Development of pseudotyped VSV-SARS-CoV-2 spike variants for the assessment of neutralizing antibodies

Abstract

Aim: Serological studies with pseudotyped viruses offer a safer alternative to live SARS-CoV-2 in evaluating neutralizing antibodies, enabling research in standard labs.

Methods: The SARS-CoV-2 Spike pseudotyped vesicular stomatitis virus (VSV) pseudoviruses were generated using Spike of Wuhan strain and two variants (B.1.1.7, B.1.351) and utilized to evaluate the serum neutralizing activity of human plasma samples of vaccinated (n = 13) and healthy people (n = 2) compared with a plaque assay with authentic virus.

Results: Neutralizing titer of convalescent plasma resulted with a good correlation (R² = 0.7).

Conclusion: We evaluated a safe and reliable pseudotyped virus system that effectively mimics authentic virus and correlates well with traditional assays. The developed system allows easier testing of variants and has the potential to improve vaccine development.

Plain language summary

Article highlights.

• Pseudotyped VSV-SARS-CoV-2 Spike can successfully measure neutralizing plasma activity of vaccinated individuals compared with plaque assays with authentic virus.

• The method of pseudovirus reported in the study can distinguish the neutralizing abilities of convalescent plasma against Spike variants such as Alpha (B.1.1.7), Beta (B.1.351) and Wuhan types.

• VSV-based pseudotyped virus system demonstrates greater sensitivity, specificity and correlation with plaque assay, crucial for assessing vaccine responses.

• BNT162b2-vaccinated individuals have exhibited higher neutralizing activity against authentic virus and pseudovirus than CoronaVac-vaccinated.

1. Introduction

Serum neutralization refers to the ability of antibodies present in the serum to neutralize the infectivity of viruses as a result of natural infection or vaccination. However, high levels of mutation observed in viral glycoprotein of different variants could prevent effective protection provided by neutralization serum. Therefore, immediate determination of serum neutralization potency against different variants is crucial [1]. Particularly, new variants have begun to emerge for disease caused by global infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (family Coronaviridae, genus Betacoronavirus, species Severe acute respiratory syndrome-related coronavirus) since 2019. It has been reported that mutations on the major neutralization target of these variants alter serum-neutralizing activity caused by early strains or vaccines [2,3]. Some of these mutations can provide an escape from immune response by decreasing monoclonal antibody (mAb) neutralization [4].

Authentic viruses involve high risk for research studies, so it is essential to use laboratories that meet at least biosafety level-3 (BSL-3) standards with negative pressure systems leading to the limitation in use of biosafety level-2 (BSL-2) laboratories by research centers. Pseudotyped virus neutralization assay is an alternative method based on packaging a reliable replication-defective pseudovirus with Spike protein to mimic the entry of the authentic virus, allowing BSL-2 conditions to perform neutralization assays [5,6]. Several convenient pseudotyped virus for SARS-CoV-2 have been reported, such as human immunodeficiency virus (HIV) (family, Retroviridae, genus Lentivirus, species Human immunodeficiency virus)-based lentiviral particles [7], murine leukemia virus (MLV) (family Retroviridae, genus Gammaretrovirus, species Murine leukemia virus)-based retroviral particles [5], or vesicular stomatitis virus (VSV) (family Rhabdoviridae, genus Vesiculovirus, species Indiana vesiculovirus)-based systems [8,9] with enhanced capability for assessing the efficacy of therapeutic drugs and vaccines [9]. These pseudovirus neutralization assays conclude very closely correlated to those of authentic virus measurements. Also, tracking the changes in Spike glycoprotein using pseudovirus neutralizing assays is relatively easy [10].

From the beginning of the pandemic, RNA sequencing data of virus glycoproteins was utilized to identify individual mutations in SARS-CoV-2 [11]. One of the first variants called B.1.1.7 was first emerged in the United Kingdom and had multiple mutations in the target regions of neutralizing antibodies such as receptor binding domain (RBD) and N-terminal domain of Spike. Subsequently B.1.351 variant has been identified in South Africa with additional mutations. B.1.1.7 and B.1.351 variants share key mutations in the RBD (N501Y, D614G), but B.1.351 has additional changes causing widespread escape from mAbs (E484K, K417N) [12,13].

