Evaluating Humoral Immunity against SARS-CoV-2: Validation of a Plaque-Reduction Neutralization Test and a Multilaboratory Comparison of Conventional and Surrogate Neutralization Assays

ABSTRACT The evaluation of humoral protective immunity against SARS-CoV-2 remains crucial in understanding both natural immunity and protective immunity conferred by the several vaccines implemented in the fight against COVID-19. The reference standard for the quantification of antibodies capable of neutralizing SARS-CoV-2 is the plaque-reduction neutralization test (PRNT). However, given that it is a laboratory-developed assay, validation is crucial in order to ensure sufficient specificity and intra- and interassay precision. In addition, a multitude of other serological assays have been developed, including enzyme-linked immunosorbent assay (ELISA), flow cytometry-based assays, luciferase-based lentiviral pseudotype assays, and commercially available human ACE2 receptor-blocking antibody tests, which offer practical advantages in the evaluation of the protective humoral response against SARS-CoV-2. In this study, we validated a SARS-CoV-2 PRNT to assess both 50% and 90% neutralization of SARS-CoV-2 according to guidelines outlined by the World Health Organization. Upon validation, the reference-standard PRNT demonstrated excellent specificity and both intra- and interassay precision. Using the validated assay as a reference standard, we characterized the neutralizing antibody response in specimens from patients with laboratory-confirmed COVID-19. Finally, we conducted a small-scale multilaboratory comparison of alternate SARS-CoV-2 PRNTs and surrogate neutralization tests. These assays demonstrated substantial to perfect interrater agreement with the reference-standard PRNT and offer useful alternatives to assess humoral immunity against SARS-CoV-2. IMPORTANCE SARS-CoV-2, the causal agent of COVID-19, has infected over 246 million people and led to over 5 million deaths as of October 2021. With the approval of several efficacious COVID-19 vaccines, methods to evaluate protective immune responses will be crucial for the understanding of long-term immunity in the rapidly growing vaccinated population. The PRNT, which quantifies SARS-CoV-2-neutralizing antibodies, is used widely as a reference standard to validate new platforms but has not undergone substantial validation to ensure excellent inter- and intraassay precision and specificity. Our work is significant, as it describes the thorough validation of a PRNT, which we then used as a reference standard for the comparison of several alternative serological methods to measure SARS-CoV-2-neutralizing antibodies. These assays demonstrated excellent agreement with the reference-standard PRNT and include high-throughput platforms, which can greatly enhance capacity to assess both natural and vaccine-induced protective immunity against SARS-CoV-2.

of several efficacious COVID-19 vaccines, methods to evaluate protective immune responses will be crucial for the understanding of long-term immunity in the rapidly growing vaccinated population. The PRNT, which quantifies SARS-CoV-2-neutralizing antibodies, is used widely as a reference standard to validate new platforms but has not undergone substantial validation to ensure excellent inter-and intraassay precision and specificity. Our work is significant, as it describes the thorough validation of a PRNT, which we then used as a reference standard for the comparison of several alternative serological methods to measure SARS-CoV-2-neutralizing antibodies. These assays demonstrated excellent agreement with the reference-standard PRNT and include high-throughput platforms, which can greatly enhance capacity to assess both natural and vaccine-induced protective immunity against SARS-CoV-2.
Several laboratory-developed (5,6,15) and commercially available (16) enzyme-linked immunosorbent assays (ELISAs) have been developed to detect anti-SARS-CoV-2 antibodies raised upon infection. While these platforms provide a high-throughput means of detecting antibodies against SARS-CoV-2, they are unable to measure the immunological function of SARS-CoV-2-specific antibodies. In contrast, the plaque-reduction neutralization test (PRNT) quantifies levels of neutralizing antibodies capable of blocking the interaction that mediates virus entry into susceptible host cells and subsequent virus replication. For SARS-CoV-2, this interaction involves binding of the receptor binding domain (RBD) of the SARS-CoV-2 spike glycoprotein with the angiotensin-converting enzyme 2 (ACE2) (17). Antibodies targeting non-RBD epitopes of SARS-CoV-2 have also demonstrated low to high levels of neutralization activity (18)(19)(20). While the conventional PRNT is often used as the reference standard for the evaluation of virus-neutralizing antibodies, this assay is time-consuming and laborious and requires containment level 3 (CL-3) facilities to work with the risk group-3 pathogen. In contrast, laboratory-developed and commercially available surrogate SARS-CoV-2 neutralization tests, including microneutralization and human ACE 2 receptor-blocking antibody tests, have been developed for high-throughput evaluation of functional immunity against SARS-CoV-2 and may offer advantages in large-scale immunity testing.
