Quantification of SARS-CoV-2 neutralizing antibody by wild-type plaque reduction neutralization, microneutralization and pseudotyped virus neutralization assays

13 Oct.,2022

 

Custom Die Casting Parts

microneutralization assay

Coronaviruses are enveloped, positive-sense, single-stranded RNA viruses that belong to the family Coronaviridae. Coronaviruses are known to cause diseases with a spectrum of severity in humans and other animals1,2. In December 2019, a new coronavirus emerged that was identified as a member of the Betacoronavirus genus. It was subsequently named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and is the causative agent of coronavirus disease 2019 (COVID-19). The spread of this virus led to a global pandemic affecting millions of people3. The search for an effective vaccine for COVID-19 is a global priority. Immunological assays, such as those for neutralizing antibodies, are important for providing information on correlates of protection for vaccine efficacy and data to support regulatory submissions for vaccine licensure.

Development of the protocol

The initial procedure developed was the plaque reduction neutralization test (PRNT), which measures neutralizing antibodies by in vitro virus neutralization. PRNT is considered the gold standard for assessing humoral correlates of protection for many viral diseases4,5,6. In the absence of phase III efficacy data, neutralizing antibody results have a key role in identifying optimal COVID-19 vaccine candidates and will be important in the regulatory licensure process. It is not yet known whether circulating neutralizing antibody levels can be considered a sole measure of protective immunity against COVID-19, but recent successful treatment of severely ill patients with convalescent plasma and monoclonal antibody therapeutics containing high levels of neutralizing antibodies suggest an important role for these antibodies7,8,9,10. However, PRNT has several limitations that make it unsuitable for large-scale studies of serum samples, such as phase III human clinical vaccine trials. PRNT is technically demanding, has very low throughput, is difficult to automate and, for SARS-CoV-2, has a long turnaround time due to the time taken for the virus to form visible plaques. In addition, the analysis of the count data is typically performed using Excel spreadsheet calculations to determine neutralization titers (e.g., Kärber formula11) or using free statistical packages such as R to perform Probit regression. These analysis solutions represent obstacles for assay validation, which may lead to difficulties in using the data as part of a regulatory submission. For these reasons, of the three assays presented in this article, only the microneutralization assay (MNA) and pseudotyped virus neutralization assay (PNA) (not the PRNT) have been qualified, as only data produced by these assays and analyzed by the validated version of the SoftMax Pro software are likely to be acceptable to regulatory agencies. The qualification results for the MNA and PNA can be found in Extended Data Figs. 1 and 2, respectively.

To improve on the limitations of the PRNT, an alternative assay called the focus-reduction neutralization test (more commonly known as an MNA) has been developed for other viruses12,13,14. This assay is performed in 96-well plates throughout and uses immunostaining to visualize infected foci, which can be counted using computer-controlled imagers (Fig. 1), substantially increasing assay throughput compared with the manual counting performed in PRNT. A further development of MNA is to replace the hazard group 3 wild-type virus with a hazard group 2 pseudotyped virus to perform a PNA. Pseudotyped viruses consist of nonreplicating genetically modified virions that comprise the structural and enzymatic core of one virus, such as vesicular stomatitis virus (VSV), while bearing the surface proteins of another virus, such as the SARS-CoV-2 spike, and encode a quantifiable reporter gene such as firefly luciferase15. In the PNA described in this protocol, the pseudotyped virus consists of a VSV that has been engineered to bear the SARS-CoV-2 spike glycoprotein on its surface and replace its G protein gene with a luciferase reporter. The last 19 amino acids of the spike protein have been deleted, which roughly corresponds to the cytoplasmic tail.

Fig. 1: General overview of the MNA described in Procedure 2.

Serum samples are diluted in a 96-well plate (Stage I), before wild-type SARS-CoV-2 virus is added to the diluted serum at BSL3 and neutralization allowed to occur (Stage II). The neutralized virus is then transferred onto Vero E6 cells (Stage III), allowed to adsorb, overlaid with viscous medium and incubated for 24 h (Stage IV). Plates are then fixed (Stage V), virus infected foci immunostained (Stage VI) and foci counted (Stage VII). Well counts are then analyzed with SoftMax Pro to determine median neutralization titers of the reference and test sera (Stage VIII).