In this paper, we evaluated human convalescent plasma with different serum neutralizing activities using pseudotyped VSV-ΔG virus carrying Spike variants (Wuhan strain, B.1.1.7 and B.1.351). Neutralization activities of human convalescent plasma samples (n = 15) were selected in a manner ensuring homogenous distribution for reliable evaluation of the pseudovirus system. First, we made a series of point mutations in spike sequence of SARS-CoV-2 to obtain global variants. Second, we generated pseudotyped viruses using Spike of ancestral Wuhan strain and two variants and utilized to evaluate neutralization activity of human serum samples. Neutralization assay using Spike pseudotyped VSV-ΔG virus was found to correlate with plaque assay using SARS-CoV-2 authentic virus. Additionally, SARS-CoV-2 Spike pseudotyped VSV-ΔG virus was sufficient to discriminate serum responses against different variants of the virus and to test vaccine response in individuals.

2. Materials & methods

2.1. Plasmid construction

The expression vector for SARS-CoV-2 Spike (pTWIST-EF1-alpha-SARS-CoV-2-S-2xStrep vector) was kindly provided by Dr. Nevan Krogan [14]. The spike gene was codon optimized to improve the expression efficiency in mammalian cells (Genbank: MN985325). A pMD2.G lentiviral plasmid containing spike gene was generated (Supplementary Figure S1). Briefly, forward primer 5′-ATAGAATTCGCCGCCACCATG-3′ and reverse primer 5′-ATAGAATTCTCATCAACTACCGCAAGAACAACAACC-3′ were used to amplify mutant Spike protein with 21 amino acid deletion in C-terminal. To clone spike gene successfully, TA cloning was applied. In the first step, PCR reaction was set up to add adenine bases to 3′OH terminal of spike gene. The amplified product was ligated into pGEM-T easy vector system. In the second step, spike gene was restricted by EcoRI and ligated into pMD2.G vector. spike gene with C-terminal 21 amino acid deletion (pmD2.SpikeCdel21) was confirmed with Sanger DNA sequencing.

B.1.1.7 (Alpha) variant including ΔH69-V70, N501Y, D614G, ΔY144 mutations and B.1.351 (Beta) variant including E484K, N501Y, D614G, K417N mutations of SARS-CoV-2 were constructed in pMD2.SpikeCdel21 plasmid by using site-directed mutagenesis with specifically designed primers (Supplementary Table S1). PCR conditions were set up by following procedures of Phusion High-Fidelity Polymerase (Thermo Sci., F530L) and Quick-change Site-directed Mutagenesis Kit (Stratagene, Agilent, 200518). Next, the template was digested with DpnI to eliminate parental DNA. The constructed vector was transferred into Escherichia coli XL-1 Blue competent cells. Following the picking up colonies, plasmids were sequenced and mutations were confirmed. Sequences of mutation primers and sequences of plasmid constructs are given in the Supplementary Information File.

2.2. Cell culture

HEK (Human Embryonic Kidney) 293T cells (ATCC, CRL-321) and Vero E6 (Monkey African Green Kidney) (ATCC, CRL-1586) cells were cultured in Dulbecco's modified Eagle medium (DMEM, PAN Biotech, P04–3500) supplemented with 10% fetal bovine serum (PAN Biotech, FBS standard, P30–3306) and %1 L-glutamine (200 mM; PAN Biotech, P04–80100) and 1% penicillin/streptomycin (Pan Biotech, P06–07100) at 5% (v/v) CO₂, 37°C incubator.

2.3. Plasma samples

Human convalescent plasma samples were obtained from volunteer donors after 2 months post-vaccination (Table 1). In order to ensure ethical standards, written consent was obtained from all participants involved in this study. All participants were thoroughly screened for any acute febrile illnesses, infections, or underlying medical conditions that could potentially interfere with the novel test. Specifically, none of the participants had a history of autoimmune diseases or infections that are known to cause cross-reactivity in immunological assays. This precaution was taken to ensure the specificity and reliability of the test results. Plasma samples were heat-inactivated at 56°C for 30 min before using for neutralization assays.

