key: cord-0681924-wwliuq71
authors: Goh, Yun Shan; Ng, Lisa F.P.; Renia, Laurent
title: A flow cytometry-based assay for serological detection of anti-spike antibodies in COVID-19 patients
date: 2021-06-19
journal: STAR Protoc
DOI: 10.1016/j.xpro.2021.100671
sha: aebfa4ff81ee67900c6d5970bbb96d7a62f3eb3d
doc_id: 681924
cord_uid: wwliuq71

One of the key public health strategies in coronavirus 2019 disease (COVID-19) management is the early detection of infected individuals to limit the transmission. As a result, serological assays have been developed to complement PCR-based assays. Here, we report the development of a flow cytometry-based assay to detect antibodies against full-length SARS-CoV-2 Spike protein (S protein) in COVID-19 patients. The assay is time-efficient and sensitive, being able to capture the wider repertoire of antibodies against the S protein.

Lead Contact *Correspondence: goh_yun_shan@IDlabs.a-star.edu.sg **Correspondence: renia_laurent@IDlabs.a-star.edu.sg Summary One of the key public health strategies in coronavirus 2019 disease (COVID-19) management is the early detection of infected individuals to limit the transmission. As a result, serological assays have been developed to complement PCR-based assays. Here, we report the development of a flow cytometry-based assay to detect antibodies against full-length SARS-CoV-2 Spike protein (S protein) in COVID-19 patients. The assay is time-efficient and sensitive, being able to capture the wider repertoire of antibodies against the S protein.

The protocol consists of four main parts. Three of the main parts are preparation steps to generate the S protein-expressing cells for the assay itself: (1) generation of transfer plasmid for transfection, (2) transfection to generate lentiviral particles, (3) transduction to generate S protein-expressing cells. The final part (4) involves the flow cytometry-based assay to detect specific antibodies against S protein.

Generation of transfer plasmid for transfection to generate lentiviral particles Note: The DNA sequence, encoding for the full length S protein, is codon-optimized (Table 1 ) and is chemically synthesized by Genscript. The lead time for the chemical synthesis of the DNA sequence by Genscript is about 2-3 weeks.

Below details the protocol to clone the S gene into the transfer plasmid, pHIV-eGFP.

For more info on the manufacturer's instructions, please refer to Table 2 at the end of this section.

Overview Day 1

Step 1  Preparation of vector Day 2

Step 2  Preparation of insert Day 3

Step 3 and 4  Ligation of insert fragment into vector backbone  Transformation of ligation mix into chemically competent bacterial cells Day 4

Step 5  Colony PCR Day 5

Step 6  Plasmid extraction Alternatives: Other plasmid extraction kits, such NucleoSpin Plasmid Mini kit (Macherey Nagel Cat# 740588.50) can also be used.

The extraction of the plasmid can be scaled up by extracting from a 100 ml culture, using a QIAGEN plasmid Maxi kit (#12162).

b. Sequence extracted plasmid using primers in Table 1 . Transfection to generate lentiviral particles HEK293T cells are transfected to generate lentiviral particles.

The culture medium for HEK293T cells is DMEM supplemented with 10 % fetal bovine serum and 1 % penicillin-streptomycin.

The lentiviral particles are generated, using transfer plasmids and the pMD2.G, pRSV-Rev and pMDLg/pRRE packaging system. This is a 3rd generation, 4-plasmid system.

Overview Day 1

Step 7  Seeding of cells Day 2

Step 8  Transfection  Medium change at the end of the day Day 4

Step 9  Harvest lentiviral particles 7. Day 1: Seed 0.5 x 10 6 HEK293T cells in 2 ml culture media into each well in 6 well plate.

The cells should be 70-80% confluent the next day (before transfection).

The cell number and transfection protocol below can be scaled by a factor of 0.4 if 12 well plate is used or by a factor of 2.5 if a 6 cm dish is used.