Full size image

In this protocol, we describe procedures for performance of the classical PRNT, a newly developed MNA and a PNA for the measurement of SARS-CoV-2 neutralizing antibodies. The National Institute for Biological Standards and Control (NIBSC) has made available the 1st WHO International Standard for anti-SARS-CoV-2 antibody (20/136)16. In combination with this international antibody standard, we expect these assays to play a pivotal role in the development and licensure of urgently needed COVID-19 vaccines.

Applications

The MNA and PRNT described in this protocol have been applied to the testing of serum from participants in randomized controlled clinical trials for the ChAdOx1 nCoV-19 (AstraZeneca; Oxford University) and Ad26.COV2.S (Janssen) COVID-19 vaccines17,18,19. They have also been applied to the testing of serum from animal models20,21 and vaccine immunogenicity studies22. The wild-type assays may also be used with little adaptation to assess the in vitro efficacy of antiviral drugs23. All three assays permit the detection and measurement of titers of SARS-CoV-2 neutralizing monoclonal antibodies24,25 and could help to elucidate the structural basis of neutralization26. These assays form a critical readout for the development and licensure of current and future COVID-19 vaccines and, moving forward, will be a key element in demonstrating the bioequivalence of future vaccines.

Comparisons between the assays

The PRNT requires 5 d of incubation to develop countable plaques, whereas, in the MNA and PNA, plates are fixed after only 24 h incubation; this considerably shortens the turnaround time of these assays. The MNA and PNA are conducted in 96-well plate format throughout, whereas the PRNT is performed in a mixture of 96-well (for the neutralization stage) and 24-well (for the plaque assay) plates. This increases the number of samples that can be tested per operator on any given testing occasion in the MNA and PNA relative to the PRNT. The use of 96-well plates throughout the assay also facilitates increased automation of the washing, staining and reading stages of the assay. Automation is a key factor in reducing operator error and increasing assay precision.

In the MNA, immunostained foci are counted using a computerized counting system and, in the PNA, luminescence of the luciferase reporter is measured with a plate reader; whereas, in the PRNT, operators perform manual plaque counting. This improves accuracy and objectivity of the reading as well as increasing throughput. In the MNA and PNA, count and luminescence measurements, respectively, are analyzed automatically using curve-fitting software, which allows the majority of test samples to have a median neutralizing dose (ND50) assigned to them. In addition, the curve-fitting method provides a confidence interval around any result derived from the curve. The software used is GxP compliant, which lends itself to validation in accordance with the Good Clinical Practice (GCP) regulations required by regulatory authorities. In contrast, the most common methods for performing this determination for PRNT and some PNA assays27 are use of the Spearman–Kärber or Reed–Muench formulae, both of which require a full neutralization response (0–100%) and neither of which easily provide confidence intervals. These traditional calculations are generally carried out by hand or with the aid of a spreadsheet, representing additional challenges for assay validation to meet the requirements of regulators.

As wild-type virus contains the complete SARS-CoV-2 replicative machinery, both PRNT and MNA can be adapted to assess the efficacy of antiviral compounds, whereas the PNA is appropriate only for measuring neutralization activity of serum or monoclonal antibody-based therapeutics.

Limitations

Limitations of the PRNT are that it is slow to perform, has low throughput and is technically demanding. In addition, both the wild-type MNA and PRNT require access to Biosafety Level 3 (BSL3) facilities and suitably trained staff, whereas the PNA can be conducted in a BSL2 facility. MNA and PNA correlate well with the ‘gold standard’ PRNT when tested with a panel of convalescent human serum; live virus PRNT with MNA (Pearson r = 0.963; P < 0.001) and PRNT with PNA (Pearson r = 0.862; P < 0.001) (Extended Data Figs. 3 and 4). During qualification, assay performance is assessed and documented for key performance metrics, e.g., specificity, linearity, bias, precision and limits of quantification. Both the MNA and the PNA have been qualified and will be utilized in regulatory submissions. However, as the pseudovirus used in PNA contains only the SARS-CoV-2 spike antigen and not the full complement of viral antigens, there is a risk that the PNA might fail to correlate well with the wild-type assays for some vaccine candidates, particularly those containing non-spike subunits of SARS-CoV-2. It is therefore crucial that, for each novel vaccine candidate, serum responses are correlated between MNA and PNA before transferring serum antibody assessments over to the PNA.