In this study, we aimed to validate a laboratory-developed SARS-CoV-2 PRNT to assess both 50% neutralization (PRNT-50) and 90% neutralization (PRNT-90) of SARS-CoV-2 according to guidelines outlined by the World Health Organization (https:// www.who.int/medicines/areas/quality_safety/quality_assurance/28092018Guideline_ Validation_AnalyticalMethodValidation-Appendix4_QAS16-671.pdf) and characterized the neutralizing antibody response in laboratory-confirmed COVID-19 specimens. Using the validated PRNT as a reference standard, along with a panel of characterized serological specimens, we conducted a small-scale multilaboratory comparison of alternate SARS-CoV-2 PRNTs and surrogate neutralization tests to identify other methods to assess the neutralizing antibody response in SARS-CoV-2 infection.
Between several assay runs, the PRNT-50 demonstrated sufficient interassay precision, with 98.2%, 98.2%, and 80.0% of results within 4-fold of, within 2-fold of, and equal to the median test result, respectively ( Table 3). The PRNT-90 also demonstrated sufficient interassay precision, with 98.2% and 90.7% of results within 4-fold and 2-fold of the median test result; 68.5% of results by PRNT-90 were equal to the median test result. Both the PRNT-50 and PRNT-90 surpassed the validation cutoff, which required $90% of results generated to be within 2-fold of the median test result. The referencestandard PRNT demonstrated no reactivity with the COVID-191/neutralizing antibodynegative (nAb2), SARS-CoV-1, hepatitis, HIV, syphilis, and preoutbreak sera that were used in the panel for the multilaboratory assay comparison.
Small-scale multilaboratory comparison of conventional and surrogate SARS-CoV-2 neutralization assays. Alternate PRNT conducted by University of Alberta (U of A) demonstrated 100% congruence (P , 0.0001) and perfect interrater agreement with the reference-standard PRNT (k -value 1.000) as shown in Table 4. This assay detected 100% of laboratory-confirmed COVID-19 nAb1 specimens and did not cross-react with any specimens from the COVID-19 nAb2, SARS-CoV-1, hepatitis, HIV, or syphilis subsets. This assay also did not show any reactivity with sera collected from healthy patients prior to the COVID-19 pandemic. The alternate PRNT conducted by Wadsworth demonstrated 97.5% congruence (P , 0.0001) and almost perfect interrater agreement with the reference-standard PRNT (k -value 0.950). This assay detected 100% of laboratory-confirmed COVID-19 nAb1 specimens and did not show any reactivity with sera other than the laboratory-confirmed COVID-19 nAb2 specimen.  a Due to various host responses, specimens within the molecularly confirmed COVID-19 subset may include both nAb1 and nAb2 specimens. NA, not applicable. b Sensitivity: percentage of molecularly confirmed COVID-19 specimens which tested positive by SARS-CoV-2 PRNT. c Specificity: specimens collected from healthy patients prior to the COVID-19 pandemic which tested negative by SARS-CoV-2 PRNT. d Accuracy: 3 specimens positive for SARS-CoV-2 nAbs diluted at 1:4, 1:16, and 1:64 and undiluted and tested 4 times. A result was considered accurate if the absolute difference between the log 2 expected and log 2 observed value was #2.00. Accuracy of the PRNT-90 could not be assessed since the expected titers of diluted specimens fell below the limit of detection of the assay (1:20). Number of specimens and 95% confidence intervals are indicated in parentheses.
The conventional virus neutralization test (cVNT) conducted by BC CDC demonstrated 95% congruence (P , 0.0001) and almost perfect interrater agreement with the referencestandard PRNT (k -value 0.900; Table 4). It detected 94.7% of laboratory-confirmed COVID-19 nAb1 specimens and the COVID-19 specimen that tested nAb2 by the reference standard also tested negative by the cVNT. The fluorescence-activated cell sorter (FACS)based surrogate virus neutralization test (sVNT) by CRCHUM was performed as described previously (6) and demonstrated 92.5% congruence (P , 0.0001) and almost perfect interrater agreement with the reference-standard PRNT (k -value 0.851). It detected 100% of laboratory-confirmed COVID-19 nAb1 specimens. However, the COVID-19 nAb2 specimen tested positive by this assay.