Table 1.

Vaccination status of human plasma samples used for neutralization assays.

Plasma samples	Vaccination
1	CoronaVac, 2 dose
2	CoronaVac, 2 dose
3	CoronaVac, 2 dose
4	CoronaVac, 2 dose; BNT162b2, 1 dose
5	CoronaVac, 2 dose; BNT162b2, 1 dose
6	CoronaVac, 2 dose; BNT162b2, 1 dose
7	CoronaVac, 2 dose
8	CoronaVac, 2 dose; BNT162b2, 1 dose
9	BNT162b2, 2 dose
10	BNT162b2, 2 dose
11	BNT162b2, 2 dose
12	CoronaVac, 2 dose
13	CoronaVac, 2 dose
14	Healthy control
15	Healthy control

2.4. Plaque reduction neutralization assay

Serum samples (non-denatured) of vaccinated (n = 13) and healthy people (n = 2) were diluted in DMEM from 1:10 to 1:320. SARS-CoV-2 virus was diluted in DMEM to a final concentration of 10^-3 pfu/ml. 300 μl serum and 300 μl diluted virus were mixed and incubated in 37°C incubator at 5% (v/v) CO₂ for 1 h. Then, the serum-virus mixture was added to Vero E6 cells and incubated in a 37°C incubator at 5% (v/v) CO₂ for 1 h. Serum-virus mixture was discarded from the plate and the plate was covered with 2% methylcellulose and DMEM at 1:2 ratio and incubated for 4 days. Methylcellulose was discarded and cells are washed with 1xPBS. Then, 4% PFA was used for cell fixation at room temperature for 20 min. After fixation, cells were stained with crystal violet solution for 30 min at room temperature. Plaques were counted after washing with PBS to remove the excessive stain. The number of plaques were used to calculate virus neutralization titers.

2.5. Generation of SARS-CoV-2 pseudotyped variants

rVSV-ΔG-G* (G*: VSV glycoprotein) and VSV-ΔG-Spike (carrying Spike protein of coronavirus instead of VSV glycoprotein) virus were produced according to the protocol provided by Whitt in 2010 [15]. BHK-21/WI-2 cells (Baby Hamster Kidney-21/ Clone WI-2, Kerafast, EH1011) were infected with vaccinia virus (ATCC, strain vTF7-3, VR-2153™) for 1 h, followed by co-transfection of five plasmids: VSV-ΔG, together with VSV accessory plasmids encoding for VSV-N, P, L and G proteins (Kerafast), all of which were under T7 promoter control. The primary transfection was performed using Lipofectamine 2000. BHK-21/WI-2 cells were transfected with Spike plasmid to assist in creating passage 1 (P1). Forty-eighth following primary transfection, supernatant containing the recovered VSV-Spike was collected, centrifuged at 1300 × g × 5 min to remove cell debris. Total supernatant was collected, centrifuged and used for sequential passaging in Vero E6 cells to eliminate parental virus [16], as shown in Figure 1. VSV-ΔG-Spike was propagated in DMEM containing 5% FBS (reduced serum media), MEM NEAA, 2 mM L-glutamine and 1% P/S.

Figure 1.

Generation of VSV pseudotyped viruses bearing truncated SARS-CoV-2 Spike proteins. (A) Preparation of rVSV-ΔG-G* in BHK21/W1 cell line. (B) Packaging of rVSV-ΔG-Sdel21 in HEK293T cell line. (C) Infection of pseudovirus carrying GFP reporter and sequential passaging of pseudovirus to eliminate parental virus (rVSV-ΔG-G*) using plasmids encoding Sdel21 and T7 polymerase transfected Vero E6. (D) SARS-CoV-2 Spike pseudovirus neutralization assay using human convalescent plasma samples and GFP signal reading.

rVSV-ΔG: Glycoprotein (G protein) deleted recombinant VSV; SARS-CoV-2: Severe acute respiratory syndrome coronavirus 2; VSV: Vesicular stomatitis virus.