A potential limitation of all neutralization assays utilizing wild-type virus relates to the testing of plasma instead of serum derived from clotted whole blood, as plasma is often derived from whole blood that has been collected into tubes containing heparin as an anticoagulant. There have been reports that heparin is an effective inhibitor of SARS-CoV-2 entry into cells, and therefore there is a risk that this sample type might give rise to erroneous results from these assays23,28. Other common anticoagulants have not been reported to exhibit nonspecific antiviral activity.

A final limitation of the PNA is that the assay is appropriate only for measuring neutralization activity of serum or monoclonal antibody-based therapeutics, whereas both wild-type virus assays can be adapted to assess the efficacy of other antiviral compounds. This is because the pseudovirus does not contain the coronavirus replicative machinery that such compounds are directed against.

Alternative methods

All measurements of neutralizing antibody titers are generally performed by PRNT, some form of the MNA, or assays using pseudotyped viruses (e.g., the PNA). PRNT tests generally follow the procedure described here with only minor changes, mainly to the source of the wild-type virus, the semi-solid overlay used (agarose, methylcellulose or AviCell) and the method of analysis. Of these, only the method of analysis is worthy of note. As discussed earlier, methods such as Reed–Muench and Spearman–Karber are often still used to estimate median neutralization doses. In the past, this was due to their ease of calculation; however, curve-fitting methods have now been demonstrated to provide more accurate results29, and their computation no longer presents a barrier. The primary advantage of PRNT over other methods is the ability to detect more subtle changes, such as those to plaque morphology, in response to a serum sample, antibody or therapeutic, as well as a reduction in total plaque number. For example, antibody enhancement may cause plaques to appear larger than normal, or certain types of incomplete neutralization may cause a reduction in plaque size.

Large increases in scale and sample throughput can be achieved by replacing the PRNT with an MNA. In addition, as MNA is performed in a 96-well format, the assays are more amenable to automation of several assay steps. The two main types of MNA differ in the method of readout; some assays (such as those presented here) are read by counting immunostained foci (spots), whereas others are read on an ELISA reader measuring a colorimetric dye such as o-phenylenediamine30. The advantage of foci counting is that microbial contamination of samples or other mishandling causing cell monolayer damage can be seen during the plate quality control stages of the assay; whereas a colorimetric assay lacks this opportunity to reject the well or sample containing the error, leading to an aberrant absorbance reading. The advantages of the colorimetric methods are that they are generally faster to read, require less expensive equipment to perform measurements, and setting up the curve-fitting software is more straightforward. In the MNA presented here, the primary focus detection antibody is directed at the SARS-CoV-2 spike receptor binding domain, which is presented on the surface of infected cells; whereas, in other assays, the primary antibody is directed against the viral nucleocapsid30, which requires the staining procedure to have an additional cell-permeabilization step that increases assay time. The advantage of using a primary antibody against nucleocapsid is that, if one is working with virus variants with many mutations in the spike protein, it is conceivable that the anti-spike antibodies may no longer bind to the spike; however, this is less likely to occur to the nucleocapsid, as it is not under the same immune pressure in the host.

All wild-type assays suffer from the limitation of requiring BSL3 facilities and highly trained staff to perform them. PNAs allow neutralization assays to be performed in a standard BSL2 laboratory. These assays retain the advantages of the MNA, of being more scalable and having higher throughput than the PRNT. The readout of the PNA is luminance, which has the same advantages and disadvantages discussed for colorimetric methods in the MNA above. Median neutralizing doses can be determined using in-house Excel calculations (e.g., Reed–Muench method) or curve-fitting functions in off-the-shelf statistical packages such as GraphPad Prism27,31, but these approaches are unlikely to be acceptable to vaccine licensing agencies. It is preferable, as presented here, to perform calculations within a GxP validated version of curve-fitting software such as SoftMax Pro.