DISCUSSION
The public health, economic, and societal impacts of the COVID-19 pandemic have led to expedited approval and emergency use of several promising vaccine candidates (https://www.who.int/news/item/15-02-2021-who-lists-two-additional-covid-19 -vaccines-for-emergency-use-and-covax-roll-out). However, since much remains unknown about long-term immunity against SARS-CoV-2, reliable methods to quantify functional humoral immune responses will be indispensable for understanding protection in both vaccinated and naturally infected individuals. In this study, we validated a laboratorydeveloped method of quantifying antibodies capable of neutralizing SARS-CoV-2 in vitro. The validated PRNT was used to characterize a subset of laboratory-confirmed COVID-19 serological specimens and subsequently used as a reference standard for the evaluation of other platforms for the analysis of the SARS-CoV-2-neutralizing antibody response.
The validated PRNT was then used as a reference standard to compare several alternate PRNTs and sVNTs for SARS-CoV-2-neutralizing antibodies. Both of the alternate PRNTs conducted by U of A and Wadsworth demonstrated excellent interrater agreement with the reference-standard PRNT. The only difference between the two assays was their result for the laboratory-confirmed COVID-19 specimen that tested nAb2 by the reference standard. The specimen tested positive by an ELISA detecting anti-SARS-CoV-2 nucleocapsid IgA, IgM, and IgG antibodies (Platelia SARS-CoV-2 Total Ab assay, BIO-RAD). However, the specimen tested negative by anti-SARS-CoV-2 spike 1 ELISAs, including ZEUS ELISA SARS-CoV-2 IgG test system and Euroimmun anti-SARS-CoV-2 ELISA (IgG and IgA). While the U of A assay agreed with the reference standard, the Wadsworth assays deemed the specimen to be positive for SARS-CoV-2-neutralizing antibodies. This observation could be interpreted in 2 different ways. First, one could consider the Wadsworth assay to demonstrate sensitivity greater than that of the reference standard and U of A PRNTs for detecting nAbs in laboratory-confirmed COVID-19 specimens. This is supported by the positive results generated by the cVNT (BC CDC) and the luciferase-based sVNT (CRCHUM). The second interpretation could consider the reference standard and U of A PRNTs to have a higher threshold for neutralizing activity. This is supported by the specimen testing negative for antibodies against the SARS-CoV-2 spike 1, which contains the RBD targeted by the majority of nAbs. Further research is needed to establish the threshold of neutralizing activity in vitro that correlates with neutralization and protection in vivo. One difference was that heat inactivation of serological specimens was conducted for the reference standard and U of A assays but not for the Wadsworth assay. Heat inactivation of serum at 56°C for 30 min is standard procedure in order to inactivate complement when evaluating neutralization (24,25). Complement has been demonstrated to enhance in vitro neutralization of viruses, including human cytomegalovirus (26), influenza virus (27), West Nile virus (28), and hepatitis C virus pseudotypes (29), directly and/or by enhancement of antibody neutralization. A recent study demonstrated that heat-inactivating serum specimens prior to immunoanalysis resulted in significantly lower observed levels of IgM and IgG anti-SARS-CoV-2 antibodies by ELISA (30). However, the effect was not characterized for PRNT and it remains unknown whether complement contributes to direct neutralization and/or enhancement of antibody-mediated neutralization in SARS-CoV-2 PRNT assays.