2.6. SARS-CoV-2 pseudovirus infection for viral entry & titration

To determine viral titer, Vero E6 cells were seeded in a 24-well-plate before the day of infection. 100 μl VSV-ΔG-Spike pseudotyped virus was diluted threefold serial dilutions up to 1:243 using DMEM 5% FBS as diluent. After 24 h, GFP positive cells were counted in the last wells to calculate the TU/ml.

For viral entry assay, medium was replaced with 200 μl cell culture medium including hydroxychloroquine (HCQ) (20–50 μM) as a SARS-CoV-2 blocker for 20 min. Then, treated wells were inoculated with SARS-CoV-2 Spike protein pseudotyped VSV-ΔG (VSV-ΔG-Sdel21). 48 h of post infection, the plate was scanned with Axio Observer Fluorescence Microscope (Zeiss).

2.7. Immunocytochemistry

Indirect immunofluorescence analysis was performed for cellular Spike protein and ACE-2 protein detection. Vero E6 cells which are infected with Spike pseudotyped VSV-ΔG virus were fixed with %4 Paraformaldehyde (PFA) and stained with anti-Spike antibody (Sino Biological 2019-nCoV Spike S1 Rabbit mAb, 40150-R007) (1:80 dilution), then incubated with 1:1000 diluted Alexa Fluor 568-conjugated goat anti-rabbit IgG secondary antibody. ACE-2 protein was detected with anti-hAce2 (R&D Human Ace2, AF933) (1:120 dilution), incubated with Alexa Fluor 568-conjugated anti-goat IgG secondary antibody (1:500 dilution).

2.8. Western Blot

Western blot (WB) analysis was performed to validate SARS-CoV-2 Spike protein expression in HEK293T cells. Total protein was subjected to SDS-PAGE followed by immunoblotting to nitrocellulose membranes (250 mA for 1.15 h). Rabbit SARS-CoV-2 (2019-nCoV) Spike S1 monoclonal antibody (40150-R007, Sino Biological) and a horseradish peroxidase-conjugated secondary antibody (1/10,000) were used for detection. Antibodies were used at 1/500 for Spike S1 subunit and at 1/10,000 for β-actin.

2.9. Pseudovirus based neutralization assay

Serum samples of vaccinated (n = 13) and healthy people (n = 2) were diluted in DMEM-5 (reduced serum media; 5% FBS, 1% L-glutamine, 1x MEM NEAA) from 1:30 to 1:2430. SARS-CoV-2 Spike pseudotyped VSV-ΔG virus was diluted in DMEM-5 to a final concentration of 10^-4 TU/ml. 100 μl diluted serum and 50 μl diluted virus were mixed in a 96 well plate and incubated in 37°C incubator for 1 h. Then, serum-virus mixture was added to Vero E6 adherent cells (80% confluent) and incubated in a 5% (v/v) CO₂, 37°C incubator for 2 days. Cells were fixed with 4% PFA and analyzed by using Tecan plate reader (Infinite M200 Pro, Life Science). The GFP fluorescence intensity was microscopically detected at 48 h post-infection by Axio Observer Fluorescence Microscope (Zeiss).

2.10. Statistical analysis

Data were analyzed using GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA) and expressed as mean ± standard deviation (SD). A one-way ANOVA test along with Bonferroni's post hoc test was utilized among study groups to compare the effect of CQ on viral entry. A p-value less than 0.05 is considered statistically significant.

3. Results

G-protein deficient VSV bearing C-terminal 21 aminoacid truncated Spike protein were packaged in HEK293T cell line (Figure 1). To fully eliminate ancestral VSV-G, Spike pseudoviruses were serially passaged in pmD2.SpikeCdel21 transfected Vero E6 cells.

Hydroxychloroquine effectively inhibits viral entry and has been proposed for prophylaxis/treatment of SARS-CoV-2 infection. Therefore, HCQ was used to display inhibition of pseudovirus entry. VSV-ΔG-Sdel21 pseudotyped virus infectivity was decreased with increasing concentration of hydroxychloroquine (20–50 μM) based on percentage of GFP-positive cell (Figure 2A & B). The number of GFP-positive cells in VSV-ΔG-Sdel21-infected Vero E6 cells were significantly reduced from 30% to 10% in highest concentration of HCQ compared with positive control (p < 0.0001). Representative image of positive control was given in Supplementary Figure S2.

