Generation of SARS-CoV-2 Spike Pseudotyped Virus for Viral Entry and Neutralization Assays: A 1-Week Protocol

The COVID-19 pandemic caused by the SARS-CoV-2 coronavirus requires reliable assays for studying viral entry mechanisms which remains poorly understood. This knowledge is important for the development of therapeutic approaches to control SARS-CoV-2 infection by permitting the screening for neutralizing antibodies and other agents that can block infection. This is particularly important for patients who are at high risk for severe outcomes related to COVID-19. The production of pseudotyped viral particles may seem like a daunting task for a non-virology laboratory without experience in the two most commonly used pseudotyping systems, namely retro/lentiviruses and vesicular stomatitis virus (VSV) which lacks the VSV envelope glycoprotein (VSVΔG). By incorporating the most up-to-date knowledge, we have developed a detailed, easy-to-follow novel protocol for producing SARS-CoV-2 spike-bearing pseudovirus using the VSV-ΔG system. We describe the infection assay which uses GFP fluorescence as a measure of infection in a 24-well live imaging system. We present results of our optimization of the system to enhance viral infection levels through the over-expression of human ACE2 receptor and the overexpression of at least one of two proteases - TMPRSS2 or Furin, as well as, supplementation with Poloxamer 407 (P407) and Prostaglandin E2 (PGE2) as adjuvants. We show that the system works efficiently in three unrelated, clinically relevant cell lines: human 293T (renal epithelial) cells, human Calu-3 (lung epithelial) cells, and the non-human primate (African Green Monkey) cell line, Vero-E6 (renal epithelial) cells. In addition, we have used this system to show infection of human induced pluripotent stem cell-derived cardiomyocytes (iPS-CMs). This system is efficient (virus generation, titration, and infection assays can be performed in 1 week), quantitative, inexpensive, and readily scalable for application in drug development and therapeutic screening approaches.


INTRODUCTION
The coronavirus disease (COVID-2019) pandemic is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (1). SARS-CoV-2 is an enveloped, positive-sense single stranded RNA virus belonging to the Betacoronavirus genus of the Coronaviridae family of viruses (2). Understanding the viral entry process is key for the development of therapeutic strategies to prevent viral spread. The spike (S) glycoprotein is present on the surface of SARS-CoV-2 and is responsible for the viral tropism through host cell recognition by binding to the human angiotensin-converting enzyme 2 (hACE2) receptor (3,4). Coronavirus S glycoprotein is a class I viral fusion protein (5) containing a homotrimeric conformation with two functional subunits termed S1 and S2 present in the N-terminal domain (NTD) and C-terminal domain (CTD), respectively (6). The S protein interacts with hACE2 through the receptor binding domain (RBD) located in the NTD (S1) (7). Once bound to hACE2, the S1 domain is cleaved and dissociated from the protein exposing the S2 domain containing the S transmembrane domain, cytoplasmic tail, and the fusion peptide (8). Spike protein cleavage between the S1 and S2 domain is mediated by several host proteases, such as Furin or the transmembrane protease serine protease-2 (TMPRSS2). This S protein cleavage triggers conformational rearrangements of the S2 domain to activate the protein for viral membrane fusion and release of the viral genetic material into the cells (8,9).
Recovered COVID-19 patients develop neutralizing antibodies (nAbs) against the Spike glycoprotein proving its high antigenicity (9,10). In vitro cellular assays using the Spike glycoprotein are therefore used to study viral infection or Spike/hACE2 interaction. However, the Spike protein alone fails to mimic the virus-host cell interaction. On the other hand, experiments with actual pathogenic SARS-CoV-2 demands strict biosafety level-3 (BSL-3) laboratory conditions (11). The use of replication-restricted pseudoviruses bearing viral coat proteins represents a safe and useful method that has been widely adopted by virologists to study viral entry, detection of nAbs in serum samples, and therapeutic development under less stringent biosafety conditions [e.g., biosafety level-2 (BSL-2)]. Therefore, pseudotyping viral systems have been widely employed to study highly infectious and pathogenic viruses such as Ebola virus, Middle Eastern Respiratory Syndrome (MERS) virus, or SARS viruses (12)(13)(14).