The other assays evaluated in this study demonstrate potential alternatives for the measurement of neutralizing antibodies against SARS-CoV-2. For example, aside from the alternate PRNTs, the cVNT (BC CDC) demonstrated the greatest interrater agreement with the reference-standard PRNT and offers practical benefits, such as a 96-well format to evaluate more samples per plate and decreased processing time due to the absence of a semisolid overlay step. However, like the PRNT, the cVNT still requires CL-3 facilities to manipulate live SARS-CoV-2 and analysis of neutralization can be affected by technician subjectivity. In contrast, the luciferase-based pseudotyped lentivirus sVNTs, the commercially available sVNT, and anti-RBD IgG ELISA are analyzed using automated platforms, decreasing processing time and observer subjectivity. In addition, these platforms can be conducted safely without a CL-3 facility, greatly expanding their use in lower-containment-level laboratories. However, cell-based assays, such as PRNT, cVNT, and luciferase-based pseudotyped lentivirus sVNTs, require highly skilled personnel and several days to conduct, hindering their practicality for use in large-scale immunity studies. While both anti-RBD IgG ELISAs and the commercially available sVNT are simple, high-throughput, and demonstrate excellent interrater agreement with the reference-standard PRNT, the anti-RBD IgG ELISA is a less expensive method to screen for the presence of potential neutralizing antibodies. Furthermore, the anti-RBD IgG ELISA has demonstrated near perfect agreement with another commercial sVNT (31). Thus, this ELISA could be used as a more cost-effective alternative to the commercially available sVNT as a screening assay prior to confirmation by PRNT, similar to a previously proposed algorithm (32).
It is crucial to recognize the small number of specimens in the panel used for the evaluation of these assays, which could limit the generalizability of the study. However, several of these protocols have been characterized further and demonstrated excellent performance in other studies (6,7,(33)(34)(35)(36)(37)(38). Importantly, our study demonstrates the proof-of-principle of using a sufficiently validated platform as a reference standard to directly compare nine serological methods to quantify and/or estimate the presence of neutralizing antibodies across several laboratories. Future evaluation of other alternate and surrogate platforms for the quantification of SARS-CoV-2-neutralizing antibodies will require a larger sample subset similar to the subset used for the validation of the reference-standard PRNT used in this study. Recently, the WHO has made recommendations to express serological data for neutralization in international units (IUs) as a way to effectively compare results of several studies (39). This standardization of results should be used in future studies using any of the assays evaluated in this study, especially in clinical studies, where it is crucial to be able to draw comparisons between trials.

MATERIALS AND METHODS
Ethics and biosafety statements. This study was approved by the Public Health Agency of Canada's Research Ethics Board (Protocol no. 2020-004P). All experiments involving SARS-CoV-2 (risk group 3 [RG-3]) were conducted in a containment level 3 (CL-3) laboratory. Procedures involving live SARS-CoV-2 were conducted using personal protective measures and operational practices for the manipulation of SARS-CoV-2 outlined by the Government of Canada (40).
SARS-CoV-2 (hCoV-19/Canada/ON_ON-VIDO-01-2/2020, EPI_ISL_425177) p3 stocks were produced by infecting Vero E6 cells at a multiplicity of infection (MOI) of 0.1. After 1 h of adsorption at 37°C and 5% CO 2 , DMEM supplemented with 2% FBS was added and flasks were placed in a 37°C incubator in an atmosphere of 5% CO 2 . The flasks were monitored daily under a light microscope to detect the presence of cytopathic effect (CPE) in infected cells and lack of CPE in the mock-infected cells. After 72 h postinfection (HPI), virus supernatant was collected and centrifuged for 10 min at 525 Â g and 4°C to remove cell debris. Aliquots of SARS-CoV-2 were prepared and transferred to a 280°C freezer for long-term storage. SARS-CoV-2 stocks were titrated using a previously described plaque assay (41).
Samples. Samples used for the validation of the reference standard SARS-CoV-2 PRNT were as follows: SARS-CoV-2 antibody-positive specimens comprised human serum and plasma samples from COVID-19 patients from Sunnybrook Hospital who were confirmed positive by molecular testing (Toronto Invasive Bacterial Diseases Network, ON; n = 218). All participants provided informed consent for collection of serum (Sinai Health Sytem REB no. 02118-U and no. 05-0016-C). The subset included 38,60,20,16,20,22,20, and 21 specimens collected at 1 to 7, 8 to 14, 15 to 21, 22 to 28, 29 to 35, 36 to 42, 43 to 49, and $50 days post-symptom onset. Of these samples, 24 were used for evaluating intraassay precision and 18 of these samples were used for evaluating interassay precision of the reference-standard PRNT. SARS-CoV-2 antibody-negative human serum samples were obtained from healthy adults prior to the COVID-19 pandemic (n = 52). The sample subset used for the multilaboratory comparison of alternative and surrogate platforms for the detection of SARS-CoV-2 were as follows: 19 specimens collected from laboratory-confirmed COVID-19 patients and confirmed positive for neutralizing antibody by reference standard SARS-CoV-2 PRNT (COVID-19 nAb1) and 1 specimen collected from molecularly confirmed COVID-19 patients and confirmed negative for neutralizing antibody negative by referencestandard PRNT (COVID-19 nAb2). In addition, the panel comprised laboratory-confirmed specimens from 2 SARS-CoV-1 patients, 5 hepatitis patients, 5 syphilis patients, and 3 HIV patients, as well as 5 specimens from healthy individuals collected prior to the COVID-19 pandemic. The latter were collected from individuals whose occupations required routine quantification of anti-rabies virus-neutralizing antibody titers postvaccination.