Figure 2.

Viral entry assay in Vero E6 cell line. (A) Vero-E6 cells were inoculated with VSV-ΔG-Sdel21 pseudotyped virus. Cells were treated with hydroxychloroquine at 20 min before inoculation. After 24 h, GFP signal detected under fluorescence microscope is shown in representative cellular images. (B) For quantification, GFP signal from images per sample was analyzed by using Image J at 24 h of post-inoculation. Values are presented as % of total population. Scale bar: 100 μm. Data are represented as means of ± S.D. of at least three independent experiments. *p < 0.05; ****p < 0.0001 - using One-Way ANOVA with Bonferroni's post-hoc test.

GFP: Green fluorescent protein; VSV: Vesicular stomatitis virus.

To validate Spike and ACE2 protein expression of infected cells, immunostaining was performed using VSV-ΔG-Sdel21 pseudotyped virus. Co-localization was not found in mock-treated hACE2 (Figure 3A-D), indicating that SARS-CoV-2 uses human ACE2 as a receptor for entry. Meanwhile, colocalization of VSV-ΔG-Sdel21 and human ACE2 receptor was investigated (Figure 3E-H). In addition, colocalization of VSV-ΔG-Sdel21 and Spike protein confirmed the existence of Spike as a surface glycoprotein in our VSV-ΔG-Sdel21 pseudotyped particles (Figure 3I-L).

Figure 3.

Colocalization of pseudovirus with hACE-2 receptor or SARS-CoV-2 Spike protein in Vero E6 cell line. Cells were incubated with anti-SARS-CoV-2 Spike, human anti-ACE2 and DAPI. (A–D) Mock-treated VeroE6 cells with hACE2 staining. (E–H) VSV-ΔG-Sdel21 infected Vero E6 cells with hACE2 staining. (I–L) VSV-ΔG-Sdel21 infected Vero E6 cells with SARS-CoV-2 Spike staining. Yellow color shows viral S protein (J) and red colors represent human ACE2 (b, f). Scale bars, 20 μm. The data in a-l are representative images of three independent experiments.

SARS-CoV-2: Severe acute respiratory syndrome coronavirus 2; VSV: Vesicular stomatitis virus.

The convalescent plasma samples were analyzed with VSV-ΔG bearing truncated SARS-CoV-2 Spike protein (VSV-ΔG-Sdel21). Administration of neutralizing antibodies against Spike pseudotyped VSV-ΔG virus reduce the number of GFP-positive cells. Neutralization activity of plasma was calculated based on GFP signal. Neutralization activities of plasma samples were depicted in Figure 4. With this assay, samples 2, 3, 4, 6, 7, 8, 9, 10 and 11 showed high neutralization activity with 50% inhibition above 1:270 dilutions.

Figure 4.

In vitro pseudovirus neutralization activity of human plasma samples (1–13) against SARS-CoV-2 Spike variants; Wuhan (A), Alpha (B), Beta (C). Neutralization curves of human convalescent plasma samples against pseudotyped SARS-CoV-2 variants; (A) VSV-ΔG-Sdel21 Wuhan type, (B) VSV-ΔG-Sdel21 Alpha variant carrying ΔH69-V70, N501Y, D614G, ΔY144 and (C) VSV-ΔG-Sdel21 Beta variant carrying E484K, N501Y, D614G, K417N mutations were measured as GFP signal in Vero E6 cells, and the percentage of neutralization was calculated. Data are represented as means of ± S.D. of at least three independent experiments. Human samples; 1 (blue circle), 2 (red circle), 3 (green circle), 4 (purple circle), 5 (orange circle), 6 (black circle), 7 (brown circle), 8 (dark blue circle), 9 (dark purple circle), 10 (dark red circle), 11 (dark green circle), 12 (dark gold circle), 13 (light green circle).

GFP: Green fluorescent protein; SARS-CoV-2: Severe acute respiratory syndrome coronavirus 2; VSV: Vesicular stomatitis virus.