Overview of the procedures

For all neutralization assays described here, the serum must first be heat-treated to inactivate complement using the method described in Box 1. For both wild-type virus assays, PRNT (Procedure 1) and MNA (Procedure 2), a stock of virus needs to be produced as described in Box 2. For the PRNT, the wild-type virus stock is subjected to a rough sighting to estimate the dilution at which to prepare the virus, followed by the fine sighting (Box 3). Once a given virus stock has been sighted into the assay in this manner, samples can be analyzed by the PRNT by following the instructions in Procedure 1. A similar process is used to test a new virus stock in the MNA but can be more simply performed by the semifine sighting (Box 4), which includes regression analysis to determine the working dose for the assay. Once the virus has been sighted into the MNA, samples can be analyzed by following the instructions in Procedure 2. Resighting of the PRNT or MNA would only be required when a new stock of wild-type virus is produced or if the virus-only control (VOC) counts were outside the specified range for the assay.

For the PNA (Procedure 3), stocks of pseudovirus should be produced by following the procedure in Box 5. New virus stocks should be subjected to the sighting procedure (Box 6). Once the optimal pseudotyped virus dilution for a given stock is known, Procedure 3 can be started. Resighting would be required in the event of a new batch of pseudotyped virus being produced or if the VOCs (pseudotyped virus control; PC) luminescence measurements were outside the specified range.

Box 1 Sample-preparation procedure for all assays

Timing 1 h

Procedure

Critical

Before assessment of neutralizing antibody activity in serum samples, it is essential to inactivate components of the complement system as they are known to interfere with these assays.

Caution

Appropriate national laws and institutional regulatory board ethical guidelines must be followed and informed consent obtained from patients for the use of human serum in these protocols

  1. 1

    Heat the water bath to 56 °C (±1 °C), and confirm with a calibrated thermometer.

  2. 2

    Briefly centrifuge serum samples at 2,000g for 10–30 s to ensure that the entire sample is collected in the bottom of each tube.

  3. 3

    Place the tubes containing the serum into the water bath for 30–40 min.

  4. 4

    Mark tubes to confirm heat inactivation has been performed.

    Pause point

    Proceed immediately to testing or freeze serum until required (≤ −16 °C) for several months at ≤ −16 °C or years at ≤−65 °C.

Box 2 Production of wild-type virus stocks for PRNT and MNA

Timing 5 d

Additional reagents and equipment required

  • 5 mm borosilicate glass beads (sterilized by autoclaving) (Z143944, cat. no. Sigma)

  • Vero/hSLAM cells (ECACC, cat. no. 04091501)

    Critical

    At the Docherty Institute, the virus (Victoria/1/2020) was isolated and grown in Vero/hSLAM cells. In our initial experiments, we found Vero E6 cells to be less satisfactory for growth (Vero E6 produce lower virus titers and are more likely to cause the furin-deletion mutation).

  • Geneticin solution, 50 mg/ml stock (Gibco, cat. no. 10131035)

Additional reagent setup required

  • Vero/hSLAM maintenance medium (10% FCS): prepare as for PRNT/MNA maintenance medium described in ‘Reagent setup’ but with the addition of 0.4 μg/ml of Geneticin

  • Vero/hSLAM infection medium (0% FCS): prepare as for Vero/hSLAM maintenance medium but omit the addition of fetal calf serum

  • Virus cultivation medium (4% FCS): prepare as for Vero/hSLAM maintenance media but add 20 ml of fetal calf serum instead of 50 ml to give a final serum concentration of 4% (vol/vol) FCS.

Procedure

  1. 1

    Day 0: seed Vero/hSLAM cells into sufficient T175 flask for desired batch size plus three additional T25 flasks. Seed flasks with 6.0 × 104 cells/cm2 in Vero/hSLAM maintenance medium and incubate at 37 °C.

  2. 2

    Day 1: using a microscope, check the cells are >90% confluent (i.e., cells are covering at least 90% of the surface area of the flask area). If they are not, leave the flasks for a further 4–6 h and repeat the check.

  3. 3

    Take one of the T25 flasks, remove the medium with a pipette and wash the cell monolayer twice with 5 ml/wash of PBS.

  4. 4

    Add 1 ml of trypsin-EDTA solution, and place the flask back into the incubator at 37 °C for ~10 min until the cells begin to visibly detach.

  5. 5

    Tap the flask firmly to detach the remaining cells, add 1 ml of Vero/hSLAM maintenance medium and thoroughly resuspend the cells.