SARS-CoV-2 plaque-reduction neutralization test. The SARS-CoV-2 PRNT was adapted from a previously described method for SARS-CoV-1 (42). Serological specimens were diluted 1:10 in DMEM supplemented with 2% FBS and inactivated at 56°C for 30 min. In a 96-well plate, sera were further diluted 2-fold from 1:10 to 1:640 in DMEM supplemented with 2% FBS in a volume of 150 mL. One hundred fifty microliters of SARS-CoV-2 diluted at 100 PFU/100 mL was added to each well, yielding final serum dilutions of 1:20 to 1:1,280 and final virus concentrations of 50 PFU/100 mL. No neutralization, 50% neutralization, and 90% neutralization controls were prepared by diluting SARS-CoV-2 at 50 PFU/100 mL, 25 PFU/100 mL, and 5 PFU/100 mL, respectively. These were incubated with no sera. DMEM supplemented with 2% FBS was used as a no-virus control. After 1 h of incubation at 37°C and 5% CO 2 , 100 mL of each sera-virus mixture containing 50 PFU of SARS-CoV-2 was added in duplicate to 12-well plates containing Vero E6 cells at 95 to 100% confluence. One hundred microliters of each control was added in triplicate to two sets of 12-well plates containing Vero E6 cells. All plates were incubated at 37°C and 5% CO 2 for 1 h. After adsorption, a liquid overlay was prepared by adding equal volumes of 3% carboxymethylcellulose (CMC) and 2Â modified Eagle's medium (Temin's modification), no phenol red (2Â MEM; Gibco, Waltham, MA) supplemented with 8% FBS, 4 mM L-glutamine, 2Â nonessential amino acids, and 1.5% sodium bicarbonate. A total of 1.5 mL of liquid overlay was added to each well and plates were incubated at 37°C and 5% CO 2 for 3 days. The liquid overlay was removed and cells were fixed with 10% neutral buffered formalin for 1 h at room temperature. The monolayer in each well was stained with 100 mL 0.5% crystal violet (wt/vol) in 20% ethanol for 10 min and washed with 20% ethanol. For each specimen, the average number of plaques was calculated for each dilution and compared with the average number of plaques for 50% neutralization and 90% neutralization controls. The reciprocal of the highest serum dilution resulting in 50% and 90% reduction in plaques compared with controls was defined as the PRNT-50 and PRNT-90 endpoint titer, respectively. PRNT-50 titers and PRNT-90 titers of $20 were considered positive for SARS-CoV-2-neutralizing antibodies, whereas titers of ,20 were considered negative for SARS-CoV-2-neutralizing antibodies. Under the assumption that these specimens are negative for neutralizing antibodies, these specimens have a value of 0 for statistical analysis purposes.
