SARS-CoV-2 Antibody Neutralization Assay Platforms Based on Epitopes Sources: Live Virus, Pseudovirus, and Recombinant S Glycoprotein RBD

Plaque reduction neutralization test (PRNT)

Vero E6 cells maintained in DMEM medium supplemented with 10% fetal bovine serum and 100 IU/ml penicillin/100 µg/ml streptomycin in humidified CO₂ incubator can be readily infected by SARS-CoV-2 virus (56). The infecting virus replicates within the cells causing the cell to be lysed and form visible plaques, which count as plaque-forming units per ml (PFU/ml). Thus, the PFU/ml corresponds to the virus infectivity level which is determined through titer evaluation. In virus-based neutralization assay using Vero E6 cells, briefly, the cells are seeded in a multiwell-plate then infected with the virus in the following day. The inoculum is removed after 1-h incubation, and the infected cells are overlaid with a semi-solid medium containing agar or carboxymethyl cellulose to prevent further sporadic infection. The inactivated CP or serum isolated from either the COVID-19 patients or vaccinated patients are mixed with virus solution corresponding to the designated PFU/ml. If the serum added to the culture contains nAbs, it will inhibit Vero E6 cell infection, which reduces viral PFU counted after 2–3 days incubation. The higher the Ab titer in the plasma or serum sample, the lower PFU will be observed (56, 57). Therefore, virus neutralization assay is also known as PRNT as well. To ease the plaque investigation, they can be stained by using a dye such as crystal violet. In addition to plaque visualization by crystal violet staining, fluorescence-based staining also can be carried out to observe the infected cells by utilization of fluorochrome-tagged secondary Abs (58). Basically, the number of cells that expressed fluorescent signals is defined as the number of infected cells (59).

PRNT, which is carried out in Vero cells, is a crucial part of evaluating vaccine efficacy, COVID-19 patients’ immunity, and the quality of CP. There are reports on the application of PRNT on CP evaluation designated for COVID-19 patient treatment. Shen et al. (60) had performed a serum neutralization assay of SARS-CoV-2 virus by nAbs of CP in Vero cells. Technically, the serum samples need to be heated at 56°C for 30 minutes to inactivate the life viruses. Serial dilution of serum is made to be incubated with the viruses to determine the nAbs titer. The mixture is then added to the confluent monolayer Vero cells in a 96-well plate in which only free viruses are able to attach ACE2 receptors. The visualized plaques are then observed by using a microscope to determine the neutralizing Ab titer.

Shen et al. (60) reported that by this test, the CP neutralizing Ab titers fall between 80 and 480. PRNT was also performed by Duan et al. (61) to determine the neutralizing activity of CP. This test was conducted in a high biosafety level (BSL)-4 laboratory of Wuhan Institute of Virology, China toward CPs collected from 40 recovered patients who were recommended as plasma donors. As the results, 39 patients showed high Ab titers at least 1:160 (61). Nevertheless, PNRT also possesses several disadvantages as its implementation requires virulent virus propagation in a BSL-3 facility. In addition, it does not allow rapid screening of plasma to meet the current high demand. In accordance with this matter, PNRT for CP is recommended but not obliged by FDA. The retained plasma samples can be tested later after the administration of CP (62).

SARS-CoV-2 pseudotyping techniques

Pseudovirus can be engineered to be either replication-competent or incompetent depending on the modification of the virus genome (47, 48, 68, 69). Compared to wild-type viruses, replication-incompetent pseudoviruses can only undergo a single infection cycle in the infected host cells. Hence, most SARS-CoV-2 pseudoviruses are engineered to form replication-incompetent ones to essentially devoid of virulent viral components and generate more safety to be handled in BSL-2 laboratories (48).

In a replication-competent pseudovirus, the wild-type spike gene is substituted with the gene encoding the designated spike gene in order the virus genome bears the complete material genetic to produce infectious recombinant pseudovirus. For instance, to produce a replication-competent SARS-CoV-2 pseudotype using VSV backbone, we need to generate a VSV genome bearing knocked out G glycoprotein gene to enable the incorporation of S glycoprotein into the VSV backbone (68). The procedure to generate pseudotyped viruses is less complicated than replication-incompetent pseudovirus because the G glycoprotein gene is directly substituted with S glycoprotein (rVSVΔG/SARS-CoV-2 S*), so the additional plasmid is not needed to complement the removal of the G-glycoprotein gene. Supporting plasmids encoding viral structural proteins: nucleoprotein (N), phosphoprotein (P), and large polymerase protein (L) are also added to increase packaging efficiency, which expression can be regulated by a T7 promoter of vaccinia virus. Therefore, to promote the expression of recombinant protein, the vaccinia virus is needed. Alternatively, the vaccinia virus also can be replaced by using plasmid, which encodes the T7 polymerase of this virus.

