Protocol and Reagents for Pseudotyping Lentiviral Particles with SARS-CoV-2 Spike Protein for Neutralization Assays

Abstract

SARS-CoV-2 enters cells using its Spike protein, which is also the main target of neutralizing antibodies. Therefore, assays to measure how antibodies and sera affect Spike-mediated viral infection are important for studying immunity. Because SARS-CoV-2 is a biosafety-level-3 virus, one way to simplify such assays is to pseudotype biosafety-level-2 viral particles with Spike. Such pseudotyping has now been described for single-cycle lentiviral, retroviral, and vesicular stomatitis virus (VSV) particles, but the reagents and protocols are not widely available. Here, we detailed how to effectively pseudotype lentiviral particles with SARS-CoV-2 Spike and infect 293T cells engineered to express the SARS-CoV-2 receptor, ACE2. We also made all the key experimental reagents available in the BEI Resources repository of ATCC and the NIH. Furthermore, we demonstrated how these pseudotyped lentiviral particles could be used to measure the neutralizing activity of human sera or plasma against SARS-CoV-2 in convenient luciferase-based assays, thereby providing a valuable complement to ELISA-based methods that measure antibody binding rather than neutralization.

Keywords: 293T-ACE2; ACE2; ALAYT; COVID-19; SARS-CoV-2; Spike; coronavirus; cytoplasmic tail; lentiviral pseudotype; luciferase; neutralization assay.

Figure 1

General approach for lentiviral pseudotyping. (A) 293T cells are transfected with a plasmid encoding a lentiviral backbone (genome) expressing a marker protein, a plasmid expressing Spike, and plasmids expressing the other HIV proteins needed for virion formation (Tat, Gag-Pol, and Rev). The transfected cells produce lentiviral particles with Spike on their surface. These viral particles can infect cells that express the ACE2 receptor. (B) We used three variants of Spike: The codon-optimized Spike from SARS-CoV-2 strain Wuhan-Hu-1, a variant containing mutations K1269A and H1271A in the cytoplasmic tail (such that the C-terminal five amino acids are ALAYT), and a variant in which the cytoplasmic tail of Spike has been replaced with that from influenza hemagglutinin (HA). (C) Spike expression on the surface of 293T cells transfected with the plasmids expressing our three Spike constructs was measured using flow cytometry 24 h post-transfection. Spike expression was measured by staining with in-house produced CR3022 antibody [43,44,45] at a concentration of 10 μg/mL followed by staining with an anti-human Fc antibody conjugated to APC (Jackson Labs, 109-135-098) at a 1:100 dilution.

Figure 2

293T-ACE2 cells are infectable with SARS-CoV-2 Spike-pseudotyped lentiviral particles. (A) Flow cytometry plot showing expression of human ACE2 by the 293T-ACE2 cells (grey shaded) at passage 12 compared to parental 293T cells (white fill) as quantified by staining with an anti-ACE2 antibody (see Section 4.2 for detailed methods). The gate was set so the parental 293T cells were 2% positive. (B) Microscope images showing ZsGreen expression in 293T-ACE2 or 293T cells at 58 h after incubation with Spike- or VSV G-pseudotyped lentiviral particles with the ZsGreen backbone. For each viral entry protein, 293T and 293T-ACE2 cells were incubated with equal volumes of virus. Cells were incubated with 1/20^th the volume of VSV G-pseudotyped lentivirus compared to Spike-pseudotyped lentivirus. The decrease in infected cells for the Spike-HAtail virus compared to the other Spike-pseudotyped lentiviruses is consistent with this virus having somewhat lower titers (see Figure 3A).

Figure 3

Titers of Spike-pseudotyped lentiviral particles in 293T-ACE2 cells. (A) Titers of the ZsGreen backbone pseudotyped with the three Spike variants or VSV G, as determined by counting green cells via flow cytometry analysis at 48 h post-infection and then calculating transduction-competent viral particles per mL from the percentage of green cells. The “n.d.” indicates that the titer was not detectable. Data shown are from a single representative example. (B) Titers of the Luciferase-IRES-ZsGreen backbone as determined by measuring relative luciferase units (RLUs). RLUs were determined at 48 h after infecting ~2.3 × 10⁴ 293T-ACE2 cells per well in 96-well plates. The RLUs per mL for the Spike-pseudotyped viruses are the average of three three-fold serial dilutions of virus starting at 50 μL virus in a total volume of 150 μL. For the VSV G-pseudotyped virus, RLUs per mL were averaged from two three-fold dilutions starting at 3 μL virus in a total volume of 150 μL. (C) Microscope images showing 293T-ACE2 cells transduced with Spike pseudotyped virus with either the ZsGreen or Luciferase-IRES-ZsGreen backbone at 60 h post-infection. As can be seen from the images, the ZsGreen backbone gave a stronger fluorescent signal than the Luciferase-IRES-ZsGreen backbone, presumably because this protein was expressed more strongly as the sole CMV-promoter driven transcript than as the second transcript driven by an IRES.

Figure 4

Neutralization assays. (A) Neutralization assay using plasma collected from a confirmed SARS-CoV-2-infected patient at 19 days post-symptom onset (“p.s.o.”). The IC50 for this plasma was 1:2076 for the Spike-pseudotype, 1:1334 for the Spike-ALAYT-pseudotype, and 1:1605 for the Spike, HAtail-pseudotype. (B) Neutralization assay using soluble ACE2 protein fused to the Fc domain from IgG (ACE2-Fc). The IC50 for ACE2-Fc was 2.49 μg/mL for the Spike-psdeudotype, 1.75 μg/mL for the Spike-ALAYT-pseudotype, and 0.25 μg/mL for the Spike-HAtail pseudotype. (C) Negative control sera collected prior to the emergence of SARS-CoV-2 did not neutralize the Spike-pseudotyped lentiviral particles. The serum from 1989 was from a person of a similar age at the time of serum collection as the confirmed SARS-CoV-2 infected patient whose plasma was tested in A. High concentrations of naïve serum seemed to enhance luciferase signal, perhaps because of components that improve cell-growth. Each point shows the average of duplicate values with error bars showing standard error.

Figure 5

Example plate layout for neutralization assays. It is possible to run full-dilution series of two sera or plasma samples in duplicate on each plate with the necessary controls. These controls include media only, cells only, and virus-only wells, as well as four wells of virus-infecting 293T cells to confirm the lack of infection with Spike-pseudotyped lentivirus in the absence of ACE2. The average signal from the “Virus Only” and “Virus + 293Ts” wells provides the background signal. The “Virus + Cells” wells represent maximum infection without any serum and provide a metric for 100% virus infectivity. Note that “cells” here refers to the 293T-ACE2 cells. The different colors simply denote different conditions, such as different serum samples or cells. These conditions are labeled in the left-most column of the figure (or in each individual cell for the “Virus + 293Ts” condition).