Validation of reference standard SARS-CoV-2 PRNT. Following guidelines outlined by the World Health Organization (https://www.who.int/medicines/areas/quality_safety/quality_assurance/28092018Guideline_ Validation_AnalyticalMethodValidation-Appendix4_QAS16-671.pdf), the precision, accuracy, and specificity of the SARS-CoV-2 PRNT were assessed using both PRNT-50 and PRNT-90 titers. The intraassay precision of the PRNT was defined as the agreement of results generated by a single assay for the same homogenous sample. Intraassay precision was assessed by testing samples collected from laboratory-confirmed SARS-CoV-2 patients (n = 24) in triplicate in a single assay run. The interassay precision of the PRNT was defined as the agreement of results generated by several assays for the same homogenous sample. Interassay precision was assessed by testing samples collected from laboratory-confirmed SARS-CoV-2 patients (n = 18) in three independent assay runs by three different analysts. Intraassay and interassay precision were considered acceptable if $90% of the observed results were within 4-fold difference of the median titer for $80% of positive samples tested. For both intra-and interassay validations, the sample subset comprised high-volume specimens representing a variety of PRNT titers. Samples collected from laboratory-confirmed SARS-CoV-2 patients were used even if negative for SARS-CoV-2-neutralizing antibody; this was to ensure that even negative results were congruent within and between assay runs. Accuracy of the PRNT was defined as the degree of agreement of test results with the true value. Accuracy was assessed by testing laboratory-confirmed SARS-CoV-2 undiluted and diluted at 1:4, 1:16, and 1:64 in PBS 5 times each within the same assay run. The expected titer for each diluted specimen was defined as the mean of the four observed titers for the corresponding undiluted specimen divided by the dilution factor (4, 16, and 64, respectively). Accuracy was considered acceptable if the absolute difference of the log 2 observed mean titer and the log 2 expected titer was #2.00 for $80% results for each dilution. Specificity of the PRNT was defined as the percentage of SARS-CoV-2 antibody-negative samples from healthy individuals collected prior to the COVID-19 pandemic, which tested negative in the assay. A specificity of $95% was considered acceptable.
Alternate PRNTs. Alternate PRNTs were conducted by Wadsworth Centre (New York, NY, USA) and the University of Alberta (Edmonton, AB, Canada; REB Pro00099761). These assays were conducted similarly to the reference-standard PRNT, with a few distinct differences in SARS-CoV-2 stock production, virus dilution, challenge and adsorption, overlay, and visualization. These differences are summarized in Table 5.
SARS-CoV-2 microneutralization test: cVNT. A cell culture-based SARS-CoV-2 microneutralization test was developed and conducted in a CL-3 facility at the British Columbia Centre for Disease Control (BC CDC). Sera were incubated at 56°C for 30 min to inactivate the complement. Each serum was subjected in duplicate to 2-fold serial dilution from 1:8 to 1:4,096 in volumes of 100 microliters in a 96-well microtiter plate. To each dilution, 100 50% tissue culture infective dose (TCID 50 ) units of SARS-CoV-2 were added in a 50-mL volume and the preparations were incubated at 37°C for 2 h. Volumes of 100 mL of each dilution were inoculated into respective wells of a microtiter plate containing monolayers of Vero-E6 cells in MEM containing 2% fetal bovine serum. The cultures were incubated at 37°C in a CO 2 incubator and examined after 72 h for the development of characteristic cytopathic effect. The reciprocal dilution below the one at which cytopathic effect could be detected was deemed the titer of neutralizing antibody. Titers of $8 were considered positive for SARS-CoV-2-neutralizing antibodies, whereas titers of ,8 were considered negative for SARS-CoV-2-neutralizing antibodies. For each assay, previously tested sera were used as positive and negative controls.
McGill University. The SARS-CoV-2 spike (S) protein pseudotyped lenti-Luc virus was produced by transfecting HEK293T cells in a 10-cm dish with 3 mg of pSPAX2 (Addgene, 12260), 6 mg of Lent-Luc (Addgene,17477), and 3 mg of SARS-CoV-2 S (provided by Shan Cen, Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences) plasmid DNA using PEI (polyethylenimine). Forty hours posttransfection (HPT), viruses in the culture supernatants were collected, clarified by centrifugation at 3,000 rpm (Beckman GR-6S centrifuge) at 4°C to remove cell debris, and aliquoted, stored at 280°C. The quantity of virus stock was determined by measuring the copy number of viral RNA with reverse transcriptase-quantitative PCR (FRT-qPCR). The infectivity of the virus stock was measured by infecting HEK293T cells engineered to express ACE2 protein. The serological samples were first inactivated at 56°C for 30 min, then diluted in 100 ml DMEM containing 2% FBS in 5-fold from 1:2.5 to 1:1,562.5, mixed with 100 ml of lenti-Luc-SARS-CoV-2-S pseudotyped virus (8 Â 10 4 copies of viral RNA, diluted in DMEM with 2% FBS). After incubation for 1 h at 37°C and 5% CO 2 , virus/serum mixture was added to HEK293T-ACE2 cells that were seeded in a 96-well plate 1 day prior to infection (18,000 cells/ well). Each serum dilution was tested in three independent infections. After 2 h infection, virus/serum inoculum was replaced with 100 ml DMEM containing 10% FBS. Cells were cultured for 48 h at 37°C and 5% CO 2 , and then 100 ml of 2Â luciferase substrate (Steady-Glo luciferase assay system, Promega) was added for 20 min to lyse cells. Luciferase activity was measured with PerkinElmer EnSpire multimode plate reader. Luciferase values of infection without exposure to patient serum serve as the control infection. Fifty percent plaque-reduction/neutralization titer (PRNT 50 ) values were calculated using the GraphPad Prism software. A sample with a half-maximal inhibitory dilution (ID 50 ) of ,1:2.5 was considered "negative" and $1:2.5 was considered "positive" for SARS-CoV-2-neutralizing antibodies.