For generating replication-incompetent pseudovirus, the gene encoding host-virus spike protein is knocked out and complemented with an additional recombinant plasmid DNA to express the spike gene of interest (64). Several protocols can be used for this purpose, including the utilization of one set of plasmids bearing VSV-ΔG virus genome, plasmids encoding N, P, L, G, and an additional plasmid bearing S glycoprotein gene. Basically, the protocol is divided into 2 steps, first, the generation of rVSV-ΔG backbone complemented with G glycoprotein, second, pseudotyping by addition of rVSV into cells overexpressing S glycoprotein. Similar to Case et al. (68), Whitt (71) also regulates the expression of VSV genes by using the T7 promoter that needs T7 polymerase of the vaccinia virus. By this method, either propagation and purification of the vaccinia virus or plasmid bearing T7 polymerase gene are needed before generating recombinant VSV (rVSV). This T7 pol plasmid also can be cotransfected together with other plasmids to generate rVSV by reverse genetic or by generating stable cells expressing this T7 polymerase prior to transfection with VSV plasmids (Fig. 1). To ease the pseudotyping process, fortunately, rVSV lacking G glycoprotein gene in the genome can be obtained from Kerafast (rVSV-ΔG/G∗ luciferase, EH1020-PM). The rVSV is ready to use for pseudotyping by adding the virus solution into cells overexpressing S glycoprotein (72). This rVSV also can be amplified by infecting the cells overexpressing G glycoprotein using this recombinant virus so that the G glycoprotein gene will complement the deleted gene in the virus genome.

Figure 1
General neutralization assay procedures by using live virus, pseudovirus and recombinant S glycoprotein RBD as epitope sources.

• Click for larger image
• Download as PowerPoint slide

Besides the utilization of VSV core, HIV lentivirus core also has been used for pseudotyping (Table 1). VSV elicits bullet-shaped conformation, while HIV lentivirus, resembling SARS-CoV-2, exhibits round-shaped conformation. Similar to VSV pseudotyped, HIV lentivirus pseudotyped also can be engineered to be replicative or non-replicative by modifying the S glycoprotein gene to substitute the HIV envelope in the virus genome or to complement it by subcloning into an additional plasmid (73, 74, 75, 76, 77).

Distinct packaging cell lines can be used to generate pseudoviruses including Vero, BHK-21, and 293T cells, in which Vero cells generate the highest pseudovirus titer (69). However, the problems arise in accordance with the lower pseudovirus titer when the S glycoprotein is expressed at a low level and slowly releases to the cell membrane. The S glycoprotein gene needs to be subcloned into a plasmid DNA vector that enables a high expression level of S glycoprotein to support the production of functional pseudoviruses (78). Alternatively, a viral vector can be used to generate stable cells that over-expressed the S glycoprotein (79). Next, to accelerate the release of S glycoprotein from the ER, about 18–19 amino acids at the C terminal domain associating with the ER-retention tail can be removed (68, 69).

Pseudovirus neutralization assay techniques

Similar to PRNT, pseudovirus neutralization assay also requires cell lines that express human ACE2 or animal ACE2, which possesses high similarity to human ACE2 to enable S glycoprotein-ACE2 interaction. Hoffman et al. (54) conducted a pseudovirus neutralization assay in a panel of cell lines and showed that most of the human cells could be used for the assay, including Caco-2 and Calu-3 cells that enable virus cell entry and replication. However, in a replication-incompetent pseudovirus, after receptor binding and cell entry, virus propagation will not occur, so that the neutralization assay by using these types of pseudoviruses also can be performed in other cell lines expressing exogenous human ACE2, for instance, HEK293T and BHK-21 cells which are transfected with a plasmid bearing human ACE2 gene to facilitate receptor binding and pseudovirus internalization without the need to support virus replication (64, 73) (Fig. 2). Co-expression of exogenous human ACE2 with TMPRSS2 was also performed to elevate the degree of pseudovirus internalization (54, 74). Pseudovirus-based neutralization assays have been widely conducted along with PRNT (37).

Figure 2
Interaction of hACE 2 with SARS-CoV-2 virus, pseudovirus and RBD either in absence or presence of nAbs I) In the absence of nAbs SARS-CoV-2 virus binds ACE 2 then followed by membrane fusion and cell entry release of its genetic material will lead to virus propagation, II) SARS-CoV-2 pseudovirus is also capable of binding ACE 2 and entering the infected cells non-replicative pseudovirus binds ACE 2 and internalizes, but does not replicate within host cells, III) RBD binds to recombinant human ACE 2 or ACE 2 sub-unit The presence of nAbs blocks ACE 2 interaction with virus, pseudovirus and RBD.

• Click for larger image
• Download as PowerPoint slide

The sensitivity of the pseudovirus neutralization assay will depend on the reporter system applied to analyze cell entry response after receptor binding. Nie et al. (48) synthesized the SARS-CoV-2 pseudotype by using VSV backbone-bearing luciferase reporter to perform a neutralization assay of serum samples with the serial dilution up to 1:3,200. Luciferase metabolizes its substrate to produce chemiluminescence light that can be analyzed using a luminometer. This reporter offers the greatest sensitivity among other reporters such as fluorescence protein GFP and miRFP. Pseudovirus-based neutralization assays also can be recalled according to the utilization of reporter, for instance, when GFP reporter is used, it can be named as GFP reduction neutralization test (68).