    Caution

    It is important that the maintenance medium is added once the cells are detached from the flask to inactivate the trypsin.

  6. 6

    Mix 100 μl of the cells with 900 μl of trypan blue, apply ~10 μl to a counting chamber and count the cells. Calculate the number of cells per cm2 of flask area, and use this figure when calculating the multiplicity of infection (MOI).

  7. 7

    Remove the medium from the remaining flasks, and wash them twice with 10 ml/wash of PBS. To each T175 flask, add 3.5 ml of infection medium. Add 0.5 ml to one of the T25 flasks (labeled ‘CPE positive control’) and 1.0 ml to the other (labeled ‘CPE negative control’).

  8. 8

    Inspect the flasks thoroughly to ensure that there are no cracks or damage before taking them to the BSL3 laboratory.

    Caution

    All infectious manipulations from this point on need to be performed in a BSL3 laboratory within a MSCIII or suitable alternative such as a flexible film isolator by trained and competent operators.

  9. 9

    Calculate the dilution required to the stock virus to achieve an MOI of ~0.0005 based on the cell count/cm2 from Step 6 and the titer of the stock virus in plaque-forming units per ml (pfu/ml). In the MSCIII, dilute the virus and add it to the flasks according to the calculations. For example, if the TC175 contains 2.0 × 107 cells and the stock virus has a titer of 1.5 × 107 pfu/ml, then the virus needs to be diluted such that 1.0 × 104 pfu (2.0x107 × 0.0005) are added to the flask. In this example, 0.67 μl (1.0 × 104/1.5 × 107 = 0.00067 ml) is too small a volume to pipette, so a series of three serial 1/10 dilutions in infection medium would be performed and 670 μl added to each T175 flask. Accordingly, 96 μl (670 × (25/175)) of the same dilution would be added to the T25 flask containing 0.5 ml of infection medium (positive control). The remaining T25 flask is left as a negative control.

  10. 10

    Return the flasks to the incubator at 37 °C, and rock the flasks gently every 10–15 min for 1 h.

  11. 11

    Take the flasks back into the MSCIII, and add 30 ml of virus cultivation medium to each T175 and 5 ml to each T25. Decontaminate the surfaces of the flasks and return them to the incubator at 37 °C. Monitor the T25 flasks (infected and uninfected) daily by phase-contrast microscopy for the presence of cytopathic effect (CPE).

  12. 12

    Harvest the virus at ~72 h postinfection when CPE observations reveal 5–20% of the cells remain attached to the flask.

    Critical step

    It is important to harvest the virus before all the cells become detached because the viability of the virus begins to fall when this happens.

  13. 13

    Add ~10 ml of sterile glass beads to each T175, and gently rock to detach the remaining cells. Transfer the cell–virus suspension into centrifuge tubes, ensuring they are properly balanced. Seal the tubes into biological containment buckets, decontaminate and bring out to the room centrifuge.

  14. 14

    Centrifuge the cell–virus suspensions at 1,500g for 10 min at 10 °C, and return the sealed biosafe buckets to the MSCIII.

  15. 15

    Pool the clarified supernatants, and dispense into 1 ml volumes into prelabeled cryovials.

    Caution

    Ensure that the cryovials used are of a suitable design for storing HG 3 agents (i.e., are capable of withstanding freezing to −80 °C and have an O-ring sealed cap).

  16. 16

    Decontaminate the cryovials, bring them out of the MSCIII and place into an ultralow freezer at ≤−60 °C. These cryovials are single-use aliquots of virus for both the PRNT and MNA procedures. The virus is stable for at least 6 months when prepared and stored in this manner.

Box 3 Procedure for wild-type virus sighting experiments for PRNT

Procedure

Critical

This sighting experiment is only performed when using a new virus stock for the first time.

Caution

Steps 3–9 involve the use of live SARS-CoV-2 virus and must be performed within a class III safety cabinet within a BSL3 facility by trained and competent operators.

Rough sighting experiment

Timing 7 d

  1. 1

    Day 0: seed a 24-well tissue culture plate with 3 × 105 Vero E6 cells/ml (500 μl/well) in PRNT/MNA maintenance medium.