Vesicular stomatitis virus (VSV) is an enveloped negativestranded RNA virus that infects a wide range of animals and less frequently humans causing mild flu-like symptoms (15)(16)(17). The simple structure, and ability to grow in a wide range of mammalian cell types with high titer has made VSV a promising viral vector and a valuable molecular biological and virological tool (18,19). The VSV genome encodes five main proteins: matrix protein (M), nucleoprotein (N), large polymerase protein (L), phosphoprotein (P) and glycoprotein (G) (20). The G protein interacts with host low density lipoprotein receptor (LDLR) for viral entry and cell fusion (21).
To study the entry of viruses that require BSL-3 or−4 containment, recombinant VSVs (rVSVs) in which the glycoprotein gene has been deleted and replaced by a reporter gene -Green Fluorescent Protein (GFP), Red fluorescent protein (RFP/DsRed), Secreted Embryonic Alkaline Phosphatase (SEAP), or firefly Luciferase (fLuc) as summarized in Table 1 -has provided a powerful tool that can be used in laboratories that do not have access to BSL-3 or−4 facilities (22). The advantages of the fluorescent protein reporters are that exact titers of the pseudotyped virus can be determined by counting individual cells infected by the pseudotypes. Because of the robust expression of viral proteins (including the reporter), the sensitivity is extremely high (at the single cell level). No substrates are needed and the equipment needed for analysis is a fluorescence microscope (with option of advanced live imaging system like one described here) or a flow cytometer. Both fLuc and SEAP provide a wide dynamic range when analyzing inhibitors of infection (either antiviral or neutralizing antibodies). The disadvantage is that both require additional steps to assay the reporters. In the case of the SEAP reporter, the pseudotyped virus needs to be purified by centrifugation through a sucrose cushion to pellet the virus away from the SEAP that is produced and in the culture supernatant of the pseudotyped virus. However, the cost of the SEAP reagents are significantly less than the reagents needed to assay fLuc, which is not secreted. Both of these reporters also require access to a plate reader that can analyze either luminescence (for fLuc) or absorbance (for SEAP).
rVSV G * Reporter can be packaged and released from cells transfected with mammalian expression plasmid encoding G protein or the envelope protein from other viruses (3,22). Thus, rVSV G * Reporter system can be used to produce single-round replication-restricted VSV pseudoviruses bearing any viral surface glycoprotein (especially from those that would otherwise require work under BSL-3 and BSL-4 containments) in BSL-2 laboratories (22).
In this study, we present an effective methodology for studying the role of the SARS-CoV-2 spike protein in determining cellular tropism and viral entry with potential applications for the screening of anti-SARS-CoV-2 neutralizing antibodies and alternative SARS-CoV-2 entry inhibitors. This approach is based on the combination of previously developed resources and approaches to address this important area of research in SARS-CoV-2 infection and therapeutic development. We have utilized a commercially available plasmid for the expression of a truncated GFP-bearing SARS-CoV-2 spike protein and recombinant VSV G particles to develop a detailed, easyto-follow protocol for making the SARS-CoV-2 spike-bearing pseudovirus. The truncated spike protein was used because it deletes the Endoplasmic Reticulum (ER)-retention signal found in the cytoplasmic domain of the spike protein. This results in a slightly higher levels of the SARS-CoV-2 spike/VSV pseudotypes (18,23). We describe the infection assay using GFP fluorescence as a measure of infection utilizing a live imaging system. We show that for optimal SARS-CoV-2 spike protein-based pseudovirus entry, we require the over-expression of human ACE2 receptor, and at least one of two proteases-TMPRSS2 or Furin. We show that the system works efficiently in 3 cell lines: 293T (human renal epithelial) cells, Calu-3 (human lung epithelial) cells, and the non-human primate cell line, Vero-E6 (African Green Monkey renal epithelial) cells, as well as, in human iPS-derived cardiomyocytes-all reported to be infected by SARS-CoV-2 (3,(24)(25)(26)(27). The system is efficient, quantitative, inexpensive, and the virus generation, titration, and infection assay can be performed in 1 week making this approach amenable for use in neutralizing antibody analysis or high content therapeutic screening.