ELISA techniques

The simplest ELISA technique is binding ELISA that utilizes the recombinant RBD. It can be conducted by coating the ELISA plate with this peptide, followed by the addition of antisera and secondary Ab either anti-IgG or anti-IgM. Shen et al. (60) have performed a binding ELISA of prospective donors’ serum one day before CP collection by utilizing recombinant S1 RBD at 200 ng/well to detect anti-SARS-CoV-2 Abs. Serial dilution of serum (1:600, 1:1,800, 1:5,400, and 1:16,200) was applied to each well to determine ELISA endpoint dilution Ab titer. All tested CP showed Ab titers equal to or more than 1,800 and were used for study ¹¹. In addition, Li et al. (82) also conducted S-RBD-specific IgG Ab measurement for the quality control of CP. They used 100 ng/well of recombinant RBD polypeptides to detect Abs in a serial diluted CP (1:160, 1:320, 1:640, and 1:1,280). As a result, all CP obtained from 64 donors had a titer higher than 1:160, which fulfills the quality control requirement based on the China national guidelines for COVID-19 CP (82). Moreover, Duan et al. (61) also conducted ELISA to examine the existence of IgG Abs in the donor’s plasma before application of CP therapy. However, RBD-based binding ELISA that had been conducted for CP analyses that is associated with nAbs titer (60, 61, 82) detects total binding Abs (bAbs) and is unable to differentiate between bAbs and nAbs (83, 84, 85). Thus, to clarify that RBD-binding Abs also act as nAbs, the ELISA technique can be modified with the utilization of recombinant human ACE2 or soluble ACE2 as decoy receptors to implicate the interaction between RBD and human ACE2. In this ELISA-based neutralization assay, the presence of nAbs in antisera will block RBD-ACE2 interaction. Thus, this ELISA technique is also known as blocking ELISA (34, 81). This RBD-ACE2-based blocking ELISA had been applied to measure the neutralization of RBD-human ACE2 by antisera after immunization of mice with a vaccine candidate RBD-Fc to evaluate its efficacy (81). This blocking ELISA also has been validated to measure nAbs level in COVID-19 patients for serosurveillance that achieves about 99% specificity and about 95% sensitivity (34).

Technically, either RBD or ACE2 can be used for coating the ELISA plate even though ELISA using RBD as coating peptide is easier to be performed to determine nAbs titers. In the procedure using RBD as coating peptide, serum samples in a serial dilution are added to the plate to give the opportunity of RBD-nAbs interaction. The nAbs content in the sample will prevent ACE2 attachment to RBD, leaving recombinant ACE2 to bind free RBD. On the other hand, when recombinant ACE2 is used as a coating peptide, nAbs in the serum samples need to be incubated with recombinant RBD before adding it to the plate. In this case, it is less practical to incubate serial dilution of samples with RBD (68).

ELISA is a widely used technique and is a simpler method in comparison to pseudovirus neutralization assay. Neutralization assay by ELISA offers a lot of advantages because it enables high throughput analysis of serum samples (86). In addition, the protocol does not involve cell culture, allowing the analysis to be performed in ordinary laboratories. Besides, there are many commercial ELISA kits available in the market along with recombinant RBD and ACE2 to be applied in the routine in-house (laboratory) ELISA. Nevertheless, even though the results obtained from this assay also show a correlation with PRNT, there is a possibility that nAbs contained in the samples do not bind RBD as the major epitopes. As the results, it contributes to the lower measured titer. Even though RBD contains important epitopes for generating nAbs, other epitopes may also generate nAbs toward SARS-CoV-2 virus. Chi et al. (87) isolated Abs from CP and identified a neutralizing Ab that, instead of RBD, binds the N-terminal domain of S glycoprotein (82, 87). Thus, the assay does not fully represent the actual nAbs titer (Table 2).

Table 2
Comparison of Ab neutralization assay platforms

• Click for larger image
• Click for full table
• Download as Excel file

WHO establishes anti-SARS-CoV-2 Abs as internal standard and reference panel to create harmonization of nAbs assessments due to various samples and assay types conducted at the different laboratories worldwide (88). This biological standardization will also lead to the determination of the international unit to express the potency of tested Abs or serum samples. To test the internal standard and reference panel candidates, participating laboratories had performed assays based on the protocols developed at their laboratories. Those protocols involve neutralization assays utilizing live virus, VSV-derived SARS-CoV-2 pseudovirus, HIV lentivirus-derived SARS-CoV-2 pseudovirus, and in-house ELISA. Neutralization assay platforms discussed in this review are summarized in Figs. 1, 2, and Table 2.