A Dual-Fluorescence Labeling Pseudovirus for Real-Time Imaging of Single SARS-CoV-2 Entry in Respiratory Epithelial Cells

The pseudovirus strategy makes studies of highly pathogenic viruses feasible without the restriction of high-level biosafety facility, thus greatly contributing to virology and being used in research of SARS-CoV-2. Here, we generated a dual-color pseudo-SARS-CoV-2 virus using an HIV-1 pseudovirus production system and the SARS-CoV-2 spike (S) glycoprotein, of which the membrane was labeled with lipophilic dye (DiO) and the genomic RNA-related viral protein R (Vpr) of the viral core were fused with mCherry. With this dual-color labeling strategy, not only the movement of the whole virus but also the fate of the labeled components can be traced. The pseudovirions were applied to track viral entry at a single particle level in four types of the human respiratory cells: nasal epithelial cells (HNEpC), pulmonary alveolar epithelial cells (HPAEpiC), bronchial epithelial cells (BEP-2D), and oral epithelial cells (HOEC). Pseudo-SARS-CoV-2 entered into the host cell and released viral core into the cytoplasm(cid:0)which clearly indicates that the host entry mainly occurred through endocytosis. The infection e�ciency was found to be correlated with the expression of the known receptor of SARS-CoV-2, angiotensin-converting 2 (ACE2) on the host cell surface. We believe that the dual-color �uorescence labeled pseudovirus system created in this study can be a useful tool in SARS-CoV-2/COVID-19 for many purposes.


Introduction
2][3] Understanding the infection mechanism has become one of top priorities.5][6] SARS-CoV-2 enters host cells mainly through endocytosis after binding to the ACE2 receptor, and the proposed model is based on lysosomotropic inhibitor treatment of infected cells. 7However, how does the virus enter the cell after binding to the receptor has never been visualized through live imaging.In addition, how sensitive are different parts of the respiratory tract to the virus remains undocumented.Our interest is to conduct systematic investigations of these issues.
][10][11] Pseudovirus strategy can provide a safe manner to study highly pathogenic viruses without high-level biosafety facility.3][14] Pseudovirus has part of the functional structure of the target virus, but have lost the function of reproducing in host cells, so they can be used to identify the function of virus components. 15,16 ARS-CoV-2 pseudovirus has been successfully applied to determine the receptor and infection pathway, evaluate neutralizing antibodies, and construct recombinant vaccine. 7,17,18 I][21][22][23] In this study, a dual-color pseudo-SARS-CoV-2 was generated by incorporating SARS-CoV-2 S protein into HIV-1 pseudovirus.Viral lipid envelope and the Vpr protein were labeled with a lipophilic membrane dye (DiO) and mCherry uorescence protein, respectively.The dual-labeled virus was used to infect the human airway epithelium cultures to image the entry pathway and infection e ciency at a single particle level in four different respiratory epithelial cells, including human nasal epithelial cells (HNEpC), human oral epithelial cells (HOEC), human bronchial epithelial cells (BEP-2D), and human pulmonary alveolar epithelial cells (HPAEpiC).The entry pathway and cell-type susceptibility of SARS-CoV-2 to respiratory tract cells were rst identi ed and systematically investigated by single virus tracking.

Results
Construction and characterization of pseudo-SARS-CoV-2 virus.The SARS-CoV-2 S plasmid, encoding the S protein, was synthesized and transfected into the 293T cells.Western blot was used to analyze the expression of SARS-CoV-2 S protein.The major band, re ecting the full-length S protein (180 kDa), was detected using the rabbit anti-S antibody.S proteins were incorporated into the pseudovirions while assembling the virus, and the e ciency was evaluated using a monoclonal mouse anti-S1 antibody.The full-length S protein was incorporated into the pseudotyped virus; however, the majority of S proteins in pseudovirions were cleaved (Fig. 1A).Pseudovirions were further characterized by negative staining in TEM.The round-shaped virus particles had an average size of 100 nm, and the cobbled surface structure of SARS-CoV-2 S proteins had an average size of 15 nm (Fig. 1B).
Pseudo-SARS-CoV-2 were overlaid onto coverslips and stained for immuno uorescence (IF) with an anti-S antibody.The lipid envelope of the pseudotyped virus was labeled using DiO and the majority of DiO co-localized with the S protein (Fig. 1D).Pseudovirions were also stained with an anti-p24 antibody and DiO, and we found that they were well co-localized.Then, pseudo-SARS-CoV-2 were exposed to combined anti-S and anti-p24 immuno uorescence staining, and S protein was found co-localized with p24 protein (Supplementary Fig. 1).These results demonstrated that SARS-CoV-2 S proteins were successfully incorporated into the pseudovirions.