Equipment and Cautions
-Personal Protective Equipment. For all the procedures with pseudovirus and lentivirus generation use laboratory coats, latex gloves, face mask, and protective eye wear -CAUTION! Use a solution of Sodium hypochlorite 10% to wash or discard all materials that are in contact with pseudovirus and lentivirus.

Production of VSV G Bearing SARS-CoV-2 Spike-19 (Truncated Spike)
This protocol is to generate VSV pseudoviruses bearing S protein (rVSV-GFP G * Spike) in 293T cell line with which to inoculate cells of interest for infection and neutralization assays (Figure 1). This protocol is based on the overexpression of a recombinant SARS-CoV-2 spike protein followed by pseudotyped VSV particle production by treating the spike protein expressing cells with rVSV G * G. In this protocol we used already produced rVSV G * G virus but these viral particles can be produced using the protocol of Whitt (22 Note: IU or Infectious Units is the titer of the virus. Because the pseudotypes undergo a single-round of infection, they don't produce plaques, but instead just infect a single cells, so we use the term infectious units rather than Plaque Forming Units (PFU). In addition, an MOI = 5 ensures that ∼99.3% of the cells are infected based on a normal Poisson distribution. We found this to be optimal for production of VSV pseudotypes. Using a higher MOI can cause more rapid cell killing since a higher number of cells are receiving more than 1 infectious unit. A higher MOI also reduces the total amount of pseudotype virus produced ostensibly because the cells are undergoing apoptosis before maximal pseudotype virus production occurs. Note: rVSV G * G viral particles can be purchased from Kerafast or produced using the protocol by Whitt (22). Briefly, 293T cells are plated in 100 mm dish and transduced with the VSV glycoprotein expression vector pCAGGS-G-Kan (16 µg) (Kerafast) using 80 µl of Lipofectamine TM 2000 using the can be performed at the same day. We strongly recommend to produce rVSV G * G working stock, in at the least 2-100 mm dishes to produce sufficient virus for the experimental plan.
11. After 24-48 h post rVSV G * G infection, the supernatant containing the VSV pseudovirus carrying the spike protein of SARS-CoV-2 is harvested, clarified by centrifuging for 10 min at ∼1,320 × g (slow deceleration mode). The virus should be used immediately or aliquoted into small batches and stored at −80 • C. Immunoblotting for both rVSV G-GFP*G (control) and pseudotyped SARS-CoV-2 viruses were performed to verify the incorporation of FLAG-tagged Spike glycoprotein into the rVSV G. Shown are representative cellular images (B) of rVSV G-GFP*G and rVSV G-GFP*SARS-CoV-2 particle inoculation into 293T-ACE2+TMPRSS2 with and without Anti-VSV-G (2 µg/mL). After 24 h, the GFP signal from 25 images per well was quantified from IncuCyte scan, and normalized to cellular area (C), and by qPCR from a single well and normalized to GAPDH (D). Infection values are presented as % GCU to signal from corresponding control group, or as transcript levels. GCU: Green Calibrated Units. Scale bar: 100 µm. Data are mean ± SEM. ****p < 0.0001 using One-Way ANOVA with Tukey's post-hoc test. Each experiment was replicated at least three times.
Note: the virus containing supernatant should not be clarified by filtration.