J o u r n a l P r e -p r o o f Transduction to generate S protein-expressing cells

Note: The transduction protocol described has been optimized using HEK293T cells. It has also been similarly applied to HEK293, EL4 and K562 cells. However, do ensure that all samples, that are going to be compared, are analyzed using the same cell line, as different cell lines might have different glycosylation modifications of the spike protein, affecting the antibody binding.

Overview Day 1

Step 10  Seeding of cells Day 2

Step 11  Transduction  Medium change at the end of the day Day 4

Step 12  Sort for GFP-positive cells 10. Day 1: Seed 0.12 x 10 6 HEK293T cells into each well in 12 well plate.

The cells should be 70-80% confluent the next day (before transduction).

11. Day 2: Transduction a. Add polybrene to viral supernatant (final concentration in well 8 µg/ml).

Note: Always include a negative control (a well where fresh culture media is added in place of the viral supernatant).

Note: It is recommended to determine the viral titer by qRT-PCR or p24 ELISA before transduction as different production lots might have different yields of virions. We have found that, if the transgene is with a phenotype detectable by flow cytometry (such as eGFP in this case), it is a better method of quantifying the viral titer than p24 ELISA or qRT-PCR (both of which measure incomplete/nonfunctional virus components in addition to functional virions 

Step-by-Step Method Details Flow cytometry-based assay to detect antibodies specific for SARS-CoV-2 S protein

All patients' plasma/serum samples are diluted 1 in 100, while all secondary and tertiary antibodies are diluted 1 in 600.

a. Remove spent media. b. Wash with PBS. c. Detach with ice-cold 2mM EDTA for 1-2 min. d. Wash twice with PBS by centrifugation at 300 x g for 5 min.

CRITICAL: Avoid using trypsin to detach the cells. The S protein is sensitive to trypsin cleavage.

Similarly, avoid using cell scrapper, as it might affect the expression of the S protein on the cell surface.

2. Seed 0.15 x 10 6 cells into each well in 96 V-bottomed well plates. a. All samples are analyzed in technical duplicates. b. Pellet the cells down by centrifugation at 300 x g for 5 min.

The assay has been optimized for 0.1 -0.25 x 10 6 cells/well. However, due to cell loss (through centrifugation), we recommend at least 0.15 x 10 6 cells/well.

3. Re-suspend cells in diluted plasma/serum samples. a. Dilute the samples at 1:100 in FACS buffer (10 % FBS diluted in PBS) prior to addition to cells. b. Ensure that negative and positive control samples are also included. Eg. Anti-spike monoclonal antibody (eg. Thermo Fisher Scientific #703958) can be used as positive controls and healthy control plasma/sera can be used as negative controls. 4. Incubate at 4 C for 30 min in the dark. 5. Wash twice with PBS by centrifugation at 300 x g for 5 min. 6. Re-suspend cells in diluted secondary antibody incubation.

a. Dilute the secondary antibodies at 1:600 in FACS buffer prior to addition to cells. b. For IgG, IgM and IgA detection, the secondary antibody is anti-human IgG, anti-human IgM and anti-human IgA Alexa Fluor 647 antibodies in FACS buffer with 1 µg/ml propidium iodide. c. For IgG subclasses detection, the secondary antibody is mouse anti-human IgG1, IgG2, IgG3 and IgG4 antibodies in FACS buffer.

Note: Other fluorophores, other than Alexa Fluor 647, can also be used. One other possible option is Alexa Fluor 405, which have little compensation issues with the GFP-positive cells and the propidium iodide staining. We have chosen Alexa Fluor 647 as there is also little compensation issues with the GFP-positive cells and the propidium iodide staining.

Note: In place of propidium iodide, DAPI can also be used for staining to differentiate live/dead cells. Alternatively, other live/dead viability dyes may be used.