Dual-labeled uorescent infectious pseudovirions.We generated dual-labeled uorescent infectious pseudovirions to visualize the dynamic entry of the SARS-CoV-2 virus into the host cells.The Vpr protein of the viral core was labeled by fusing with the mCherry uorescence protein, and the Vpr-mCherry complex was encapsulated into the pseudo-SARS-CoV-2 during virus assembly.Immuno uorescence staining with an anti-S antibody or anti-p24 antibody veri ed the successful labeling of the Vpr proteins with mCherry uorescence protein (Fig. 1D and Supplementary Fig. 1).The second color was obtained by labeling the lipid envelope with DiO, and uorescence co-localization of Vpr-mCherry and DiO veri ed the successful construction of dual-uorescent pseudo-SARS-CoV-2 (Fig. 1D).
Pseudo-SARS-CoV-2 with/without uorescence were analyzed using RT-PCR to determine whether uorescent labeling affected the viral infectivity.The results showed that the titer activity of single-labeled particles (Vpr-mCherry or DiO) and dual-labeled particles were similar to that of unlabeled pseudovirions, demonstrating that the labeling did not impair virus transmission (Fig. 1C).
Real-time imaging of pseudo-SARS-CoV-2 entry into upper respiratory cells, human nasal epithelial cells (HNEpC).Pseudo-SARS-CoV-2 was used to infect different respiratory epithelial cells to conduct real-time imaging of the viral entry process.The dynamic entry of dual-labeled pseudo-SARS-CoV-2 was rst tracked in the upper respiratory cells, human nasal epithelial cells (HNEpC).Only the co-localized signals of Vpr-mCherry and DiO were considered as single virus.A virus particle was observed on the surface of the HNEpC cell membrane, exhibiting the entry of pseudo-SARS-CoV-2 into HNEpC cells.Fig. 2A-2B and Supplementary Movie 1 show the trajectory of pseudovirions.The virus particle was rst attached to the HNEpC cell membrane and was rapidly transported into the cytoplasm (Fig. 2C).The results of mean square displacement (MSD) indicated that the pseudovirions were endocytosed into HNEpC cells through active transport (Fig. 2D).
A common virus, vesicular stomatitis virus glycoprotein (VSV-G) pseudovirions, was used as the control virus.The control virus was also labeled with DiO and Vpr-mCherry and used to infect HNEpC cells.We observed similar endocytic patterns in HNEpC cells (Supplementary Fig. 1A-1D and Supplementary Movie The release process was further con rmed by using VSV-G pseudovirions as the control virus to infect HNEpC cells.While tracking the dual-labeled control virus, Vpr-mCherry (red) separated from DiO (green) in the host cells (Supplementary Fig. 1E-1I and Movie S10).These results suggested that the viral core of pseudo-SARS-CoV-2 escaped from the envelope and were released into the cytoplasm of HNEpC cells.
The entry process of pseudo-SARS-CoV-2 into lower respiratory cells, human pulmonary alveolar epithelial cells (HPAEpiC).The entry process of dual-labeled pseudo-SARS-CoV-2 was tracked in HPAEpiC cells, and the dynamic behavior of the virus was visualized through real-time imaging in the lower respiratory cells.Fig. 3A-3B and Supplementary Movie 3 show the trajectories of virus particles in HPAEpiC cells.Initially, the virus particles attached to the cell membrane and then were rapidly transported into the cytosol (Fig. 3C).The MSD results suggested that the pseudovirions were endocytosed into the host cells through active transport (Fig. 3D).
Then, the release process of pseudo-SARS-CoV-2 was imaged in human HPAEpiC cells.We visualized the separation of the Vpr-mCherry signal (red) from the DiO signal (green) during virus transportation in the cytoplasm, indicating that the viral core was released from the envelope (Fig. 3E-3F and Supplementary Movie 4).The different dynamic trajectories, velocities, and MSD for these two parts during the separation behavior in these cells are shown in Fig. 3G-3I, demonstrating the dynamic release of the viral core into the cytoplasm.
The control virus, DiO/Vpr-mCherry dual-labeled VSV-G pseudovirions were used to infect HPAEpiC cells and showed similar entry and release processes (Supplementary Fig. 3 and Supplementary Movie 11-12).
The results suggested that pseudo-SARS-CoV-2 entered HPAEpiC cells through endocytosis.