Note: Use the supernatants of rVSV G * G and rVSV G * SARS-CoV-2 Spike for immunoblotting to confirm the spike incorporation into the rVSV G virus. Also, establish a neutralization assay using anti-VSV-G as part of proof of concept (Figure 2).

Generation of Lentivirus for Over-expression of Human ACE2, TMPRSS2, and Furin in Cells of Interest
For efficient infection of SARS-CoV-2 pseudovirus in cells of interest, the cells need to express the machinery needed for Spike protein recognition and priming. Therefore, cells of interest need to over-express human ACE2, TMPRSS2, and Furin as shown in  Note: Alternative methods can be used to concentrate the virus in the absence of an ultracentrifuge, such as using the LentiX TM concentrator solution (Takara Bio).
10. After centrifugation, the supernatant is removed in a biosafety cabinet and the tube is placed inverted on a clean tissue paper for 3 min. Add 200 µl of PBS as final volume and thoroughly resuspend the pellet. Aliquot the resuspended virus into fresh tubes (20 µl/tube) and store at −80 • C until needed.
Note: Avoid freeze-thaw cycles because it reduces considerably the lentiviral titer. Use ice to thaw the small aliquots. The present protocol uses 2 adjuvants to improve lentiviral transduction. Poloxamer 407 (P407) is a hydrophilic non-ionic surfactant, which together with other poloxamers are commonly used to improve drug delivery and as an adjuvant in several cells lines since they can act by membrane fluidization, increasing lipid exchange and reducing membrane viscosity (30,31). PGE2 acts by enhancing intracellular cAMP leading to an improvement in viral transduction. The combination of these two molecules (P407 and PGE2) has been previously shown improve lentiviral transduction in CD34+ cells with high-efficiency and without significant cytotoxicity (32    Note: This method of inoculation yields to moderate infection levels, an inherent characteristic of the SARS-CoV-2 when used to pseudotype either VSV or HIV, as shown by others (3,34).
3. Inoculate with the SARS-CoV-2 spike protein pseudotyped VSV G by adding 100-200 µl of complete media to bring the final volume to 400 µl. 4. Twenty to twenty four hours post infection, scan the plates using the Incucyte ZOOM imaging system (Sartorius) and analyze the images (Figures 4, 5 (Figure 6). See Live Cell Image Acquisition and Analysis section.
Live Cell Image Acquisition and Analysis 1. Twenty to twenty four hours post infection, scan the plates with a 10X objective using the IncuCyte ZOOM imaging system. 2. Open the Incucyte TM ZOOM software and set up the system to obtain 25 images per well using the green and phase channels. 3. When the scanning is complete, create an image set to quantify all parameters of phase-contrast area as well as fluorescent measures. Divide the Total Green Object Collection of Cells for RNA and Protein Work 1. After scanning, the cells can be washed in cold PBS, scraped, and collected in 80-100 µl RIPA Lysis buffer and Extraction Buffer or Cell Disruption Buffer from the Mirvana Kit, and used for RNA extraction for qPCR or for immunoblotting. It is important to collect the cells immediately after the IncuCyte scan for accurate comparison between GFP expression by immunoflourescence vs. qPCR. In case of delays between the scan of the plate and the qPCR, there will be some differences in the signal from both methods. 2. For qPCR, extract RNA from cell lysates using Mirvana Paris Kit or an alternative kit, and perform reverse transcription and cDNA amplification using a cDNA kit. Quantify gene transcripts by qPCR using Taqman assays for GFP and GAPDH and a "Fast" qPCR mastermix on an ABI 7900HT thermocycler. For each group, plot fluorescence against the number of cycles on a logarithmic scale and normalize data normalized to the endogenous GAPDH control. The normalized Cycle Thresholds (delta CTs) can be converted to absolute transcripts using an equation where a CT of 40 corresponds to 1 transcript.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by University of Miami IRB. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
JC and GL performed the experiments. JC and LS analyzed the data and prepared the figures. JC, GL, and LS interpreted the results of experiments. JC drafted the