To confirm data obtained by pseudovirus neutralization assay, we performed a Plaque Reduction Assay (PRA) against the SARS-CoV-2 Wuhan variant by using the same convalescent plasma samples. This assay relies on quantifying the titer of neutralizing antibody through the number of plaques forming units (pfu) generated in a monolayer of virus-infected cells. Isolated SARS-CoV-2 was used to infect Vero E6 cells in the presence of decreasing concentrations of convalescent plasma and the 50% plaque reduction titer was determined. Several samples (# 7, 8, 9 and 11) showed high neutralization activity over a PRA₅₀ of 1:320 (Table 2).

Table 2.

PRA titers and vaccination profile of human plasma samples.

Plasma samples	Vaccination	Neutralization titer by plaque assay
1	CoronaVac, 2 dose	1:10
2	CoronaVac, 2 dose	1:20
3	CoronaVac, 2 dose	1:20
4	CoronaVac, 2 dose; BNT162b2, 1 dose	1:80
5	CoronaVac, 2 dose; BNT162b2, 1 dose	1:80
6	CoronaVac, 2 dose; BNT162b2, 1 dose	1:80
7	CoronaVac, 2 dose	1:320
8	CoronaVac, 2 dose; BNT162b2, 1 dose	1:1280
9	BNT162b2, 2 dose	1:640
10	BNT162b2, 2 dose	1:160
11	BNT162b2, 2 dose	1:320
12	CoronaVac, 2 dose	1:40
13	CoronaVac, 2 dose	1:10
14	Healthy control	0
15	Healthy control	0

The neutralization assay results were plotted against each other and showed linearity with a measure of certainty of R² = 0.7 (p = 0.0006) for pseudotype against the plaque reduction assay with life virus, as shown in Figure 5A. The measure of certainty (R²) was decreasing for pseudotype bearing Spike protein of Alpha and Beta variants against the plaque reduction assay with Wuhan strain (Figure 5B & C). Linearity test of pseudovirus bearing Spike protein of Alpha variant was less significant for the comparison of the pseudovirus bearing Spike protein of Wuhan strain versus the PRNT₅₀ (R² = 0.31, p = 0.048).

Figure 5.

Comparison of pseudovirus neutralization assays with the plaque reduction assay. Pseudotyped virus neutralization assays of Wuhan (A), Alpha (B) and Beta variants (C) was plotted against the virus neutralization activities determined by plaque reduction assay (PRNT₅₀) with authentic Wuhan virus. The R² value indicates the certainty of the values to be at the trend line and shows with R² = 0.7 comparability of the two assays.

4. Discussion

Viral epidemics can easily spread around the world as recently witnessed for COVID-19 epidemic. Efficient viral tools are necessary for developing vaccines or therapeutic drugs; however, handling infectious SARS-CoV-2 requires BSL-3 facilities, which is a significant limitation in research and development applications for therapeutic purposes. The pseudovirus system is a promising approach to evaluate viral inhibitors and neutralizing antibodies against SARS-CoV-2, in convalescent plasma [5]. Although, Lentivirus and VSV are common vectors used in pseudovirus construction, VSV-ΔG system is more preferable due to the higher virus titers compared with retrovirus systems [6].

VSV is a glycoprotein (G)-enveloped negative-stranded RNA virus that infects a wide range of animals and rarely humans causing mild symptoms of flu. VSV can effectively integrate and express irrelevant transmembrane proteins onto the surfaces of recombinant virus particles. In addition, their easily modifiable small genome (11 kb) and abundant replication in a wide range of cell lines have favored the use of VSV in pseudovirus systems [17]. On the other hand, the shape of VSV used in neutralization assays might not reflect the density and distribution of Spike protein, because the viral surface geometry of the authentic spherical SARS-CoV-2 virus is different from bullet shape VSV [5].