Single-particle tracking of pseudo-SARS-CoV-2 in the other two respiratory epithelial cells.The duallabeled pseudo-SARS-CoV-2 was also used to infect human bronchial epithelial cells (BEP-2D), a type of lower respiratory cells, and human oral epithelial cells (HOEC), a type of upper respiratory cells.The endocytic patterns of pseudo-SARS-CoV-2 in these two cell lines were visualized through single-particle tracking.Virus particles were attached to the cell membrane, rapidly transported into the cytosol, followed by the sequential release of the viral core from the envelope (Fig. 4-5 and Supplementary Movie 5-8).The control group captured similar phenomena through real-time imaging of dual-labeled VSV-G pseudotyped virus in both BEP-2D cells and HOECs (Supplementary Fig. 4-5 and Supplementary Movie 13-16).The results suggested that pseudo-SARS-CoV-2 entered BEP-2D cells and HOEC cells through endocytosis.
The endocytic entry of pseudo-SARS-CoV-2 in different types of respiratory epithelial cells was visualized and it was found to exhibit similar sequential patterns.
The receptor ACE2 was critical for the e ciency of SARS-CoV-2 productive infection.We performed a high-throughput analysis of dual-labeled pseudo-SARS-CoV-2 in different respiratory epithelial cells to study the e ciency of viral productive infection.The infected cells were xed at different time points, and viral entry e ciency was analyzed in HNEpC, HPAEpiC, BEP-2D, and HOEC, respectively.At each time point, 500 host cells were randomly selected for statistical analysis, and the results were collected during 0-180 min post-infection window.In the cytoplasm, the obvious increase in the co-localized signals of DiO and Vpr-mCherry at 0-60 min was due to the entry of virus particles into the host cells.Also, there was a signi cant decrease in the co-localized dots at 60 -90 min, indicating that the viral core was released from the envelope during this period.The productive infection of pseudo-SARS-CoV-2 showed a noticeable difference in HNEpC, HPAEpiC, BEP-2D cells, and HOEC (Fig. 6A).However, similar levels in productive infection were observed among the four cells types infected with VSV-G pseudotyped virus (control) (Fig. 6B).The results suggested that the differences between pseudo-SARS-CoV-2 and VSV-G pseudovirions were probably mediated by the interaction between S proteins and the SARS-CoV-2 receptors on these respiratory epithelial cells.
ACE2, the receptor for SARS-CoV-2, in these respiratory epithelial cells was tested using western blot.In Fig. 6C, the expression of the ACE2 receptor on these cells was consistent with SARS-CoV-2 productive infection based on statistical analysis.Next, pseudo-SARS-CoV-2 with equal numbers of RNA copies were used to infect respiratory epithelial cells to further determine the e ciency of virus entry (Fig. 6D).At 2 h post-infection, total RNA was extracted from the host cells, and intracellular viral RNA copy number was analyzed by RT-PCR.The intracellular viral RNA copy number of pseudo-SARS-CoV-2 in each host was positively correlated with the expression of the ACE2 receptor, indicating that the ACE2 receptor was critical for the e ciency of SARS-CoV-2 productive infection.

Discussion
In this work, we constructed dual-color pseudo-SARS-CoV-2 to investigate the entry pathway of the virus as well as the infection e ciency in respiratory epithelial cells under the conditions of BSL-2.Western blot and TEM results showed SARS-CoV-2 S protein had been incorporated into the pseudovirions and form the cobbled surface structure in the HIV-1 pseudovirus system.Additionally, the lipid envelope staining or immuno uorescence staining was performed to verify the uorescence labeling of virus particles.Results suggested that the constructed pseudo-SARS-CoV-2 possessed high uorescence co-localization e ciency.These particles also owned high infectivity, due to the labeling strategies exerted no signi cant interference to viral structure and function.Thus, dual-uorescent pseudo-SARS-CoV-2 provides a suitable tool for imaging the dynamic process of virus entry and capture details at the single-particle level.
As an air-transmitted virus, SARS-CoV-2 infection must happen through interactions between the virus and respiratory cells.Whether it only infects particular types of respiratory cells remains elusive.By observing patterns of virus entry among four types of respiratory cells (HNEpC, HOEC, BEP-2D cells and HPAEpiC), we found that pseudo-SARS-CoV-2 exhibit identical sequential process, suggesting there is no or little preference among these cells.Meanwhile, we showed that SARS-CoV-2 productive infection in different types of respiratory epithelial cells occurred through endocytosis.This result is consistent with report from Ou et al who showed that drugs blocking endocytosis can prevent entry of SARS-CoV-2. 7atistical analysis and RT-PCR results showed that there was a signi cant difference in the infection e ciency of various respiratory epithelial cells, and pseudo-SARS-CoV-2 preferentially infected HNEpC.