7. Incubate at 4 C for 30 min in the dark. 8. Wash twice with PBS by centrifugation at 300 x g for 5 min. 9. For IgG and IgM detection, add 100 µl FACS buffer to the well. Re-suspend well and analyze by flow cytometry. 10. For IgG subclasses detection, re-suspend cells in diluted tertiary antibody incubation.

a. Dilute the secondary antibodies at 1:600 in FACS buffer prior to addition to cells. b. The tertiary antibody is anti-mouse Alexa Fluor 647 antibodies in FACS buffer with 1 µg/ml propidium iodide (PI; Sigma-Aldrich #P4170). 11. Incubate at 4 C for 30 min in the dark. 12. Wash twice with PBS by centrifugation at 300 x g for 5 min. 13. Add 100 µl FACS buffer to the well. Re-suspend well and analyze by flow cytometry.

a. Cells were gated on: (1) FSC-A/SSC-A to exclude cell debris (Figure 2A ), (2) FSC-A/FSC-H to select for single cells ( Figure 2B ), (3) FSC-A/PI to select for live cells (PI-negative population, Figure 2C ), (4) FITC/Alexa Fluor 647 ( Figure 2D -H). Binding is determined by the percentage of GFP-positive S protein-expressing cells that are bound by specific antibody, indicated by the events that are Alexa Fluor 647-and FITC-positive (Gate 2). A sample is defined as positive when the binding is more than mean + 3SD of the healthy controls.

Note: Cells are read on LSR4 laser (BD Biosciences), however, the cells can be read on any other cytometers with the following specifications (Table 3) . Using this assay, we are able to analyze the S protein-specific antibody profile of symptomatic and asymptomatic COVID-19 patients (Goh et al., 2021) . While the antibody levels are lower in asymptomatic patients, the assay is highly sensitive and detect 97% of the asymptomatic infections. We also found that IgG1 is the dominant IgG subclass in both symptomatic and asymptomatic patients.

Quantification of S protein antibody by flow cytometry Specific antibody binding to cells were determined by LSRII 4 laser (BD Biosciences) and analyzed using FlowJo (Tree Star).

1. Gate the cells based on the following: a. Forward (FSC) and side (SSC) scatter parameters, FSC-A/SSC-A, to exclude cell debris ( Figure  2A Binding is determined by the percentage of GFP-positive S protein-expressing cells that are bound by specific antibody, indicated by the events that are Alexa Fluor 647-and FITCpositive (Gate 2).

2. Define a sample as positive when the binding is more than mean + 3SD of the healthy control individuals. The thresholds using the healthy control readings is based on the normal-like distribution of the healthy control reading where a mean + 3SD threshold would mean that there is less than a 0.13% chance of a false positive. Note: In Goh et al. (Goh et al., 2021) , our sample size of healthy control individuals was 22 and the Receiver Operating Characteristic (ROC) curves were constructed from each of the antibody binding with the healthy control individuals and SARS-CoV-2 patients as the true negatives and true positives respectively using the pROC library in R version 3.6.4.

Limitations Similar to all serological assays, the risk of false positive diagnosis is one of the limitations of the assay. However, the assay consists of seven tests (IgM, IgA, IgG, and four IgG subclasses), allowing internal validation. Nevertheless, borderline positive results should be interpreted with caution. One other limitation of the SFB assay is the need for advanced planning. The assay is a cell-based assay, hence the dependence on cell culture requires careful planning ahead to ensure sufficient cell count. This limits the application of the assay for HTS. We suggest performing different serological assays in parallel: (1) this would complement each other to provide better diagnosis, and (2) other serological assays that allows high throughput screening application, could serve as the first round of screening, and the more sensitive SFB assay could provide confirmation and further investigation of borderline/discrepant samples. As the SFB assay is a cell-based FACS assay, the acquisition of the samples can time-costly, especially when the sample size is large.

Problem 1:

Inefficient digest of vector backbone (step 1 of Before you begin).

Potential Solution:

Set up the digest reaction with 10 U of enzymes in excess per 5 µg vector.

Problem 2:

No colonies following DNA ligation (step 3 and 4 of Before you begin).