Besides, there may be residual VSV-ΔG-G* virus interfere with the pseudovirus due to its packaging process in the generation of pseudovirus. Parental VSV-G may cause additional infection apart from VSV-ΔG-Spike pseudovirus and lead to false-positive results [6]. To solve this issue, serial passaging during pseudovirus generation and antibodies against G protein could be utilized [16,18]. Yahalom-Ronen et al. [16] showed that at least 5 passages of recombinant VSV-ΔG-Spike virus ensure elimination of residual VSV-G. They also reported that sequential passage of pseudotyped virus results in increased prevalence of Spike protein structures per single particle. Therefore, in wet-laboratory processes, we applied 6 serial passaging of VSV-ΔG-Spike pseudovirus with Spike protein in Vero E6 cells to remove virions bearing G protein.

In advance of pseudovirus neutralization assay, we showed SARS-CoV-2 Spike based pseudovirus entry by using HCQ, a blocker for SARS-CoV-2 [19]. Increasing concentration of HCQ led to a decrease in GFP signal in Vero E6 cells. There is a limitation to use GFP reporter because several studies have shown that fluorescent proteins, such as GFP or DsRed, exhibit slower kinetics and reduced sensitivity compared with Firefly luciferases [20]. Notwithstanding, due to robust expression in the cell and ease to follow, we followed intensity of GFP fluorescence to determine VSV titer and inhibition.

The constructed model consisting of Vero E6 cells and VSV-ΔG-Spike can replicate the entry of authentic SARS-CoV-2. The serum neutralizing titer of vaccinated convalescent patients measured by the VSV-ΔG-Spike pseudovirus assay, which is safer and rapid, shows a strong correlation with SARS-CoV-2 PRA₅₀ assay. In our pseudovirus study, Alpha and Beta variants of Spike protein were constructed and tested against human convalescent plasma samples listed in Table 1. It is shown that the pseudovirus assay based on the VSV-ΔG system was successful in discrimination of different SARS-CoV-2 variants. In consistency with literature, pseudovirus neutralization assay exaggerated the potency of samples with low neutrality activity, but increased the inhibition limit, enabling the identification of samples with high neutrality potency [21]. Each vaccinated human convalescent serum sample was assayed for neutralization against Alpha (B.1.1.7), Beta (B.1.351) and Wuhan (Wild type) viruses. Overall, the neutralizing activity against Alpha Spike pseudotyped virus remained largely unchanged but was significantly reduced against Beta Spike pseudotyped virus. The main contributing factor is suggested to be E484K mutation in Beta variant which provides neutralization resistance of the virus. Studies indicate that this mutation in receptor binding motif is in an immunodominant epitope of Spike protein [22,23].

When CoronaVac (Sinovac) and BNT162b2 (Pfizer-BioNTech) vaccines were compared, BNT162b2-vaccinated plasma samples were highly neutralizing according to PRA₅₀ and pseudovirus neutralization assay. In consistent with literature, the human plasma showed a slight decrease in neutralizing titers against the Alpha variant pseudovirus. However, the overall level of neutralizing activity remained largely intact [24]. The fact that BNT162b2-immune plasma largely maintains its ability to neutralize pseudovirus with the Alpha Spike suggests that the Beta variant is unlikely to evade protection provided by the BNT162b2 vaccine. However, pseudovirus neutralization assay showed a substantial drop-in neutralization activity against the Beta variant, consistent with conclusions being reached by others [25,26].

An effective strategy called as heterologous prime-boost which combines different vaccines against the same antigen was utilized in order to enhance the immune response against SARS-CoV-2. A wide range of studies were reported with the increased binding and neutralization effect of antibodies followed by the heterologous regimen [27–30]. Moreover, compared with homologous vaccination, a vector-based or mRNA vaccination after a different type of vaccine such as inactivated or virus vector vaccine supports the immune system with a better immune response and good tolerability [31,32]. Our results demonstrate that, consistent with the literature, the participants vaccinated with 2 doses of CoronaVac followed by one dose BNT162b2 provided higher neutralization capacity against especially the Alpha variant. The heterologous prime-boost regimen demonstrates a substantial increase in neutralization titers compared with the CoronaVac-only regimen and shows competitive results compared with the BNT162b2-only regimen, suggesting that the heterologous regimen potentially offers improved protection due to the boosted immune response, making it a promising strategy for enhancing vaccine efficacy.