This could be attributed to the different levels of expression of SARS-CoV-2 receptors on these cells, because pseudo-SARS-CoV-2 infection required interaction between S protein and ACE2.In contrast, VSV-G pseudotyped virus failed to differentiate cells with different amount of ACE2 receptors.The results were consistent with early reports that the nose contained the highest percentage of ACE2-expression cells in the airways, and the nasal surfaces were the dominant initial site for SARS-CoV-2 respiratory tract infection. 2,16,24 Ostudy provides an excellent tool to systematically visualize the progression of SARS-CoV-2 infection in live cells.In the future, other viral components could be included in the pseudovirions system to investigate additional processes of the viral life cycle at the single-particle level.On the other hand, although we found endocytosis via interaction between SARS-CoV-2 S protein and ACEs is the major driving force of SARS-CoV-2 infection, we have not evaluated effects of endocytosis inhibitors, neutralizing antibody, or mutations of S protein during virus entry into cells.These relevant studies should be performed using live SARS-CoV-2 virus under the conditions of BSL-3.
In conclusion, we constructed a dual-color uorescence labelled pseudo-SARS-CoV-2 to study virus entry and cell susceptibility under BSL-2 conditions.Dynamic process of virus entry was visualized, imaged, and analyzed in live cells at single virion level.We observed similar patterns of entry pathways in different respiratory epithelial cells through endocytosis.There was a positive correlation between infection e ciency of the SARS-CoV-2 and expression level of the ACE2 receptor.The method may be further developed to elucidate the life cycle of SARS-CoV-2 at a single particle level.
Virus infection and uorescence imaging.Respiratory epithelial cells were plated into a confocal dish and incubated with pseudo-SARS-CoV-2 (DiO-mCherry) for 30 min at 4°C.Unbound particles were removed and replaced with fresh media before incubation at 37℃ in a 5% CO 2 atmosphere.Infected cells were imaged under an UltraView Vox spinning disk confocal laser scanning system using a Nikon Ti-e microscope with 60× objective.DiO and mCherry were excited at 488 nm and 561 nm, respectively.Realtime imaging was usually performed at an interval of 10 s for 15 min to minimize photobleaching.
Transmission electron microscopy analysis.Pseudo-SARS-CoV-2 was concentrated using the PEG 20000 solution, followed by ultra ltration using an Amicon Ultra-4 centrifugal filter device (50 kDa, Millipore).The carbon support copper grid was placed on the puri ed pseudovirions and incubated for 10 min.Then, the pseudovirions loaded on the carbon support copper grid were negatively stained with 2% phosphotungstate (PTA) for 30 s.The samples were observed on a 200 kV JEM-F200 transmission electron microscope.
Real-time PCR.Cellular total RNA of the infected cells was extracted 24 h post-transfection, and quantitative real-time PCR analysis was performed using a HiScript II One-Step qRT-PCR Probe Kit (Vazyme Biotech).To con rm the virus entry, cellular RNA was extracted 2 h post-transfection and analyzed using RT-PCR.The DNA sequences used were as follows: The primer HIV sense (TTT GAC TAG CGG AGG CTA GAA G), HIV antisense (CCC TGG CCT TAA CCG AAT TTT), and a speci c TaqMan probe (FAM-CGC TTA ATA CCG ACG CTC TCG C-TAMRA).

9 ).
The results suggested that pseudo-SARS-CoV-2 entered HNEpC cells through endocytosis.The release of viral core from the envelope during endocytic entry revealed by single particle tracking.During the endocytic entry of pseudovirions, we visualized the release of viral core from the envelope into the cytoplasm in various respiratory epithelial cells.The release was rst observed in HNEpC cells by temporal tracking of dual-labeled pseudo-SARS-CoV-2.Virus particles with co-localized signals of Vpr-mCherry and DiO were imaged in the cytoplasm of HNEpC cells.During virus transportation, Vpr-mCherry (red) was separated from DiO (green), indicating the release of the viral core from the envelope (Fig.2E-2Fand Supplementary Movie 2).Fig.2G-2Ishow the dynamic trajectories, velocities, and MSD of Vpr-mCherry and the envelope with DiO.Both uorescent dots had different trajectories, velocities, and MSD after their separation.The results suggested that the viral core successfully escaped from the endosomes and was released into the cytoplasm of HNEpC cells, and the dynamic process was necessary for productive infection.

Figure 4 image
Figure 4