    Critical step

    Maintain cultures of Vero E6 cells such that they are in logarithmic growth (slightly subconfluent) when they are trypsinized, diluted and seeded into plates. Also ensure that the cells are homogeneous and well suspended.

    Critical step

    To ensure even distribution of the cells, rock the 24-well plates at least five times from front to back and side to side to ensure even distribution of the cells, and do not swirl, which can cause the cells to cluster in the middle of the well.

  2. 2

    Stack and seal the plates into a lid-lock plastic box containing several layers of paper towel that have been moistened with ~100 ml of sterile water (for a 5.5 L box). Incubate the plates at 37 °C for a minimum of 20 h before use in the assay.

  3. 3

    Day 1: in a class III safety cabinet in a BSL3 laboratory, thaw a single-use vial of virus using hand heat and make up tenfold serial dilutions of the virus stock from 10-1 to 10-5

    Critical step

    Check that the cells seeded on day 0 are ≥85% confluent and are forming an even monolayer before starting this step.

  4. 4

    Follow Steps 14-30 of Procedure 1, but add 100 µl of virus dilution in duplicate to cells in Step 18 rather than virus mixture from a neutralization plate. After plaque counting, do not perform neutralization dose calculation (Step 31); instead, continue from Step 5 below.

  5. 5

    Calculate the arithmetic mean of plaque counts, and select a dilution with an average between 40 and 100 plaques.

  6. 6

    Calculate the pfu/ml of the virus stock using the following equation:

    $${\mathrm{Pfu}}/{\mathrm{ml}} = {\mathrm{no}}{\mathrm{.}}\,{\mathrm{of}}\,{\mathrm{plaques}}/\left( {{\mathrm{Dilution}} \times {\mathrm{Infection}}\,{\mathrm{volume}}\left( {{\mathrm{ml}}} \right)} \right)$$

  7. 7

    To create a working stock of virus for PRNT, thaw a single-use vial of virus using hand heat and dilute to ~2 × 104 pfu/ml in PRNT/MNA diluent.

  8. 8

    Dispense into 1 ml volumes into prelabeled cryovials

    Caution

    Ensure that the cryovials used are of a suitable design for storing HG 3 agents (i.e., are capable of withstanding freezing to −80 °C and have an O-ring sealed cap with external thread).

    Critical step

    Ensure that vials of virus are stored in aliquots and frozen in a timely manner (ideally no longer than 90 min from thawing to refreezing) to minimize the drop in virus titers from being held at room temperature for too long.

  9. 9

    Decontaminate the cryovials, bring them out of the MSCIII and place them into an ultralow freezer at ≤−60 °C. These cryovials are single-use aliquots for PRNT assays, but will require a further fine-sighting experiment to determine the required working dilution (as described below).

Fine sighting experiment

Timing 7 d

Caution

Steps 14–16 involve the use of live SARS-CoV-2 virus and must be performed within a class III safety cabinet within a BSL3 facility by trained and competent operators.

  1. 10

    Day 0: seed three 24-well tissue culture plates with 3 × 105 Vero E6 cells/ml (500 μl/well) in PRNT/MNA maintenance medium.

    Critical step

    Maintain cultures of Vero E6 cells such that they are in logarithmic growth (slightly subconfluent) when they are trypsinized, diluted and seeded into plates. Also, ensure that the cells are homogeneous and well suspended.

    Critical step

    To ensure even distribution of the cells, rock the 24-well plates at least five times from front to back and side to side, and do not swirl, which can cause the cells to cluster in the middle of the well.

  2. 11

    Stack and seal the plates into a lid-lock plastic box containing several layers of paper towel that have been moistened with ~100 ml of sterile water (for a 5.5 L box). Incubate the plates at 37 °C for a minimum of 20 h before use in the assay.

  3. 12

    Day 1: add 75 µl/well PRNT/MNA assay diluent into columns 1–5 and 7–11 of a 96-well V-bottom plate. Into columns 6 and 12, add 150 µl PRNT/MNA assay diluent (negative, NVC); this is the mock neutralization plate.

    Critical step

    Check that the cells seeded on day 0 are ≥85% confluent and are forming an even monolayer before starting this step.

  4. 13

    Prepare 15 tubes with PRNT/MNA assay diluent according to Extended Data Fig. 5 to make a series of virus dilutions from 1/2 to 1/16.