CRCHUM. The luciferase-based lentiviral sVNT was conducted as described previously (6). Singleround lentivirus particles expressing SARS-CoV-2 spike and luciferase were rescued by calcium phosphate transfection of 293T cells with the lentiviral vector pNL4.  .8], 1 mM ATP, and 1 mM dithiothreitol) and 50 mL of 1 mM D-luciferin potassium salt (Prolume). The neutralization half-maximal inhibitory dilution (ID 50 ) was determined using a normalized nonlinear regression and represents the plasma dilution to inhibit 50% of the infection of 293T-ACE2 cells by recombinant virus. A sample with an ID 50 of ,1:50 was considered negative and $1:50 was considered positive for SARS-CoV-2-neutralizing antibodies. Mount Sinai. The luciferase-based lentiviral sVNT was conducted as described previously (34,35). Lentivirus particles were rescued from HEK293T cells in a 6-well plate containing 2 mL growth medium (10% FBS, 1% penicillin/streptomycin [pen/strep] in DMEM). Cells were transiently cotransfected with 1.3 mg psPAX2 (Addgene, 12260), 1.3 mg pHAGE-CMV-Luc2-IRES-ZsGreen-W (BEI, NR-52516), and 0.4 mg HDM-IDTSpike-fixK (BEI, NR-52514) using 8 mL JetPrime (Polyplus-transfection SA, 114-01) in 500 mL JetPrime buffer. After 8 HPT, the medium was replaced by 3 mL of DMEM containing 5% heat-inactivated FBS and 1% pen/strep, and the cells were incubated for 16 h at 37°C and 5% CO 2 . Cells were then transferred to 33°C and 5% CO 2 for an additional 24 h. At 48 HPT, the supernatant was collected, centrifuged at 500 Â g for 5 min at room temperature, filtered through a 0.45-mm filter, and frozen at 280°C until use. For the neutralization assay, 2.5-fold serial dilutions of the serum samples were incubated with diluted virus at a 1:1 ratio for 1 h at 37°C before being transferred to plated HEK293-ACE2/TMPRSS2 cells and incubated for an additional 48 h at 37°C and 5% CO 2 . After 48 h, cells were lysed, and Bright-Glo luciferase reagent (Promega, E2620) was added for 2 min before reading with a PerkinElmer Envision instrument. A sample with an ID 50 of ,10 21.5 was considered negative and $10 21.5 was considered positive for SARS-CoV-2-neutralizing antibodies.
Commercially available hACE2 receptor-blocking antibody assay. The commercially available SARS-CoV-2 sVNT (L00847; GenScript, Piscataway, USA) was performed according to the manufacturer's instructions at McGill University (Montreal, QC, Canada). Briefly, serological specimens were incubated with horseradish peroxidase-conjugated RBD (HRP-RBD) at 37°C for 30 min. The mixtures were added to the hACE2-coated capture plate and incubated at 37°C for 15 min. Plates were then washed, removing HRP-RBD-neutralizing antibody complexes and allowing unbound HRP-RBD and HRP-RBD-nonneutralizing antibody complexes to remain bound to hACE2. 3,39,5,59-Tetramethylbenzidine (TMB) solution was added and allowed to incubate at room temperature for 15 min, and the reaction was stopped by stop solution. The optical density (OD) of each well was measured by spectrophotometry at 450 nm. The