The DNA ligation can be optimized by: 

a Vivaspin-20 centrifugal device (100 kD MWCO, Sartorius # VS2042)

Possibly due to significant cell loss throughout the assay. In this case, increase the cell number per and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact : Laurent Renia

Sensitive detection of total anti-Spike antibodies and isotype switching in asymptomatic and symptomatic individuals with COVID-19

The authors would like to thank the study participants who donated their blood samples to this study. The authors also wish to thank Ding Ying and team for their help in patient recruitment, Dr Danielle Anderson and her team at Duke-NUS for their technical assistance, and also to Dr Anis Larbi at the Singapore Immunology Network (SIgN), A*STAR for providing the healthy donor samples. Author Contributions YSG conceptualized study, designed and conducted the experiments, analyzed the data, and wrote the manuscript. LFPN and LR conceptualized study and wrote the manuscript. All authors revised and approved the final version of the manuscript.

A patent application for the SFB assay has been filed (Singapore patent #10202009679P: A Method Of Detecting Antibodies And Related Products). The authors declare no other competing interests.

J o u r n a l P r e -p r o o f are Alexa Fluor 647-and FITC-positive (Gate 2). (D) PBS control; (E) COVID-19 patient plasma, 1:100 diluted. Table 1 . DNA sequence of codon-optimised SARS-CoV-2 S gene and primers used to sequence full length SARS-CoV-2-S protein SARS-CoV-2 S gene Codonoptimised ATGTTTGTATTCTTGGTACTTCTCCCATTGGTATCTTCTCAATGCGTTAACCTTACCACACGCACCCAACT  GCCCCCGGCCTACACTAATAGCTTTACGCGGGGTGTCTACTATCCCGACAAAGTCTTTCGATCCAGTGT  GCTCCACTCCACCCAGGATCTTTTCCTTCCCTTTTTTTCTAATGTTACGTGGTTCCACGCAATCCATGTAT  CCGGTACGAATGGGACAAAACGCTTTGACAATCCAGTGCTGCCATTTAATGATGGAGTGTACTTTGCA  TCTACCGAGAAGAGTAACATCATCAGAGGATGGATCTTCGGAACGACCTTGGACTCCAAAACGCAATC  CTTGCTTATCGTTAACAATGCAACGAATGTTGTCATCAAAGTTTGCGAATTCCAATTCTGTAACGATCCC  TTCCTCGGTGTTTATTATCATAAAAATAATAAATCTTGGATGGAAAGTGAGTTCCGCGTATACAGTTCC  GCCAATAATTGTACCTTCGAATACGTAAGTCAACCGTTCTTGATGGATCTGGAAGGTAAACAGGGTAA  CTTTAAGAACCTTCGGGAGTTTGTTTTTAAGAACATAGACGGCTACTTTAAGATCTATAGTAAACATAC  GCCAATTAACTTGGTTAGAGATCTCCCGCAGGGGTTTTCAGCATTGGAGCCGCTCGTCGACCTCCCCAT  AGGTATAAATATAACTCGGTTTCAAACACTGCTGGCGCTCCACCGCAGCTACCTGACGCCTGGGGATT  CTTCTTCCGGTTGGACTGCAGGCGCTGCTGCATATTATGTAGGGTACCTGCAACCGAGAACCTTTCTCC  