  5. 14

    In a class III safety cabinet in a BSL3 laboratory, thaw a single-use vial of virus (as generated in Procedure 3, Stage I) using hand heat and add the appropriate amount of virus stock to each tube, as per Extended Data Fig. 5. Mix well using a vortex mixer.

  6. 15

    Transfer 75 µl of virus dilution in quadruplicate into the mock neutralization plate, columns 1–5 and 7–11.

  7. 16

    Stack and seal the plates into a lid-lock plastic box containing several layers of paper towel that have been moistened with ~100 ml of sterile water (for a 5.5 L box). Incubate the plates at 37 °C for 1 h (1.5 h max) to allow neutralization to occur.

    Caution

    Ensure that the sealed box is externally decontaminated before removing it from the safety cabinet or flexible film isolator.

  8. 17

    Follow Steps 14–30 of Procedure 1, but do not perform neutralization dose calculation after plaque counting (Step 31); instead, continue from Step 18 below.

  9. 18

    Calculate the arithmetic mean of the quadruplicate plaque counts for each dilution. The predicted dilution for the working stock should be the dilution at which ~70 pfu per well are counted.

    Critical step

    This predicted dilution is the working dilution for the virus stock and this sighting procedure will need to be repeated should a new working stock be produced.

Box 4 Wild-type virus semi-fine sighting and regression for the MNA

Timing 5 d

Procedure

Caution

Steps 4–8 involve the use of live SARS-CoV-2 virus and must be performed within a class III safety cabinet within a BSL3 facility by trained and competent operators.

  1. 1

    Day 0: seed 96-well tissue culture plates with 2.5 × 105 cells/ml (100 μl/well) in PRNT/MNA maintenance medium.

    Critical step

    Maintain cultures of Vero E6 cells such that they are in logarithmic growth (slightly subconfluent) when they are trypsinized, diluted and seeded into plates. Also ensure that the cells are homogeneous and well suspended.

  2. 2

    Stack and seal the plates into a lid-lock plastic box containing several layers of paper towel that have been moistened with ~100 ml of sterile water (for a 5.5 L box). Incubate the plates at 37 °C for a minimum of 20 h before use in the assay.

  3. 3

    Day 1: into columns 1–11 of a V-bottom plate, pipette 120 μl/well of PRNT/MNA assay diluent. Leave column 12 empty (negative, NVC); this is the virus dilution plate.

    Critical step

    Check that the cells seeded on day 0 are ≥85% confluent and are forming an even monolayer before starting this step.

  4. 4

    Into columns 1–11 of a second V-bottom plate, pipette 75 μl/well of PRNT/MNA assay diluent. Into column 12, pipette 150 μl/well of diluent (negative, NVC); this is the mock neutralization plate.

  5. 5

    In a class III safety cabinet in a BSL3 laboratory, add 60 μl of the virus stock to each well in column 1 of the virus dilution plate.

  6. 6

    Using a multichannel pipette set to 60 μl, perform a threefold dilution series of the virus from column 1 to column 11, mixing each dilution six to eight times before transferring to the next column.

  7. 7

    Set the multichannel pipette to 75 μl, and transfer the contents of the virus dilution plate into the mock neutralization plate (columns 1–11 only; column 12 is the NVC).

  8. 8

    Place the mock neutralization plate into the sealed, humidified box, decontaminate and incubate for 1 h at 37 °C.

  9. 9

    Follow Steps 14–36 of Procedure 2, but do not perform neutralization dose determination after foci counting; instead, continue from Step 10 below

  10. 10

    Input the foci count data (eight replicates per dilution) against the reciprocal of the dilution (i.e., 3, 9, 27, 81 and so on) into a statistical software package such as MiniTab. Where any of the counts at a dilution are >450/well, exclude/remove those data before proceeding with the regression analysis.

  11. 11

    Log10-transform both the counts and the dilution factors, and perform a linear regression.

  12. 12

    Rearrange the equation for the regression to predict the dilution factor required to achieve a virus focus count between 125 and 200 per well.

    Critical step

    This predicted dilution is the working dilution for the virus stock used, and this procedure will need to be repeated should a new working stock be produced.