Despite high sensitivity, small sample sizes may be unavoidable for establishing neutralization assays. To develop simple pseudovirus-based neutralization assays, small sample sizes have been utilized previously, offering rapid and sensitive assay for SARS-CoV-2 in BSL-2 laboratory [7,33–36]. In this pilot study, a comparable sample size of human convalescent plasma showing homogenously distributed neutralization potency was used for evaluating the pseudovirus system and vaccine response. Here, we utilized pseudoviruses carrying different Spike variant to test neutralizing performance against vaccinated human plasma samples. We also considered the effects of different vaccines on neuralization profiles of each sample. In brief, vaccinated human convalescent serum sample was assayed for neutralization against Alpha, Beta, and Wuhan pseudoviruses to determine their potency which is used for comparison with PRA using authentic Wuhan type virus. Although the pseudovirus assay overestimated the effectiveness of samples with low activity, it allowed for the identification of samples having high neutralization activity accurately. Moreover, the assay did not show any decrease in activity against the Alpha variant, but all samples lost activity against the Beta variant. It is observed that pseudovirus assay successfully discriminates neutralization potency of human plasma samples against the Spike variants. Therefore, the neutralization assay could be accepted as a sensitive method both for the different neutralization activities of samples and different Spike variants. In the light of results, our pseudovirus neutralization assay can be used to screen large numbers of plasma samples. The capacity to analyze a substantial quantity of plasma samples is expected to grow in significance as widespread vaccination initiatives commence on a global scale. This is especially true if the virus transitions into an endemic phase, necessitating periodic vaccination. Consequently, studies like this offer a means for monitoring the neutralizing antibody response at a population level among vaccinated individuals over time. This surveillance could inform decisions regarding the need for additional vaccine doses in the future, particularly among high-risk patient cohorts. Neutralizing antibody responses, even though not the only immunologically significant indicator of vaccine efficacy, are more easily quantified in a considerable number of samples when compared with evaluations of T cells [37]. Noticeably, this study demonstrates the accuracy of the pseudovirus system that mimics emerging variants, and obtained results showed this assay is sensitive particularly for those with multiple mutations in the RBD.

5. Conclusion

In conclusion, pseudovirus-based assays have shown a strong correlation with the authentic virus-based assay, even though pseudoviruses only contain the envelope proteins of the authentic virus. However, the plaque assay requiring the laboratory with the top level of security and equipment critically limited the number of study samples to test in the neutralization assay in this study. On the other hand, the pseudovirus-based neutralization assays offer notable advantages due to their practicality with smaller serum sample volumes and their safety, which requires less stringent security measures [26]. Taken together, the constructed pseudovirus system that is reliable and feasible based on a VSV packaging system could be greatly beneficial for developing SARS-CoV-2 vaccines and therapeutic drugs as well as for testing the inhibition potency of convalescent plasma samples.

Funding Statement

This project is financially supported by Sabanci University Nanotechnology Research and Application Center (SUNUM) and Scientific Research Projects Coordination Unit of Istanbul University-Cerrahpasa.

Supplemental material

Supplemental data for this article can be accessed at https://doi.org/10.1080/17576180.2024.2411920

Author contributions

AN Cimen and GC Torabfam conducted all the experiments, drafted and edited the manuscript. S Cetinel and O Kutlu designed, supervised the study and revised the manuscript. YT Tok, E Yucebag, N Arslan, D Saribal, G Esken, O Dogan, MA Kuskucu, B Mete, G Aygun, F Tabak, F Can, O Ergonul, K Midilli, S Cetinel and O Kutlu interpreted the data. All authors read and approved the final version of the manuscript.

Financial disclosure

This project is financially supported by Sabanci University Nanotechnology Research and Application Center (SUNUM) and Scientific Research Projects Coordination Unit of Istanbul University-Cerrahpasa. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Ethical conduct of research

This study was approved by the Ethics Committee of Istanbul University-Cerrahpasa Medical Faculty, with the approval number 83045809-604.01.02. All participants provided informed consent prior to their inclusion in the study. All procedures used in the study were carried out in accordance with relevant guidelines and regulations.