TTAAGTACAACGAGAATGGCACTATTACGGACGCTGTCGATTGTGCACTCGACCCCTTGAGTGAGACG  AAGTGTACACTGAAAAGCTTTACTGTTGAAAAGGGAATATATCAGACATCCAACTTTAGAGTTCAGCC  AACAGAATCCATCGTTCGATTTCCCAATATTACAAATCTCTGTCCGTTCGGAGAGGTCTTTAATGCTACC  CGATTCGCGTCAGTATACGCCTGGAACAGAAAGAGAATTTCTAACTGTGTTGCAGATTATAGTGTCCT  GTATAATTCTGCGTCTTTTAGCACTTTTAAGTGCTACGGCGTTAGCCCCACTAAGTTGAACGACCTTTGT  TTCACTAACGTGTATGCCGACTCATTCGTCATAAGAGGCGACGAAGTTAGACAAATTGCACCGGGCCA  GACGGGAAAGATTGCGGACTACAACTATAAATTGCCTGACGACTTTACAGGATGTGTCATCGCCTGGA  ATAGTAATAACCTTGACTCCAAAGTCGGTGGCAATTACAATTACTTGTACCGGCTGTTCAGGAAGTCTA  ATCTCAAACCTTTTGAGCGAGATATCAGCACGGAAATTTATCAAGCTGGTAGCACTCCATGTAACGGG  GTTGAGGGTTTTAATTGTTATTTTCCATTGCAATCATATGGATTCCAACCGACTAACGGTGTTGGGTAT  CAACCATACAGAGTGGTGGTTTTGTCATTTGAACTTCTGCATGCCCCTGCAACAGTGTGCGGACCGAA  GAAGAGTACGAACCTTGTAAAGAACAAGTGCGTCAACTTCAACTTTAATGGTCTGACGGGTACCGGCG  TTCTGACGGAATCCAATAAAAAGTTCTTGCCCTTTCAGCAGTTCGGGCGAGATATCGCCGACACTACTG  ATGCGGTGCGAGATCCTCAGACACTTGAGATCCTCGATATTACCCCATGTAGTTTTGGTGGTGTGTCTG  TGATTACACCCGGCACCAATACGTCAAATCAGGTCGCAGTCTTGTACCAAGACGTGAACTGCACCGAA  GTTCCTGTAGCCATTCACGCTGATCAATTGACACCGACATGGAGGGTGTACTCCACCGGATCTAACGT  GTTCCAGACCCGCGCGGGGTGTCTTATCGGCGCAGAACATGTGAACAACTCTTACGAATGTGATATTC  CTATCGGTGCAGGCATCTGTGCCTCATACCAGACACAAACGAACTCACCAAGGAGGGCAAGGTCAGT  AGCCTCACAAAGCATAATAGCCTATACGATGAGTCTTGGTGCGGAGAACTCTGTGGCGTACTCTAATA  ACTCTATCGCCATACCGACTAACTTCACCATTTCTGTTACGACCGAGATCCTCCCAGTTTCCATGACTAA  GACAAGTGTGGATTGTACAATGTACATCTGCGGCGACAGTACTGAGTGCAGTAACCTGCTTCTGCAGT  ACGGGTCCTTCTGCACACAACTTAACCGGGCGCTGACTGGTATAGCGGTTGAACAAGACAAGAACACT  CAAGAGGTCTTCGCACAAGTAAAACAAATATACAAAACACCACCTATTAAAGATTTCGGCGGGTTTAA  TTTTAGCCAAATCCTTCCAGACCCCAGCAAACCCTCTAAGCGCAGCTTCATTGAGGATCTGCTGTTTAA  CAAGGTCACCCTGGCAGACGCGGGCTTTATCAAGCAATACGGTGACTGCCTGGGGGATATCGCGGCT TGCGCTCCAGATACCGTTCGCGATGCAGATGGCGTATAGGTTTAATGGAATTGGTGTCACGCAAAACG  TTCTCTATGAAAACCAGAAGCTGATAGCAAATCAGTTCAATTCCGCGATTGGTAAGATACAAGATTCAT  TGTCTAGTACGGCCTCTGCACTCGGAAAACTCCAAGATGTAGTGAACCAAAACGCCCAAGCCCTGAAT  ACACTCGTAAAACAGCTCTCTAGTAATTTTGGGGCCATTTCCTCCGTATTGAACGACATCTTGAGTCGC  TTGGATAAGGTAGAAGCAGAAGTACAAATTGACCGGTTGATCACGGGCAGACTTCAATCACTTCAGAC  TTATGTTACTCAGCAGCTTATACGAGCTGCAGAAATTCGCGCCTCTGCGAACCTGGCCGCCACTAAAAT  GTCAGAATGTGTACTGGGACAGAGCAAACGGGTGGATTTCTGCGGAAAGGGCTATCATCTGATGAGT  TTTCCCCAGTCTGCGCCTCATGGTGTAGTATTTCTTCATGTCACATATGTACCAGCCCAAGAAAAAAATT  TCACAACGGCGCCCGCGATTTGCCATGACGGTAAGGCGCATTTTCCTCGCGAGGGCGTTTTCGTGTCT