Box 5 Creation and production of pseudotyped virus for PNA

Timing 2 d

Procedure

  1. 1

    Day 0: seed ES-293 cells at 0.8 × 106 cells/ml in ESF-SFM medium and incubate in an orbital shaker incubator overnight at 37 °C/5% CO2.

  2. 2

    Day 1: dilute 1250 µg of plasmid DNA containing the spike Δ CT from SARS-CoV-2 Wuhan strain into OptiPro SFM to a total of 20 ml and mix gently.

  3. 3

    In a separate tube, dilute 1250 µl of FreeStyle Max transfection reagent in OptiPro SFM to a 20 mL and mix gently.

  4. 4

    Immediately add the diluted plasmid and transfection reagent together to reach a final volume of 40 ml. Incubate the mixture for 10 min at room temperature.

  5. 5

    Slowly add the DNA/transfection reagent mixture to the cells, and incubate in an orbital shaker (48 h at 37 °C/5% CO2).

  6. 6

    Determine the cell concentration and adjust the concentration to 5.0 × 106 cells/ml.

  7. 7

    Add the pseudotyped ΔG-luciferase (G*ΔG-luciferase) rVSV at an MOI = 3. Incubate for 2 h without shaking.

  8. 8

    Centrifuge at 300g for 5 min at room temperature, and resuspend the pellet in prewarmed ESF-SFM medium.

  9. 9

    Incubate in an orbital shaker incubator for 24 h at 37 °C/5% CO2.

  10. 10

    Collect the culture and centrifuge at 1,000g for 5 min at room temperature.

  11. 11

    Transfer the supernatant on a 0.2 µm filter funnel.

  12. 12

    Store at −80 °C in 1 mL aliquot (single-use aliquot).

    Critical step

    For each batch of pseudotyped virus, a titration on Vero E6 cells is performed using an anti-luciferase antibody to enable detection of foci-forming units instead of relative light units (RLUs).

    Critical step

    A neutralization assay with and without anti-VSV antibody is used to demonstrate the absence of any residual nonpseudotyped ΔG-luciferase (G*ΔG-luciferase) rVSV. If there is no residual nonpseudotyped virus, both titration curves (with and without anti-VSV antibody) will overlap.

Box 6 Pseudotyped virus dose sighting

Timing 2 d

Procedure

Critical

This stage is performed on each new lot of pseudotyped virus to determine the amount of SARS-CoV-2 pseudotyped virus particles to be used during the PNA.

  1. 1

    Day 0: seed 96-well plates with 2.0 × 105 cells/ml (100 μl/well) of Vero E6 cells in PNA maintenance medium, and incubate overnight at 37 °C/5% CO2.

  2. 2

    Day 1: ensure that the cells seeded on day 0 are >80% confluent.

  3. 3

    Perform a double-dilution series of the pseudotyped virus in duplicate, and assess in the PNA without the addition of neutralizing serum—in effect, performing a mock assay by following Steps 8–14 of Procedure 3.

  4. 4

    Perform a linear regression of mean RLUs against fold dilution of virus for the two dilutions flanking the 150,000 RLUs.

  5. 5

    Interpolate from the regression the dilution required to achieve 150,000 RLUs.

    Critical step

    Based on the titer of the batch, the dilution to be used for the PNA is adjusted such that a target of 150,000 RLUs/well is used. The standardized pseudotyped virus input makes the PNA stable and robust over time.

Experimental design

To perform the procedures described in this protocol, commercial individual and/or pooled human serum containing SARS-CoV-2 neutralizing antibodies are required. These have been available from NIBSC since December 202016. To perform the PRNT and MNA assays, implementers will need an appropriate wild-type SARS-CoV-2 strain or isolate and an operational Advisory Committee on Dangerous Pathogens containment level 3 laboratory (BSL3). To perform the PNA, implementers will need to obtain stocks of the pseudotyped virus or have the necessary expertise to generate these stocks.

For all procedures described here, experience in working with cell culture would be a distinct advantage. For all manipulations involving wild-type SARS-CoV-2 in the PRNT and MNA, operators must be competent in performing standard virological techniques at BSL3. For the PNA, implementers would need the molecular biological expertise to generate the pseudotyped virus stocks if they cannot obtain them commercially.