for another 20 min. The platelet pellets were suspended in Tyrode’s buffer supplemented with 100 nM PGE1. The purity of the isolated platelets was determined by measuring surface CD61 expression (GTX61848, Genetex) using flow cytometry (CytoFLEX S).

In vitro human leukocyte stimulation

Human isolated leukocytes, PBMCs and neutrophils (2 × 106 cells/mL) were seeded in 48-well tissue culture plates and treated with the indicated conditions. Notably, to reduce potential interference from various components in the serum, we employed a preadsorption model for stimulation when treating cells with serum samples. Briefly, cells were preabsorbed with diluted sera in the presence or absence of ACE2 at 37 °C for 30 min. Then, the cells were centrifuged at 500 × g for 5 min to remove the supernatant and resuspended in RBD-containing medium. To inhibit mAb 127-induced NETosis, neutrophils were incubated with mAb 127 and RBD and cotreated with inhibitors (dasatinib, SML-2589; compound C, 171260; Ly294002, 19–142; Cl-Amidine, 50652, Sigma‒Aldrich). The levels of tumor necrosis factor-α (TNF-α), interleukin 1β (IL-1β), IL-6, and IL-8) and MPO in the supernatants were quantified by ELISA (R&D Systems Inc., Minneapolis, MN), and the degree of adhesion or NET formation was measured by immunofluorescence staining.

Neutrophil adhesion assay

The degree of neutrophil adhesion was determined as previously described [22]. After the indicated treatments, nonadherent neutrophils were removed by washing twice with PBS. The adherent cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100 in PBS. The number of adherent cells and the expression of a human granulocyte activation marker (CD66b) were detected by a mouse anti-human CD66b antibody (GTX19779, GeneTex) and Hoechst 33,342 (Invitrogen, Carlsbad, CA) and visualized by fluorescence microscopy (Olympus FluoView FV1000, Japan) and quantified using ImageJ.

Quantification of NET formation

The degree of NET formation was determined as previously described [23]. Isolated neutrophils under the indicated treatment were spun onto slides by a cytocentrifuge. After fixation and permeabilization, slides were incubated with mouse anti-human CD66b antibody (GTX19779, GeneTex) or rabbit anti-human citrullinated histone antibody (ab5103, Abcam, Cambridge, United Kingdom) and rabbit or mouse anti-human MPO antibody (A1374, ABclonal; GTX75318, GeneTex) and Hoechst 33,342. The degree of NET formation was determined by quantifying the overlap of CD66b or histones and MPO and nuclei using fluorescence microscopy (Olympus FluoView FV1000) and ImageJ.

NET purification and treatment with NET-containing supernatants

Isolated human neutrophils (5 × 106 cells/mL) were treated with 500 nM phorbol 12-myristate 13-acetate (PMA) for 4 h, and the NETs were purified as previously described [24]. Purified NETs were used at a concentration of 1 μg/mL as a positive control treatment. For the supernatant treatment, isolated human PBMCs, human umbilical vein endothelial cells (HUVECs) or isolated human platelets were incubated with the indicated conditioned medium and RPMI medium at a ratio of 1:2. PBMCs and HUVECs were stimulated for 24 h, while platelets were stimulated for 15 min.

Endothelial cell permeability assay

HUVECs were grown in Transwell plates (0.4 μm; Corning B.V. Life Sciences, the Netherlands) with 10% fetal bovine serum (FBS)-containing EGM-2 medium until a monolayer was formed. The upper chambers were reconstituted with the indicated conditioned media and RPMI medium at a ratio of 1:2 or with EGM-2 medium alone. After 24 h of incubation, the media in the upper chambers were replaced with serum-free media containing streptavidin-HRP (1:100, R&D Systems). After 15 min, medium from the lower chamber was collected, and HRP activity was measured by adding TMB reagent (Clinical Science Products). Color development was detected by a VersaMax microplate reader (Molecular Devices) at 450 nm.

Flow cytometry analysis of platelet activation

Platelet activation was determined by measuring P-selectin (CD62P) surface expression. After the indicated treatment, isolated human platelets (1 × 107 cells/100 μl) were incubated with FITC-conjugated anti-human CD62P (BD Bioscience, San Diego, CA) or FITC-conjugated isotype-matched antibodies (BD Bioscience) and analyzed using flow cytometry (CytoFLEX S).

HUVEC, neutrophil and platelet coculture model

In a 24-well plate, confluent HUVECs on coverslips were incubated with a total of 5 × 105 neutrophils and 1 × 107 platelets, followed by treatment with the recombinant RBD protein in the presence or absence of the indicated antibodies (mAbs 127, LGSV201, or isotype cmIgG) or dasatinib for 3 h. The cells were washed twice with PBS/2% FBS, fixed with 4% paraformaldehyde, and permeabilized with 0.1% Triton X-100. After blocking, the cells were incubated with a rabbit anti-human CD61 monoclonal antibody (GTX61848, Genetex) and a mouse anti-human CD66b monoclonal antibody (GTX19779, Genetex) and then incubated with fluorescently labeled antibodies against rabbit or mouse IgG (A-11017, A-11072, Invitrogen) and Hoechst 33342 for 1 h and visualized by fluorescence microscopy (Olympus FluoView FV1000, Japan).

Purification and preadsorption of immunoglobulin G (IgG) from COVID-19 patient serum

Human total IgG in the collected sera from COVID-19 patients or healthy donors (100 µl) was purified using Pierce Protein G Plus Agarose (Thermo Fisher Scientific). Elution was performed with 0.1 M glycine–HCl (pH 2.7) and immediately neutralized with neutralizing buffer (1 M Tris–HCl, pH 9.0). Purified human IgG underwent an exchange with PBS through Amicon Ultra filters with a 30 kDa molecular weight cutoff (Sigma‒Aldrich). To confirm that the concentration of purified total human IgG retained the characteristic features of CR Abs for neutrophil treatment, a range of 50–200 µg/mL of purified human IgG from different individuals was tested using both indirect ELISA and BSA- or RBD-preadsorption ELISA. For IgG treatment from COVID-19 patients, human isolated neutrophils underwent preadsorption with 100 µg/mL purified human IgG in the presence or absence of ACE2 at 37 °C for 30 min. Subsequently, the cells were centrifuged at 500 × g for 5 min to eliminate the supernatant and cultured in 10 µg/mL RBD-containing medium at 37 °C for 24 h. For inhibitor treatment, neutrophils were incubated with 10 µg/mL RBD and cotreated with 50 nM dasatinib at 37 °C for 24 h. The levels of IL-8 and MPO in the supernatants were quantified by ELISA (R&D Systems Inc.), and the degree of adhesion or NET formation was measured by immunofluorescence staining.

Statistical analysis

All statistical analyses were conducted using either an unpaired or paired Student’s t test for comparisons between two independent groups. One-way ANOVA or two-way ANOVA and post hoc Tukey tests were used for multigroup comparisons. Prism software (GraphPad Software Inc., CA) was utilized for statistical analysis. The data are presented as the means ± standard deviations (S.Ds.) from at least three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 and ns indicates no significance based on 95% two-tailed confidence intervals.