Development of an automated, high-throughput SARS-CoV-2 neutralization assay based on a pseudotyped virus using a vesicular stomatitis virus (VSV) vector

Abstract

ABSTRACTThe global outbreak of COVID-19 has caused a severe threat to human health; therefore, simple, high-throughput neutralization assays are desirable for developing vaccines and drugs against COVID-19. In this study, a high-titre SARS-CoV-2 pseudovirus was successfully packaged by truncating the C-terminus of the SARS-CoV-2 spike protein by 21 amino acids and infecting 293 T cells that had been stably transfected with the angiotensin-converting enzyme 2 (ACE2) receptor and furin (named AF cells), to establish a simple, high-throughput, and automated 384-well plate neutralization assay. The method was optimized for cell amount, virus inoculation, incubation time, and detection time. The automated assay showed good sensitivity, accuracy, reproducibility, Z' factor, and a good correlation with the live virus neutralization assay. The high-throughput approach would make it available for the SARS-CoV-2 neutralization test in large-scale clinical trials and seroepidemiological surveys which would aid the accelerated vaccine development and evaluation.

Keywords: SARS-CoV-2; VSV; high-throughput; neutralization assay; pseudovirus.

Figure 1.

(A) Schematic diagram of D614G and D614G with a C-terminal truncation of 21 amino acids. The numbers in the figure represent the positions of amino acids; (B) SARS-CoV-2-Sdel21-GFP pseudovirus packaging flow chart. The figure is drawn using biorender. (https://app.biorender.com/).

Figure 2.

Optimization of the S protein plasmids and selection of sensitive cells. (A) The investigation of the effects of truncation of amino acids from 18 to 24 at the C terminus of the spike plasmid of D614G on the pseudovirus titres; (B) Transfection of furin, TMPRSS2, cathepsin L and pcDNA3.1 using transient transfection of ACE2 in 293 T cells. The above plasmids were transiently transfected for a single time according to 10 µg of plasmid in each T25 bottle (if two plasmids were transfected at the same time, the total plasmid amount was 10 µg). Cells were harvested 24 h later for a cell sensitivity test. The number of GFP-positive cells after virus infection in different cells was counted with the Biotek instrument; (C) Comparison of the infection efficiency of SARS-CoV-2-D614Gdel21-GFP in 293 T cells that were transiently transfected with ACE2 and Furin (*293T-ACE2-Furin), 293 T cells which were stably transfected ACE2 and Furin (293T-ACE2-Furin), named AF, 293T-ACE2, and VeroE6 cells. All data are obtained through three independent repeated experiments. The bar value represents the mean with SD. Cells with * are represented by transiently transfected cells.

Figure 3.

Optimization of cell number. (A) Optimization of the number of cells added in the pseudovirus titration assay; (B-D) To optimize the seeding cell number, 1 × 10³–1.6 × 10⁴ cells/well were added and the cells were infected with an inoculant dose of 0.2 MOI, and the NT₅₀ values of Sample 1, Sample2 and Sample 3 were calculated with non-linear regression, i.e. log(inhibitor) vs. response (four parameters). All data are obtained through three independent repeated experiments.

Figure 4.

Optimization of pseudovirus inoculated dose. (A-C) The optimal inoculant dose of the pseudovirus was investigated using doses from 0.025–0.8 MOI. The amount of cell addition were 2.00 × 10³ cells/well, and the NT₅₀ values were calculated with non-linear regression, i.e. log(inhibitor) vs. response (four parameters). All data are obtained through three independent repeated experiments.

Figure 5.

Optimization of pre-incubation and incubation time for pseudovirus neutralization assay. (A) The effect of different pre-incubation times on the NT50 values in the SARS-CoV-2-D614Gdel21-GFP fluorescent pseudovirus neutralization assay; (B) The number of GFP-positive cells detected at 0, 8, 12, 24, 36, 48, and 72 h were compared in the pseudovirus titration test; (C) Comparison of the NT50 values of Sample 1, Sample 2, and Sample 3 at the end of different incubation times. All data are obtained through three independent repeated experiments. The bar value represents the mean with SD.

Figure 6.

Establishment of the automatic fluorescent pseudovirus neutralization test based on the VSV system in a 384-well plate. This figure is drawn using biorender.

Figure 7.

Methodological validation of the fluorescent pseudovirus 384-well plate neutralization assay. (A) Sensitivity of the pseudovirus assay; (B) Specificity of the pseudovirus assay; (C) Reproducibility of the pseudovirus assay; (D) The number of GFP-positive cells was obtained by analyzing 100 positive wells and 100 negative wells using the 384-well plate fluorescent pseudovirus neutralization assay. All data are obtained through three independent repeated experiments. The bar value represents the mean with SD.

Figure 8.

Comparison of the 384-well plate GFP pseudovirus assay, 96-well plate Fluc and live virus assay. (A) NT₅₀ values were obtained by analyzing 60 guinea pig serum samples using the two neutralization test methods (384-well plate fluorescent pseudovirus neutralization assay and 96-well plate Fluc pseudovirus neutralization assay) after calibration with international standards for consistency comparison; (B) Correlation between the 384-well plate and the 96-well plate neutralization assay for comparison of all the 60 samples. The qualitative comparison of the results from the two assays showed concordance of 49 positive and 10 negative samples (98.0%, Kappa = 0.942). The fitted regression line is presented by the equation y = 0.3986x + 12.988, R² = 0.9903; (C) Agreement between the 384-well plate and the 96-well plate neutralization assay for 50 positive samples and 10 negative samples analyzed by Bland-Altman plot; (D) Correlation analysis of EC₅₀ values between the 384-well plate pseudovirus neutralization assays and authentic neutralization assay. The fitted regression line is presented by the equation y = 2.9559x - 0.0874, R² = 0.84.